An investigation of swift trust formation in disaster management teams
Matthew McLaren
A thesis in fulfilment of the requirements for the degree of
Doctor of Philosophy
Construction Project Management
Faculty of the Built Environment
2018
THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet
Surname/Family Name : McLaren
Given Name/s : Matthew
Abbreviation for degree as give in the University calendar : PhD
Faculty : Built Environment
School : Construction Project Management Thesis Title: An investigation of swift trust formation in disaster management teams
Abstract 350 words maximum: Due to the increasing vulnerability of developing nations to natural disasters, it is not uncommon for these countries to ask for international aid. Consequently, many developed nations send disaster management personnel to assist developing nations when responding to a natural disaster. Natural disasters present multi-national disaster management teams with numerous challenges and impose intense pressures, which provide little time for normal team formation processes to occur. The development of trust within multi-national disaster management teams is particularly important given the need for rapid responses under extreme time and resources pressures, where decisions often must be made outside of normal protocols. The newness of the teams, and the varied backgrounds from which they are drawn, further complicates this process of trust formation. However, within the disaster management literature, there is a lack of empirical evidence investigating how trust forms for local and international disaster management personnel responsible for coordinating a national operation. Addressing this lack of research and the poor conceptualisation of this important challenge, this thesis mobilises Meyerson, Weick and Kramer’s (1996) theory of Swift Trust in Temporary Systems to investigate the process of trust formation between disaster management personnel working in an Emergency Operations Centre. The research employs a single case study approach, focussing on the disaster response to Tropical Cyclone Winston in Fiji (2016). It utilises a triangulation of qualitative methods using semi-structured interviews, participant observation and document analysis. Through deductive content analysis, it was found that swift trust did not exist between the disaster management personnel responding to Tropical Cyclone Winston. Rather, the results showed that trust formation was more aligned with the organisational theories of trust and dependent upon a colleague’s role, competency, and personality. Particularly, when forming trust, local disaster management personnel were looking for their international colleagues to humbly integrate into the operation and work under the national leadership. These results are significant because they provide new theoretical insights into trust development within multi-national disaster management teams, which have not been previously explored in the literature. In addition to this theoretical contribution, the findings provide important practical insights to help international disaster management personnel facilitate more effective disaster responses by building greater levels of trust with their local counterparts.
Declaration relating to disposition of project thesis/dissertation
I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.
I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).
17/10/2018
Signature Witness Signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.
FOR OFFICE USE ONLY Date of completion of requirements for Award:
ii
ORIGINALITY STATEMENT
‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’
Signed ……………………………………………......
Date 17/10/2018
iii COPYRIGHT STATEMENT
‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'
Signed ……………………………………………......
Date 17/10/2018
AUTHENTICITY STATEMENT
‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’
Signed ……………………………………………......
Date 17/10/2018
iv ACKNOWLEDGEMENTS
To my supervisor Professor Martin Loosemore; I am indebted to you for the personal investment you have made in me throughout this research process. I can definitively say that I would not have finished if it were not for your dedication to see me through. Thank you for encouraging me when I was discouraged and guiding me step-by-step to the end. It has been worth it thanks to you.
To my co-supervisor Professor Ed Blakely; thank you for being the first person who said ‘yes’ to my research. You have given your time so generously over these years and I thank you for making my field work in Fiji possible; without which I would not have a thesis.
To my wife Emily; who would have thought that through this process we would decide to get married. Thank you for being my number one fan and continual provider of perspective. I truly believe that this is a thesis with every page lined with your support, reassurance and care.
To my parents, Rob and Judy, thank you for always being there for me through this process; no matter the cost to yourselves. Your selflessness has not gone unnoticed. To my friends, housemates, and mezze-lab colleagues over the years, thank you for listening, asking questions, and being on the team. I am so grateful for every one of you.
I would also like to thank the participants in this research from the Emergency Operations Centre in Fiji. Thank you for letting me conduct research in such a sensitive and operationally stressful time.
To the University of New South Wales – thank you for the opportunity to study. I particularly want to thank UNSW Canberra for providing me with a space to finish my PhD.
Finally, this research would not have been financially possible if it was not for the Australian Government providing me with an Australian Postgraduate Award.
v ABSTRACT
Due to the increasing vulnerability of developing nations to natural disasters, it is not uncommon for these countries to ask for international aid. Consequently, many developed nations send disaster management personnel to assist developing nations when responding to a natural disaster. Natural disasters present multi-national disaster management teams with numerous challenges and impose intense pressures, which provide little time for normal team formation processes to occur. The development of trust within multi-national disaster management teams is particularly important given the need for rapid responses under extreme time and resources pressures, where decisions often must be made outside of normal protocols. The newness of the teams, and the varied backgrounds from which they are drawn, further complicates this process of trust formation. However, within the disaster management literature, there is a lack of empirical evidence investigating how trust forms for local and international disaster management personnel responsible for coordinating a national operation.
Addressing this lack of research and the poor conceptualisation of this important challenge, this thesis mobilises Meyerson, Weick and Kramer’s (1996) theory of Swift Trust in Temporary Systems to investigate the process of trust formation between disaster management personnel working in an Emergency Operations Centre. The research employs a single case study approach, focussing on the disaster response to Tropical Cyclone Winston in Fiji (2016). It utilises a triangulation of qualitative methods using semi-structured interviews, participant observation and document analysis.
Through deductive content analysis, it was found that swift trust did not exist between the disaster management personnel responding to Tropical Cyclone Winston. Rather, the results showed that trust formation was more aligned with the organisational theories of trust and dependent upon a colleague’s role, competency, and personality. Particularly, when forming trust, local disaster management personnel were looking for their international colleagues to humbly integrate into the operation and work under the national leadership. These results are significant because they provide new theoretical insights into trust development within multi- national disaster management teams, which have not been previously explored in the literature. In addition to this theoretical contribution, the findings provide important practical insights to help international disaster management personnel facilitate more effective disaster responses by building greater levels of trust with their local counterparts.
vi TABLE OF CONTENTS
Defining natural disasters ...... 1
The effects of natural disasters on developing nations ...... 4
1.2.1. Human impacts ...... 4
1.2.2. Economic impacts ...... 5
1.2.3. The impact of urban growth on natural disasters in developing nations ...... 9
Disaster management ...... 12
1.3.1. Mitigation and preparedness in developing nations ...... 13
1.3.2. Response and recovery in developing nations ...... 14
International assistance after a natural disaster in a developing nation ...... 16
The Emergency Operations Centre ...... 19
Trust and disaster management ...... 20
Problem statement and aims ...... 25
Defining trust ...... 27
The conditions for trust to exist ...... 29
Trust and distrust ...... 30
Theories of trust ...... 31
2.4.1. Sociological theories of trust ...... 31
2.4.2. Economic theories of trust ...... 37
2.4.3. Organisational theories of trust ...... 39
The theory of swift trust in temporary systems ...... 44
2.5.1. Defining temporary systems ...... 45
2.5.2. Disaster management and the EOC as a temporary system ...... 46
2.5.3. The theoretical foundations of swift trust ...... 49
2.5.4. The trust criteria of swift trust ...... 51
vii 2.5.5. Factors influencing the resilience of swift trust ...... 52
2.5.6. Studies considering swift trust ...... 53
2.5.7. Swift trust and natural disaster management ...... 54
2.5.8. The need to investigate swift trust in an EOC ...... 56
The philosophical position ...... 58
3.1.1. Ontology and epistemology ...... 58
Single case study research design ...... 64
3.2.1. Case study type ...... 65
3.2.2. Single case study design ...... 66
Three-phase research design ...... 85
Phase 1 – Contextualisation and the temporary system boundary...... 88
3.4.1. Stage 1 – Determining a respondent’s original organisation ...... 89
3.4.2. Stage 2 – Determining when a respondent arrived in the EOC ...... 89
3.4.3. Stage 3 – Determining a respondent’s role within the EOC ...... 90
Phase 2 – Data collection methods ...... 90
3.5.1. The use of semi-structured interviews ...... 91
3.5.2. Pilot study ...... 92
3.5.3. Refined final data collection strategy ...... 98
3.5.4. Additional data collection methods ...... 99
3.5.5. Reflexivity ...... 103
Phase 3 – Data analysis ...... 105
3.6.1. The use of content analysis as a method of qualitative analysis ...... 106
3.6.2. Context – the analytical construct ...... 107
3.6.3. The process of content analysis ...... 110
Introduction ...... 118
Formed to address a task that is non-routine and requires interdependence ...... 120
viii 4.2.1. Formed to address the task of managing and coordinating the response and recovery…………………………...... 120
4.2.2. Non-routine ...... 124
4.2.3. Required interdependence ...... 128
A task that is defined by clear and specific goals and has a time limit on its duration………………………………...... 136
4.3.1. The goals of the EOC being specific and clear...... 137
4.3.2. The EOC’s limited duration ...... 138
Diversely skilled people drawn from a labour pool with overlapping networks ...... 138
4.4.1. The diversity of roles and skills of the respondents in the EOC ...... 138
4.4.2. The labour pool and overlapping networks ...... 142
Members that have a limited working history but the potential to work together again………………………………...... 149
4.5.1. The respondents’ work history ...... 149
4.5.2. Potential for working together again in the future ...... 153
Assembled by the contractor ...... 155
4.6.1. The Director as the contractor ...... 156
4.6.2. Whether the Director assembled the respondents in the EOC ...... 158
Discussion of Research Proposition 1 ...... 160
4.7.1. Formed to address a task that is non-routine and requires interdependence. 161
4.7.2. A task that is defined by clear and specific goals and has a time limit on its duration…………………………………...... 165
4.7.3. Diversely skilled people, the labour pool and overlapping networks ...... 165
4.7.4. Members that have a limited working history but the potential to work together again………………………………...... 167
4.7.5. Assembled by the contractor ...... 168
The EOC as a temporary system ...... 169
Swift Trust Criterion #1 - Reputation ...... 174
ix 5.1.1. Professional and personal reputations ...... 174
5.1.2. Untrustworthy behaviour damaging reputations ...... 177
5.1.3. Additional reasons for not expecting untrustworthy behaviour ...... 180
Swift Trust Criterion #2 – Interactions around role ...... 181
5.2.1. Interactions within the EOC based on a colleague’s role ...... 182
5.2.2. Colleague’s having clear roles ...... 184
5.2.3. Role behaviour ...... 189
5.2.4. The assumption of role as a swift trust criterion ...... 192
5.2.5. The importance of personality ...... 194
Swift Trust Criterion #3 – Categorical assumptions versus evidence-based information……………………………… ...... 196
5.3.1. Trust criteria used by respondents within the EOC ...... 196
5.3.2. Trust criteria specifically applied by locals to international disaster management personnel ...... 204
5.3.3. Culture as a categorical assumption used in the development of swift trust .. 208
5.3.4. The military’s involvement as a categorical assumption used in the development of swift trust ...... 212
5.3.5. Addressing Swift Trust Criterion #3 ...... 214
Swift Trust Criterion #4 – Confirmation of assumptions ...... 215
5.4.1. Validation of the trusting criteria applied to a colleague ...... 216
5.4.2. Expectations of behaviour ...... 218
Swift Trust Criterion #5 – Moderate interdependence ...... 221
5.5.1. The degree of interdependence between members in the EOC ...... 221
5.5.2. Interdependence due to group inclusion ...... 225
Swift Trust Criterion #6 – The contractor ...... 227
5.6.1. How the Director’s reputation influenced trust formation ...... 228
5.6.2. The Director’s reputation and its impact on international respondents extending trust………………………………...... 230
Discussion of Research Proposition 2 ...... 232
x 5.7.1. Reputation ...... 233
5.7.2. Interactions around role ...... 235
5.7.3. Categorical assumptions versus evidence-based information ...... 238
5.7.4. Confirmation of assumptions ...... 240
5.7.5. Moderate interdependence ...... 242
5.7.6. The contractor ...... 243
Did swift trust exist in the EOC?...... 244
The reputation of the contractor ...... 249
Optimism ...... 252
A presumption of trust ...... 256
6.3.1. The assumption that trust is presumed ...... 257
6.3.2. Whether behaviour affirmed a presumption of trust ...... 259
6.3.3. The impact of signage on the presumption of trust ...... 260
Assumed qualifications ...... 261
6.4.1. Whether the respondents knew their colleagues’ qualifications ...... 261
6.4.2. Whether the respondents thought their colleagues were qualified ...... 262
Hedges ...... 267
Improvisation ...... 271
Focussing on the task ...... 274
Discussion of Research Proposition 3 ...... 275
6.8.1. The reputation of the contractor ...... 276
6.8.2. Optimism ...... 277
6.8.3. A presumption of trust ...... 279
6.8.4. Assumed qualifications ...... 280
6.8.5. Hedges...... 282
6.8.6. Improvisation ...... 283
6.8.7. Focussing on the task ...... 284
xi The applicability of the swift trust resilience factors for trust in the EOC ...... 285
Overview of the literature ...... 287
Research proposition 1 ...... 288
Research proposition 2 ...... 291
Research proposition 3 ...... 295
Research limitations ...... 297
Further research ...... 298
7.6.1. Further research into swift trust in temporary systems ...... 298
7.6.2. Further research into trust formation and disaster management after a natural disaster………………………………...... 300
xii LIST OF FIGURES
Figure 1.2.1 Comparison of changes in mortality rates between developing and developed nations ...... 5
Figure 1.2.2 Economic damages compared against national GDP, 2009-2013 (adapted from EM-
DAT (2015); UN (2015); UNEP/UNISDR (2013); WorldBank (2015a)) ...... 7
Figure 1.2.3 A comparison of the rate of urban growth for developing and developed nations
(adapted from United Nations (2014)) ...... 9
Figure 3.2.1 The path of TC Winston across Fiji ...... 72
Figure 3.2.2 The TC Winston Emergency Operation Structure ...... 79
Figure 3.3.1 Research design - single case study approach ...... 87
Figure 3.4.1 The process for selecting personnel from the EOC to be included within the temporary system ...... 89
Figure 3.5.1 The Network Mapping Worksheet ...... 95
Figure 3.6.1 Content analysis process - adapted from Krippendorff (2013) ...... 110
Figure 3.6.2 Coding framework - adapted from Krippendorff (2012, 2013) ...... 111
Figure 4.2.1 The expanded EOC and new reporting structure adopted after TC Winston ...... 125
Figure 4.2.2 The EOC network ...... 130
xiii LIST OF TABLES
Table 1.1.a Disaster groups, definitions and types (adapted from CRED (2014) and Guha-Sapir,
Hoyois, and Below (2014)) ...... 3
Table 1.6.a Trust criteria used to measure and manage trust for NGOs (ECB, 2007, p. 7) ...... 24
Table 2.5.a Qualification of an EOC as a Temporary System...... 47
Table 3.1.a The possible philosophical positions for this research ...... 60
Table 3.2.a Information concerning each respondent interviewed within the EOC ...... 75
Table 3.5.a Ad-hoc conversations with specific respondents within the EOC in addition to the interviews ...... 102
Table 4.2.a The tasks required for the response and recovery, their respective objectives, and the team(s) responsible for their implementation (NDMP, 1995) ...... 122
Table 4.2.b The new roles adopted by the permanent EOC staff as part of the operation ..... 126
Table 4.2.c Pre-disaster tasks and objectives for the EOC ...... 126
Table 4.2.d The number of connections for each respondent within the EOC as determined from their Network Mapping Worksheets ...... 131
Table 4.2.e The number of connections for each team within the EOC ...... 134
Table 4.4.a The respondents’ roles, their unique role expertise, and secondary skills ...... 139
Table 4.4.b The respective organisations that each respondent worked for prior to the national operation...... 144
Table 4.4.c The organisations each respondent was connected to through their colleagues in the
EOC ...... 146
Table 4.4.d The number of organisations each respondent was connected to in the EOC ..... 147
Table 4.5.a The amount of pre-existing and new relationships for the respondents within the
EOC ...... 151
Table 4.5.b Comparison of new and pre-existing relationships for surge staff and permanent EOC staff ...... 152
xiv Table 4.6.a The person/organisation that asked each surge staff respondent into the EOC .. 159
Table 5.2.a A summary of the respondents’ trust criteria applied to local and international colleagues ...... 193
Table 5.3.a A summary of the respondents’ trust criteria applied to local and international colleagues ...... 197
Table 5.5.a The respondents’ recorded reliance on their colleagues compared against their perceived reliance on their colleagues ...... 224
Table 6.2.a Each respondent’s perceived ‘degree of impact’ for an operational problem in the
EOC ...... 253
Table 6.4.a A summary of whether the respondents thought their local and international colleagues were qualified for their roles in the EOC ...... 266
xv
INTRODUCTION
The impacts of natural disasters are not homogenous, with developing nations having higher mortality rates and comparative economic losses compared to developed nations. Subsequently, developed nations send international disaster management personnel to assist the developing nation respond to and recover from a natural disaster. However, there remains little research regarding how trust forms for disaster management personnel during the national response and recovery. Therefore, to highlight this gap in the literature, this chapter will first define natural disasters. It will then discuss the impacts of natural disasters on developing nations. Thirdly, it will explain the process of disaster management, particularly in developing nations, and how international personnel are sent to help. Finally, it will consider the importance of trust in this process and present the problem statement and research aims.
Defining natural disasters A disaster is defined as, “a situation or event which overwhelms local capacity, necessitating a request to a national or international level for external assistance; an unforeseen and often sudden event that causes great damage, destruction and human suffering” (Guha- Sapir, Vos, Below, & Ponserre, 2011, p. 7). Similarly, the United Nations Office for Disaster Risk Reduction defines a disaster as, “A serious disruption of the functioning of a community or a society at any scale due to hazardous events interacting with conditions of exposure, vulnerability and capacity, leading to one or more of the following: human, material, economic and environmental losses and impacts.” (UNISDR, 2017) The term disaster is used in many different contexts and can be confused with other terms such as emergency, crisis, catastrophe or hazard. For clarity,
1 an emergency is, “a minor event that can cause a few casualties and a limited amount of property damage” (Lindell, Perry, & Prater, 2007, p. 15). A crisis is a broad term which describes any, “unforeseen situation…accompanied by a high degree of uncertainty” (Demiroz & Kapucu, 2012, p. 93). A catastrophe is defined as a ‘major’ disaster with some greater implications observed at, “the organisational, community and societal levels” (E. L. Quarantelli, 2005, p. 3). A hazard is defined as a, “threatening event, or probability of occurrence of a potentially damaging phenomenon within a given time period and area.” (IFRC, 2018) For this research, the word disaster will refer to disasters, catastrophes, and hazards.
Considering disasters, there are two major types (Lindell et al., 2007):
Natural – an event that occurs in nature such as an earthquake, flood or hurricane.
Technological – events that are caused by the accidental failures of technologies, such as a hazardous material leak.
This research focusses on natural disasters for two reasons. Firstly, they are non-deliberate and uncontrollable, causing greater losses (particularly mortalities). Secondly, they can involve multi- national teams when responding to and recovering from the damages.
There are many different types of natural disasters, occurring at different frequencies in different areas of the world (Table 1.1.a).
2 Table 1.1.a Disaster groups, definitions and types (adapted from CRED (2014) and Guha-Sapir, Hoyois, and Below (2014))
Disaster group Definition Disaster type
Earthquake, mass Geophysical A hazard originating from solid earth. movement, volcanic activity
A hazard caused by short-lived, micro to meso scale extreme weather and Extreme temperature, Meteorological atmospheric conditions that last from fog, storm minutes to days.
A hazard caused by the occurrence, Flood, landslide, wave Hydrological movement, and distribution of surface action and subsurface freshwater and saltwater.
A hazard caused by long-lived, meso to macro scale atmospheric processes Drought, glacial lake Climatological ranging from intra-seasonal to multi- outburst, wildfire decadal climate variability.
In addition to Table 1.1.a, it is also important to recognise that natural disasters can also be categorised by their speed of onset (Lin Moe & Pathranarakul, 2006), as opposed to their group. Slow-onset natural disasters include those caused by global temperature variations, sea level rise, drought, etc. Rapid-onset natural disasters include events such as earthquakes, flash floods, tsunamis, and hurricanes, which have minimal lead time before the event occurs (Cutter et al., 2008; Lin Moe & Pathranarakul, 2006). Significantly, both types of natural disasters can be exacerbated by the effects of climate change (G. O'Brien, O'Keefe, Rose, & Wisner, 2006).
Drawing from the largest public disaster database, the latest report from the Belgium Centre for Research on the Epidemiology of Disasters (CRED) stated that between 2005-2014, there was a global, annual average of 380 disasters, with 76,416 deaths, 199 million victims, and economic damages of $159.7 billion (Guha - Sapir, Hoyois, & Below, 2016). However, the spread of natural disasters is not homogenous and these losses are not equally distributed (Coppola, 2006), with developing nations being adversely impacted much more than developed countries.
3 The effects of natural disasters on developing nations There is contention about what defines a developing nation (also referred to as a lower developed nation or the majority world). However, to avoid generalisations and for specificity in this research, nations classed as low-income or middle-income economies are defined as developing (based on the World Bank measures of gross national income (GNI) per capita (WorldBank, 2015b)). This thesis defines low-income nations as those that have a GNI per capita of $1,045 or less, and middle-income economies as those that have a GNI per capita of more than $1,045 but less than $12,746 (all figures are in 2013 US dollars) (WorldBank, 2015a).
Although natural disasters impact both developing and developed nations, developing nations are subject to higher human impacts and greater economic losses (as a proportion of national Gross Domestic Product). Additionally, due to increased rates of urban growth, developing nations are more susceptible to urban poverty, which compounds the impacts of natural disasters. Urban growth in developing nations can lead to poorly built infrastructure in dangerous areas, sustained environmental degradation, and greater consequences from the effects of climate change. The human and economic impacts in developing nations are discussed, as well as the problems with urban growth.
1.2.1. Human impacts In 2013, 50% of the top ten countries which suffered natural disaster mortalities were, “low income or lower-middle income economies…[yet they] accounted for 88% of global disaster mortality” (Guha-Sapir et al., 2014, p. 1). More generally, “while only 11 percent of the people exposed to natural hazards live in countries classified as [having] low human development, they account for more than 53 percent of total recorded deaths” (UNDP, 2004). The data presents a clear picture that developing nations are most at risk to natural disaster mortalities. This trend affecting developing nations is not new, it has been happening for decades. Between 1960-1990, on average, the annual death toll dropped by 75% for developed countries, whilst developing nations had a 400% increase (Lindell, 2013).
4 change +400% proptionalpercentage -
-75% Mortality
1955 1960 1965 1970 1975 1980 1985 1990 1995
Developing Nations Developed Nations
Figure 1.2.1 Comparison of changes in mortality rates between developing and developed nations
The developing world, “includes a disproportionate number of…disaster-prone countries” (Benson & Clay, 2004, p. 25), resulting in a higher risk of mortality in those nations (EM-DAT, 2015; UNEP/UNISDR, 2013).
Furthermore, developing nations are also vulnerable to social losses from natural disasters. These can include loss of employment (Tseng & Chen, 2012) or loss of a business (Shirley, Boruff, & Cutter, 2012), as well as other livelihood factors such ease of access to medical services (Shirley et al., 2012), loss of structures, and loss of animals and crops (Lindell, 2013). Furthermore, citizens of developing nations subject to the impacts of a natural disaster may have to temporarily or permanently relocate (Hunter, 2005) due to the damage incurred. This process may be voluntary, however it may also be forced (Hunter, 2005); resulting in involuntary displacement.
1.2.2. Economic impacts Considering the economic losses specific to developing nations, it is not accurate to equate these nations’ losses ‘dollar for dollar’ with the losses of developed nations. This is because economic loss should be a proportional measure, and not based solely on the financial magnitude of the loss. Therefore, to properly understand the impact of the financial damages suffered by a nation due to a natural disaster, the losses incurred can be evaluated through their roll-on effect against the nation’s GDP. It is believed that natural disasters have negative effects on national GDP
5 (Hochrainer, 2009), with some natural disasters leading to declines in national per-capita GDP of 2% (Raddatz, 2007). A second measurement used to determine the impacts of the financial costs of a disaster on a nation is by calculating that nation’s economic loss as a percentage of their national GDP. “A natural disaster will have a strong impact on a country’s economic performance if the size of damages compared with the size of the economy…is high” (Charvériat, 2000, p. 20). By applying such a measurement, it is possible to more accurately compare the impacts of natural disasters on the economies of both developed and developing nations, and understand economic losses in terms of a nation’s respective financial capability.
6
Figure 1.2.2 Economic damages compared against national GDP, 2009-2013 (adapted from EM-DAT (2015); UN (2015); UNEP/UNISDR (2013); WorldBank (2015a))
7 Figure 1.2.2 presents a compilation of data concerning the nations deemed most at risk from natural disasters (UNEP/UNISDR, 2013). From this data set, the graph depicts the top 10 nations for the years of 2009-2013, in terms of:
economic damages incurred; and/or
economic damages incurred as a percentage of the national GDP.
In support of Charvériat (2000), the data shows that the economies of developing nations are especially susceptible to the financial paralysis from natural disasters:
“When we take a local view and put the productivity losses of a city in relation to the GDP of an entire country…a smaller country with only one or a handful of urban centres can end up in the first ranks because these cities play an essential role to their home country’s national economy.” (Sundermann, Schelske, & Hausmann, 2013, p. 14)
This disproportionate level of loss compared to their national GDP, coupled with the high level of exposure to and subsequent impacts of natural disasters, can create a cycle of continual loss for developing nations. This process can also negate the limited progress made toward developmental goals (Hochrainer, 2009; UN, 2005), impeding a nation’s development and keeping communities in a constant state of poverty (Sharma, Surjan, & Shaw, 2011).
A common form of protection against the economic damages from natural disasters is insurance. However, developed nations are in a stronger position to utilise such a safeguard, and natural disaster insurance is rarely present within developing nations (Rasmussen, 2004). From 1980 – 2004, developing nations with low national incomes had only 1% of their losses covered by insurance, as opposed to 30% for high-income developed nations (Linnerooth-Bayer, Mechler, & Hochrainer, 2011). The economies of developing nations, “are not as diversified in their productive capacity and financial markets and are thus less able to successfully insure against the external shocks of a natural disaster” (D. Kellenberg & Mobarak, 2011, p. 306). Besides this, Linnerooth-Bayer et al. (2011) found that there can be distrust from citizens in developing nations towards the insurance industry and concerns with misunderstanding and misinformation. Gurenko (2004, p. xxix) summarises the aforementioned economic issues within developing nations when they write,
“developing countries…suffer from…widespread low risk awareness, underdeveloped insurance markets, excessive reliance on the government as the reinsurer of last resort, low incomes, inadequate information about expected economic losses from natural
8 disasters, high distribution costs and lack of public trust in government-sponsored insurance programmes.”
1.2.3. The impact of urban growth on natural disasters in developing nations The impacts of natural disasters within developing nations are being compounded by the growing number of people living within cities and urban centres. Understanding the concept of population growth in cities requires definitions of two key parameters. The first is urbanisation, which is, “a shift in the distribution of population from rural to urban locations” (Clark, 2000, p. 18). This natural migration compounds the second process, called urban growth. Urban growth is defined as the process where more people living in cities will fuel the rate at which natural population growth will occur within these hubs (Clark, 2000).
As of 2013, 53% of the world’s population lived in urban settlements (WorldBank, 2015c). Looking forward,
“The population living in urban areas is projected to gain 2.6 billion, passing from 3.6 billion in 2011 to 6.3 billion by 2050. Thus, the urban areas of the world are expected to absorb all the population growth expected over the next four decades.” (UN, 2012, p. 1)
Figure 1.2.3 A comparison of the rate of urban growth for developing and developed nations (adapted from United Nations (2014))
As shown in Figure 1.2.3, it is projected that 95% of urban population increase in the coming decades will occur within developing nations. By 2030, the data predicts that the developing world will hold 80% of the global urban population (UN-HABITAT, 2006). Comparing the urban
9 population exposed to natural disasters in the 10 most at risk developing nations between 1950 and 2030 shows a projected 1600% increase in the number of people at risk (UNEP/UNISDR, 2013; United Nations, 2014).
It is necessary to state that although developing nations are responsible for the majority of urban growth, rural areas in developing nations are also susceptible to the impacts of natural disasters. Particularly, natural disaster mitigation can be more difficult in rural communities due to them having less people, potentially weaker economic and social institutions, and poor infrastructure and planning (D. K. Kellenberg & Mobarak, 2008). Additionally, many rural communities’ reliance on subsistence agriculture means that a natural disaster can have devastating consequences on their livelihood (Davies, Guenther, Leavy, Mitchell, & Tanner, 2009). “Rural residents may be more vulnerable due to lower incomes and more dependent on locally based resource extraction economies.” (Shirley et al., 2012, p. 247) Lastly, natural disaster mitigative measures in urban centres can also directly impact rural communities; such as in Bangladesh, where flood water in Dhaka are diverted to surrounding regions.
Refocussing on urban centres, Blakely, Birch, and Anglin (2011) comment on the problems that developing nations could face due to the impacts of natural disasters on highly populated urban areas (also addressed by Hewitt (1998)):
“Disasters in urban areas are exponentially more harmful owing to the high population densities and economic importance of cities. Urban disasters yield enormous mortality and injury rates, tear apart the layered fabric of densely-settled neighbourhoods, destroy complex economies and interdependent social systems, and wreak havoc with costly infrastructure.” (Blakely, Birch, & Anglin, 2011, p. 7)
Sundermann et al. (2013, p. 1) further emphasise the risk developing nations face as their cities continue to grow, despite being susceptible to natural disasters. They state that, “the growing concentration of people, assets and infrastructure means that the loss potential in urban areas is high and rising.” El-Masri and Tipple (2002) continue by stating that high urban growth rates in developing nations, coupled with low incomes or poverty, inadequate access to appropriate resources, and rushed planning, has resulted in many people living in poorly constructed housing on dangerous sites, such as floodplains and mountain sides which are more exposed to natural disaster risk (Alexander, 1997; Blakely, Birch, & Anglin, 2011; El-Masri & Tipple, 2002; Fay, Ghesquiere, & Solo, 2003; Sanderson & Jones, 2017; Sundermann et al., 2013; Uitto, 1998). This process, termed urban poverty, only aggravates the effects of natural disasters as people are forced, due to their limited economic means, to, “acquire land and build housing outside of the
10 official system of land-use controls and building standards” (Bull-Kamanga et al., 2003, p. 197). That is, in developing nations, large populations are forced, “to concentrate in perilous, high- risk urban areas that contain little or no defence against disasters” (Haddow, Bullock, & Coppola, 2017, p. 332).
As a result of these poor land use practices and sporadic planning (due to urban growth), there is a risk that the capacity of a city and its surrounding land will be exceeded. The gradual deterioration of the surrounding natural environment, combined with the pressures of high populations, are expected to increase the effects of natural disasters on developing nations (Alexander, 1997; D. Kellenberg & Mobarak, 2011). Furthermore, the damaging effects of sprawling urban growth on natural land fertility (Lindell et al., 2007) can create a diminishing cycle for those living in urban centres. As people are forced to over-exploit the land, natural degradation worsens, which can lead to further poverty; as the environment offers less return each year (El-Masri & Tipple, 2002).
A secondary effect of this environmental degradation, due to urban growth, is that developing nations are becoming more exposed to the effects of climate change. It is predicted that climate change will increase the occurrence (Lindell et al., 2007) and intensity of natural disasters, and developing nations are particularly vulnerable (Mirza, 2003) due to informal housing, poor services, and inadequate local government investment (Revi et al., 2014). Campbell-Lendrum and Corvalán (2007) add that, “construction patterns in many developing cities have resulted in a combination of degradation of natural protection, poor quality housing construction on exposed slopes, and extensive ground coverage of concrete without adequate drainage.” Such changes can exacerbate the impact of to disasters such as flash floods (Campbell-Lendrum & Corvalán, 2007). Much development is also occurring along coastal regions, where rising sea levels are expected to increase the risk of flooding, hurricanes and storm surges to these areas as well as many small island developing nations (Campbell-Lendrum & Corvalán, 2007; Pelling & Uitto, 2001). As shown in Bangladesh, urban growth can also be driven by the migration of environmental refugees from rural to urban centres (Bates, 2002; Hunter, 2005). Furthermore, environmental refugees can also seek refuge in other, neighbouring developing nations, increasing pressure on these places. For example, up to 17 million Bangladeshis moved to India due to frequent storms, floods and droughts (Reuveny, 2007). Other examples of climate refugees include those affected by flood and drought in China (Hunter, 2005) and the victims of the Soufriere Hills volcano in the Caribbean (Bates, 2002).
11 As a result of the above forces, it is expected that urban centres within developing nations will be increasingly affected by natural disasters in the coming years (Blakely, Birch, & Anglin, 2011). “In a world of on-going population growth, rapid urbanisation…and an approach to investment that discounts disaster risk, the potential for future losses is enormous” (UNISDR, 2013, p. iii). Therefore, “the need for disaster management is nowhere more urgent than in the world’s sprawling urban centres” (Sundermann et al., 2013, p. 5). Therefore, disaster management must become an integral part of these nations’ forward planning and development.
Disaster management Disaster management is a process made up of four distinct stages; mitigation, preparedness, response, and recovery (Abrahams, 2001; Coppola, 2006; Waugh & Streib, 2006); also respectively referred to as reduction, readiness, response, and recovery (Ritchie, 2008; Ronan et al., 2008). These stages should not be considered as linear but rather linked and interdependent (Coppola, 2006; E. L. Quarantelli, 2003). The four stages are described as follows (E. L. Quarantelli, 2000):
Mitigation - the policies and actions undertaken before a disaster in order to prevent or limit the impact when it occurs. Examples include building codes, land use planning, and insurance.
Preparedness - the steps and processes that take place when a disaster is imminent. Examples include warnings being issued and evacuations.
Response - the actions taken immediately following the disaster to deal with the present emergency. Actions include search and rescue, and medical aid.
Recovery - the activities undertaken after the response period of the disaster. These include rebuilding infrastructure and reopening businesses.
As previously stated, disaster management needs to be integrated into developing nations’ planning and forward progression (Mirza, 2003). Although some resources are available, such as the United Nations’ Disaster Resilience Scorecard for Cities (UN, 2014), there still remains a large gap between the disaster management capabilities of developing and developed nations.
Similarly, socioeconomic stress, poor physical infrastructure, low levels of education, insufficient commercial resources, and a lack of local aptitude also hinder the ability of a developing nation to strengthen their disaster management capabilities (Al-Nammari & Alzaghal, 2015; Henderson, 2004). To address this problem, this section will first consider some of the challenges developing
12 nations face in the pre-disaster stages of mitigation and preparedness. It will then do the same for the post-disaster stages of response and recovery. Finally, it will consider the role the international community plays in sending disaster management personnel to help manage the response and recovery after a natural disaster in a developing nation.
1.3.1. Mitigation and preparedness in developing nations At present, developing nations do not invest in preparedness and mitigation to the same degree as developed nations. There are multiple factors that may weaken a developing nation’s ability to mitigate against and prepare for a natural disaster, including (D. Kellenberg & Mobarak, 2011; McEntire, 2004):
low rates of education;
weak institutions;
poor access to health care for citizens;
the occupancy of hazardous sites and unsafe buildings;
poor building standards;
corruption;
a lack of advanced warning and emergency response systems;
nominal insurance; and,
resource poor governments with insufficient disaster institutions.
Regarding mitigation and preparedness, Kreimer (2001) comments that, “the damage caused by disasters can be reduced by changing perceptions and behaviour so that all members of a society place a high priority on safety in planning and development.” However, “although the importance of disaster preparedness and mitigation is widely recognised by almost all of the world’s countries…countries ranking lower on development indices place disaster management very low in budgetary priority” (Coppola, 2006, p. 19).
In terms of better mitigation and preparedness within developing nations, Kreimer (2001) recommends higher design standards for important buildings such as schools and hospitals, education for citizens (especially about natural disasters), enforcement of building codes and land use planning, as well as income generating programs supported by the insurance industry.
13 However, the effects of poverty remain, and mitigating against or preparing for a disaster may be financially impossible (Pelling, 2003; Pelling, Özerdem, & Barakat, 2002). Mitigation and preparedness are necessary within developing nations, however there still remains a, “growing gap between rich and poor nations in their ability to mitigate the effects of natural disasters” (Kreimer, 2001, p. 403).
Conversely, some developing nations, in response to their high exposure to disasters, have started to strengthen their mitigation and preparedness measures (Khan & Rahman, 2007; Twigg, 2004). Examples include the establishment of the Coordination Centre for the Prevention of Natural Disasters in Central America and the Caribbean Disaster Emergency Response Agency to mitigate and prepare for natural disasters impacting Latin America and the Caribbean respectively (Twigg, 2004). Similarly, “the development of laws and policies, improved institutional frameworks and planning, and a growing number of risk reduction initiatives [are being made in]…developing countries.” (Twigg, 2004, p. 62)
1.3.2. Response and recovery in developing nations After a natural disaster occurs, the response commences, followed by the recovery; which may last many years. For example, Aceh and Nias in Indonesia set a five year plan for recovery and reconstruction following the 2004 Indian Ocean earthquake and tsunami (Hayashi, 2011). However, these stages of disaster management can also be hindered by poverty. As previously outlined, “higher rates of poverty in the populations of developing nations not only provide fewer resources for the poor to reduce their exposure to both hazards and risks, but they prolong disaster recovery” (Henderson, 2004, p. 111).
As such, developing nations are generally confronted with a sizeable challenge when financing the response and recovery. Kreimer (2001) comments on the different economic approach taken by developed countries compared with developing countries:
“In the developed countries, efforts are made ex ante to reduce losses and to repair damage, while their economies continue to grow. In addition, highly efficient and effective banking systems, insurance arrangements, the private sector, and tax payers all can play a role in promoting rapid recovery and providing incentives for reducing catastrophic losses.” Kreimer (2001, p. 403)
Furthermore, after a natural disaster, there can be a large influx of money from international donors to help a developing nation respond and recover. However, this may have a negative effect on long term disaster management. Although temporarily aiding the post-disaster effort,
14 this provision of money can act as a disincentive for developing nations to invest in mitigation and preparedness. It can contribute to their governments and citizens establishing an ethos of dependence upon external support; despite the known benefits of pre-disaster management (Dilley, 2005; Linnerooth-Bayer et al., 2011).
In addition, these nations can end up diverting funding designated to national development into reconstruction; at the expense of completing projects that will help support long-term economic growth and development (Coppola, 2006; Gurenko, 2004). Unfortunately, natural disasters in developing nations can result in the, “diversion of foreign exchange and public finance to disaster response and recovery…[which] may retard local, regional, or national economic development projects” (Henderson, 2004, p. 112). This diversion of funding highlights the growing need for developing nations to incorporate disaster management into their development goals and to recognise that development and disaster management hold the unified purpose of improving national capability (Kellett & Peters, 2014; McEntire, 2004; Stenchion, 1997; UNDP, 2004). A final concern when financing the response and recovery of a natural disaster is corruption (Coppola, 2006; Kahn, 2005; D. Kellenberg & Mobarak, 2011; E. L. Quarantelli, 1999), which can be more prevalent in developing nations, and may hinder the post- disaster management effort (Strömberg, 2007).
Apart from funding the response and recovery for a natural disaster, developing nations also face several other challenges, such as the use of aging or poorly constructed infrastructure (Henderson, 2004; Pelling, 2003; Sundermann et al., 2013). When rebuilding after a disaster, these weak buildings and public services can be replaced by similarly poor infrastructure in order to recover as quickly as possible. As Haas, Kates, and Bowden (1977, p. 268) state, “there is already a plan for reconstruction, indelibly stamped in the perception of each resident – the plan of the pre-disaster city. The new studies, plans and designs compete with the old.” However, this short term approach to recovery can hinder economic growth and future development of the nation (D. Kellenberg & Mobarak, 2011) as it can leave a country just as vulnerable to future natural disasters as it was before (Lindell et al., 2007).
In the overall management after a disaster, many developing nations’ public servants (who will be involved in the response and recovery) are not well trained, and lack the skills and experience necessary to conduct effective post-disaster management (Henderson, 2004). Similarly, developing nations can suffer disruptions to their central governing systems (Carter, 2008), which impacts their ability to manage the response. Even at a local government level, developing nations can lack the ability to manage a response due to, “ineffective planning and
15 implementation, ambiguous lines of authority, blurred methods of information and dissemination, lack of inter-organisational coordination, neglect of recovery problems, and violation of established planning procedures” (Kusumasari, Alam, & Siddiqui, 2010, p. 444).
Conversely, there has been some movement by developing nations to increase their disaster management capabilities. Indonesia, Bangladesh and Pakistan all now have disaster management departments as part of their governments and subsequent national legislature (Swithern, 2014). However, there still remains, “a critical difference between the experience in the developing world and that in developed countries with respect to coping with disasters” (Kreimer, 2001, p. 403) and their management.
International assistance after a natural disaster in a developing nation International participation in developing nations’ response and recovery efforts occurs because of these nations’ higher susceptibility to natural disasters (Benson & Clay, 2004; Katoch, 2006) and weaker natural disaster management systems (Kreimer, 2001). As E. L. Quarantelli (2000) writes,
“in some developing countries… there exists at best only a nominal social organisation at the national level [concerned with disaster management], few if any local units, and no formal planning of any kind. In these societies, when disasters occur, the relevant crisis time functions are carried out by international relief organisations.” (E. L. Quarantelli, 2000, p. 15)
Given developed nations generally have more advanced disaster management processes, they often offer to send disaster management personnel to aid developing nations after a natural disaster (İsbir & Genç, 2008; Katoch, 2006; E. L Quarantelli, 1997; Wenger, Dykes, Sebok, & Neff, 1975). Furthermore, developed nations can also have established relationships with developing nations; which can provide a base prerogative to aid after a natural disaster (Khan & Rahman, 2007). Supporting personnel may be drawn from many different international groups, and include (Bui, Cho, Sankaran, & Sovereign, 2000; Collins, 2009; Coppola, 2006; Drabek & McEntire, 2003; Henderson, 2004; İsbir & Genç, 2008; Katoch, 2006; Lindell et al., 2007; McEntire, 2007; Patterson, Weil, & Patel, 2010; Pettit & Beresford, 2005; Raju, 2013):
the United Nations;
the World Bank;
16 the World Health Organisation (WHO);
the International Federation of Red Cross and Red Crescent Societies (IFRC);
the North Atlantic Treaty Organisation (NATO);
the World Meteorological Organisation;
other national governments;
other national militaries;
international NGOs;
international disaster management experts;
international financial institutes; and,
international media outlets.
The typical roles that these imported organisations and personnel carry out include (as examples) (Carter, 2008):
monitoring and warning of the disaster;
surveying and assessing damage in the disaster affected regions;
providing specialist emergency roles (such as technical expertise and medical treatment); and,
providing emergency equipment, supplies, transport, and additional logistical support.
It is important to note that as the disaster response and recovery are the responsibility of the host nation and their government, international disaster personnel can only become involved once requested (İsbir & Genç, 2008). However, although such help may be needed and requested by a developing nation, the lack of predictability about which international organisations may offer assistance after a disaster means that such help, if offered, may not be optimally utilised (Carter, 2008). Furthermore, a large influx of external personnel and organisations can overwhelm the management capabilities of a developing nation; which adds complexity on top of simply responding to the disaster. Unfortunately, developing nations can lack the capability to coordinate a large-scale response involving international participants (Coppola, 2006).
17 Conversely, some developing nations such as China, Iran, Cuba, India and Malaysia are making considered efforts to improve their disaster management capabilities (Katoch, 2006; Roosli, O’Keefe, & Mydin, 2013). However, it still remains that, “major disasters – in particular those that occur in developing countries – often require extensive help involving worldwide assistance” (Bui et al., 2000, p. 427). For example, within weeks of the 2004 Asian tsunami, Indonesia had approximately 400 international non-governmental organisations (INGOs) responding (M. Perry, 2007), and Tamil Nadu in India also had an extensive amount of participants (Raju, 2013). Four days after the 2003 earthquake in Bam, Iran, around 1,300 international participants from 34 different countries arrived to assist in the response (Katoch, 2006). Four years earlier in the 1999 Marmara earthquake, Turkey had many international participants (İsbir & Genç, 2008). For this earthquake, considering NGOs alone, there were, “almost 80 international [and] 100 national NGOs [who] took part with 3,622 personnel in the disaster response state” (İsbir & Genç, 2008, p. 94).
A developing nation can be inundated with an extensive amount of foreign post-disaster personnel. However, it is worth reiterating the importance of locally-led post-disaster phases of response and recovery (Henderson, 2004; Patterson et al., 2010; M. Perry, 2007). As such, international participants need to work alongside developing nations, even though these nations, according to Quarantelli (1998):
can be reactive rather than proactive in their disaster management;
have limited capability to adapt to the requirements of a natural disaster; and,
have very few post-disaster organisations at the local level, and of those that do exist, there is minimal political backing.
This presents a complex situation for a developing nation, where international disaster personnel are necessary, however they, and their management styles, must be incorporated alongside the local disaster managers; who can lack the required capacity and skills to respond to a disaster (İsbir & Genç, 2008; E L. Quarantelli, 1998). The task of coordinating these respondents is the responsibility of the developing nation (Carter, 2008). However, it is also important to recognise that this task of coordination is made more difficult due to:
cultural, organisational, jurisdictional, and legal barriers (Janssen, Lee, Bharosa, & Cresswell, 2010);
the use of different languages (Coppola, 2006);
18 the fact that personnel may not have met each other and frequently change throughout the response (Kapucu & Garayev, 2011);
different operational priorities and degrees of influence for each organisation involved (Kapucu & Garayev, 2011); and,
a potential lack of appreciation for the specialised skills that are offered by international disaster personnel (Marjanovic & Nimpuno, 2003).
Therefore, by accepting international assistance after a natural disaster, a developing nation becomes host to a complex response environment where the local and international responding organisations, who generally operate autonomously, must now form an interdependent network (Janssen et al., 2010) that will temporarily exist for the life of the response. More specifically, the overall coordination of the response will be undertaken by a temporary group consisting of local and international disaster management personnel who:
may not have met each other before;
are subject to the social and technical barriers of a multi-national response;
must work with compressed timeframes; and,
are under pressure to ensure there is effective coordination between all post-disaster participants (Henderson, 2004).
Therefore, to ensure effective collaboration whilst facing many technical and social challenges, this temporary group of local and international disaster management personnel generally will coordinate the response and recovery from a centralised location, called the Emergency Operation Centre.
The Emergency Operations Centre Most response and recovery strategies for a natural disaster operate out of a ‘hub’ or “nerve centre” (Stephens & Grant, 2001, p. 290) called an Emergency Operations Centre (EOC). The EOC, “is an ad hoc team that assembles in a central location during an emergency to support on- scene emergency response personnel and coordinate the response across incident command centres” (Militello, Patterson, Bowman, & Wears, 2007, p. 25). At a high level, the purpose of an EOC is to ensure coordination (R. W. Perry, 1995) and communication between political leaders, response agencies, the general public, and external sources of assistance such as disaster management personnel (Stephens & Grant, 2001). Additionally, inclusion in the EOC is selective,
19 based on technical expertise and political authority (R. W. Perry, 2003), and all personnel report to the head of the EOC (Anglin, 2011; R. W. Perry, 2003). This person is generally provided by the host nation and oversees the delivery of the response.
The specific purpose of an EOC, according to R. W. Perry (1995, p. 39), is to undertake several core functions, including:
coordination between responding resources, including local and international disaster management personnel;
creation of policies to guide the response and recovery;
overseeing the management of disaster operations; and,
collation and dissemination of information.
(For a more detailed explanation of the purpose and roles of an EOC, refer to Table 2.5.a). Therefore, the EOC can be considered a selective, temporary organisation that is created to specifically facilitate the coordination and interpersonal communication between the local and international disaster management personnel leading the response and recovery. It is also the place where local and international disaster management personnel must balance the requirements of the response and recovery with the technical and social challenges present in a multi-national operation.
Trust and disaster management There are many technological and social dimensions to resolving the numerous challenges involved with coordinating a complex, temporary network such as an EOC, which is subject to continual pressure in a resource-constrained environment. However, at the heart of any effective response and recovery is the trust that develops between those within the EOC, who are charged with managing and coordinating the operation (Comfort, 2005). Indeed, Waugh (1999, p. 188) goes so far as to say that, “trust…may be the key element in assuring effective collaboration in a disaster operation.” He also states that, “the effectiveness of emergency managers may depend more upon interpersonal skills, trust and trustworthiness…than technical skill and legal authority” (Waugh, 1999, p. 14). Similarly, Janssen et al. (2010) state that disaster management is underpinned by reciprocal trust. More specifically, trust is an essential parameter for coordination and cooperation after a disaster (Kapucu, Arslan, & Demiroz, 2010), and personnel that have built trust are, “more cooperative and productive in collaborative
20 decision-making processes” (Kapucu & Garayev, 2011, p. 373). Noran (2011) also argues that without trust, collaboration during an operation can be hindered.
Although the literature considers trust important in disaster management, there is limited understanding of what factors impact trust formation in the temporary group of local and international disaster management personnel in the EOC. Research exploring trust in the general context of disaster management response and recovery is limited, and largely fails to consider the factors that determine the development of trust for this select, temporary group of post-disaster personnel.
Particularly, many studies into trust and natural disasters use the popular two-stage game protocol produced by Berg, Dickhaut, and McCabe (1995) as a method of data collection for trust and trustworthiness. Cassar, Healy, and Von Kessler (2011) investigated the effects of the 2004 Asian tsunami on trust levels within select villages of Thailand. Their work found that people within these villages were more trusting after the natural disaster if they received help from either friends or financial institutions. Similarly, Castillo and Carter (2011) have shown that individuals in 30 small Honduran communities were more trusting of those within their communities after Hurricane Mitch. Veszteg, Funaki, and Tanaka (2014) also found that mutual trust increased for Japanese citizens after the 2011 Tohoku earthquake and tsunami in Japan. Conversely, Fleming, Chong, and Bejarano (2014) found that although trust between local Chileans decreased after the 2010 earthquake, there was an increase in trust towards non-local strangers. Another study was conducted in Sri Lanka by Conzo (2014) who determined that trust between victims affected by a disaster can be altered depending on how a citizen remembers their experience of a natural disaster. All of these investigations focussed on measuring an overall change in trust for post-disaster participants due to a natural disaster. However, the studies only showed that trust changed. No study provided a comprehensive analysis of what factors impacted trust formation during a disaster response. Considering these examples from game theory, there remains a gap in the literature to understand why trust increased in some scenarios but decreased in others. That is, what factors impact trust formation in a disaster management context.
Considering other research of trust in disaster management, Yoder-Bontrager (2014) wrote that in Honduras, for disaster response to be effective, there needs to be trust between the different levels of government, private and non-profit agencies. They also found that for NGOs in a disaster, coordination happens most naturally when the NGOs trust one another. Andrabi and Das’ (2010) study of the 2005 earthquake in Northern Pakistan recorded an increase in victims’
21 trust towards international post-disaster personnel as a function of the amount of aid received. They stated that, “trust in foreigners is malleable, responds to humanitarian actions by foreigners and is not a deep-rooted function of local preferences” (Andrabi & Das, 2010, p. 1). Although there is a general recognition that trust in international disaster personnel can be influenced by the expertise they provide, there is little empirical research considering other trust factors that are valued by their local counterparts.
A deeper analysis of trust in natural disasters has been considered by Chandrasekhar (2010, 2012). This work considers the significance of trust as one of four primary factors (trust, power, legitimacy, and urgency for action) influencing stakeholder participation in recovery-decision making in Nagapattinam (India) after the 2004 Indian Ocean tsunami. Importantly, the author provides an example of the role of trust between local leadership and international post-disaster personnel. They state that the collaboration between the District Administration (the local bureaucracy) and the international NGOs had the greatest impact on the post-tsunami recovery because it was based upon high levels of trust; as well as the other factors of power, legitimacy and urgency for action. This research provides an insight into the importance of trust between international and local disaster. However, the question that is not addressed by the author is; ‘how much of a role did trust play, as compared to the other factors, in the collaboration between these disaster personnel?’
Another benefit of Chandrasekhar’s (2010, 2012) research is that it provides a set of indicators which can be used to measure trust and what may cause it to change. These indicators are drawn from the work of Kumar and Paddison (2000) and White and Eiser (2006) and are tailored toward stakeholder recovery-decision participation. They are as follows (Chandrasekhar, 2010, p. 21):
the presence of, and changes in, networks of social exchange and delegation;
reciprocity and protection of other stakeholder’s interests over time;
full and frank information sharing;
meeting the expectations of all stakeholders;
stakeholders behaving as expected;
not feeling vulnerable; and,
shared values of stakeholders.
22 These indicators are used by Chandrasekhar (2010) to measure trust. However, what is not provided is any form of weighting behind each of these trust indicators and whether some are more important than others, and if so, why? There is also a lack of justification as to why some of the indicators originally included by Kumar and Paddison (2000) were excluded from Chandrasekhar’s (2010) experiment, and why the indicators chosen were the best fit for the research.
In contrast to Chandrasekhar’s (2010, 2012) study, a more comprehensive set of criteria for measuring and managing trust in disaster management is provided by The Emergency Capacity Building Project (2007), an alliance of international NGOs who work in natural disaster response and recovery. Their research is targeted at enabling an NGO to build trust within a diverse team that responds to a natural disaster. The criteria they use to measure and manage trust are displayed in Table 1.6.a.
23 Table 1.6.a Trust criteria used to measure and manage trust for NGOs (ECB, 2007, p. 7)
Trust based on a perception that team members are competent and 1. Competence will not let a colleague down. 2. Openness with Trust based on the observation that other team members share information information important to the team proactively and clearly. Trust based on the observation that other team members: maintain 3. Integrity promises, are team-orientated, and behave towards a colleague in accordance with a moral code. Trust based on the observation that other team members are 4. Reciprocity trusting and co-operative towards a colleague. Trust based on background, values, approaches, interests, and 5. Compatibility objectives held in common. Trust based on the belief that other team members are concerned 6. Goodwill about a colleague’s overall welfare. Trust based on the observation that the behaviour of team 7. Predictability members is consistent over time and in different contexts. Trust arising from the feeling that a colleague has nothing to fear 8. Well-being from other members of the team. Trust based on the observation that other team members actively 9. Inclusion include a colleague in their social and work activities. Trust based on the observation that other team members share 10. Accessibility their true feelings and are personally relatable.
The Emergency Capacity Building Project have utilised the criteria above to produce a toolkit which can be used to measure and build trust within an NGO team. The strength of this toolkit is that it attempts to utilise these criteria to understand how trust changes within an NGO; rather than just asking if trust exists or not. However, although this study does provide 10 trust criteria, as with Chandrasekhar (2010, 2012), it does not provide weightings of importance for each criterion. Furthermore, it does not provide any insights into the factors that determine whether, and how, trust develops for local and international disaster management personnel charged with managing the response and recovery.
Therefore, although providing valuable insights into trust within natural disasters, none of these case studies provide results specifically concerning the process by which trust develops for the
24 imported international personnel and the local disaster management personnel; especially within a developing nation. It remains to be understood how trust forms for the local and international personnel within the EOC, charged with coordinating the national response. In particular, it is important to understand how trust forms for local and international disaster management personnel, given developing nations will continue to rely on the expertise provided by developed nations after a natural disaster and collaboration and coordination in disaster management relies on trust.
Problem statement and aims The gap that exists within the literature can be summarised as follows:
The enormous economic, social, ecological and health impacts of natural disasters are not equally spread, with developing nations on average suffering much higher mortality rates and economic losses as a percentage of national GDP than developed nations. While there has been considerable research in disaster management in developing nations, there has been little research into understanding the temporary group of local and international disaster management personnel in the EOC charged with coordinating the response. In particular, there has been virtually no research into the formation of trust between members of this group; who are critical in determining the effectiveness of the response and recovery.
Within this context, the aim of this research is to explore what factors impact trust formation for local and international disaster management personnel who are working together in the response and recovery after a natural disaster, in a developing nation.
More specifically, it will explore:
1. What are the characteristics of the temporary system of local and international disaster management personnel in the EOC, charged with coordinating the response?
2. What factors influence trust formation, and are these factors different for local and international disaster management personnel?
3. What factors may impact the resilience of trust formed for local and international disaster management personnel?
The overall objective of this research is to extend the theories of trust to a disaster management context in order to provide new conceptual insights for disaster managers when building trust
25 during a national disaster operation. The ultimate aim of this research is to enable disaster management personnel to more effectively develop trust within their response teams, leading to heightened cooperation and collaboration during the response and recovery. Ideally, this will lead to a better operation after a natural disaster.
26
THE THEORIES OF TRUST
Considering this research’s focus on trust formation in disaster management, this chapter provides a definition of trust and a critique of the major theories of trust in the context of temporary disaster management teams. It argues that none of these theories are appropriate for understanding how trust forms for local and international disaster management personnel during the response and recovery from a natural disaster. Conceptualising disaster management teams as a temporary system, this chapter contributes to disaster management literature by showing that the theory of swift trust in temporary systems is most appropriate to understand how trust forms in a disaster management context. The chapter concludes by presenting three research propositions that test the applicability of swift trust as a theoretical model to predict the formation of trust between disaster management personnel working on the response and recovery of a natural disaster.
Defining trust When defining trust, the literature currently has a “widespread lack of agreement on a suitable definition” (Hosmer, 1995, p. 380). However, the majority of existing definitions of trust define trust as either an intention or behaviour (McKnight & Chervany, 1996, p. 27):
Trusting intention – “the extent to which one party is willing to depend on the other party in a given situation with a feeling of relative security, even though negative consequences are possible.”
Trusting behaviour – “the extent to which one person voluntarily depends on another person in a specific situation with a feeling of relative security, even though negative consequences are possible.” 27 The key difference between these two categories is a ‘willingness to depend’ (an intention to trust) and actual ‘voluntary dependence’ (trusting behaviour). Within a disaster management context, an example of trusting intention would be a logistician personally deciding to trust that they will be given the correct list of supplies to be sent to an affected community by a colleague. An example of trusting behaviour would be the logistician actually sending the supplies because they trust the list provided by their colleague.
Regarding trusting intention, one of the most widely used definitions of trust is provided by Mayer, Davis, and Schoorman (1995, p. 712) who argue that trust is based on intentionality. They write that, “trust is not taking risk per se, but rather it is a willingness to take risk”. As such, they define trust as,
“the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party.” (Mayer et al., 1995, p. 712)
Multiple authors have used this definition in their work (McKnight, Cummings, & Chervany, 1998; Saunders, Dietz, & Thornhill, 2014; Wiethoff & Lewicki, 2005) and others have grounded their work in a definition of trust which is similarly based on intention and ‘willingness to be vulnerable’; and not on behaviour (Lewicki, McAllister, & Bies, 1998; Lewicki & Tomlinson, 2003; Mishra, 1996). In a similar vein to Mayer et al. (1995) and trusting intention, Rousseau, Sitkin, Burt, and Camerer (1998, p. 395) define trust as, “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behaviour of another.”
Contrary to these definitions, Fehr (2009) and Lewis and Weigert (1985) espouse the concept of trusting behaviour. The latter states that trust is a combination of three distinctive dimensions: cognition, emotion, and behaviour. They believe that, “these dimensions…are interpenetrating and mutually supporting aspects of the one, unitary experience and social imperative that we simply call ‘trust’” (Lewis & Weigert, 1985, p. 972). Put simply, Lewis and Weigert (1985) believe that a holistic definition of trust requires not only an intention to trust, but also trusting behaviour.
Thus, the literature recognises the significance of behaviour in trust, however it does so as a method of validation for the trustor concerning their original trusting intention toward the trustee (Crisp & Jarvenpaa, 2013; Hung, Dennis, & Robert, 2004; Lewicki, Tomlinson, & Gillespie,
28 2006; Luhmann, 1979; Mayer et al., 1995). That is, the trustor can determine whether their initial willing intention to be vulnerable to the trustee was appropriate based upon the behaviour of the trustee; and decide whether to continue trusting the trustee in the future (Mayer et al., 1995).
Although other studies demonstrate that behaviour is an important factor in trust, this research has chosen to use the definition provided by Mayer et al. (1995) as it is most appropriate to a disaster management context. This is because, in natural disaster management, it is likely that local and international disaster management personnel who are working together in an Emergency Operations Centre (EOC) during the response and recovery will not have met before (Smith & Dowell, 2000; Törnqvist, Sigholm, & Nadjm-Tehrani, 2009).
When coupled with the time and resource pressures posed by a disaster, and the different cultures, languages, priorities, and organisational systems thrown together in an EOC, new challenges are presented to the formation of trust, which are outside team contexts in permanent organisations where most definitions of trust have been developed to apply. That is, during the early stages of the response and subsequent recovery, there is little time for each party to develop relationships based on trusting behaviour. So, in these early stages, trust must be based primarily upon an intention to trust, and not trusting behaviour. Albeit over time, there will be opportunities to re-evaluate this initial intention with examples of the other party’s behaviour. As such, trust is an intention defined as,
“the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party.” (Mayer et al., 1995, p. 712)
The conditions for trust to exist The literature holds a general consensus that trust requires two primary conditions in order to exist: risk and dependence. In order for trust to be necessary, there must be an element of risk involved for the trustor (Kramer & Lewicki, 2010; Mayer et al., 1995). Luhmann (2000) argues that trust is necessary when there is a risk that more could be lost than gained from a situation. Secondly, trust requires a degree of dependence (Lewicki et al., 1998; Rousseau et al., 1998) where the trustor depends upon the trustee. There are two primary forms of dependence discussed in the literature. These are explained in the context of natural disaster management:
29 Inter and intra organisational dependence (McKnight et al., 1998; Mishra, 1996; Pirson & Malhotra, 2011) such as that between a local and international NGO providing clean water to a damaged community; or different national governmental departments needing to share data with each other concerning the number of affected communities and evacuees in emergency shelters.
Two-person dependence (or dyadic dependence) (Lewis & Weigert, 1985), which is two people depending on each other. This could be a local disaster manager depending on the expert knowledge of an international disaster manager in planning the response strategy within the EOC. Another example could include two logistics officers from different NGOs depending on each other to provide transport and communications support to colleagues in the field.
Taking these two pre-requisites for trust of risk and dependence and applying them to natural disaster management reveals an interesting gap in the literature. The practical risks associated with disaster response and recovery are well documented, such as financing the recovery, prevention of loss of life, infrastructure prioritisation (Blakely, Birch, Anglin, & Hayashi, 2011; Coppola, 2006; Lindell et al., 2007). Further, it is well documented that disaster management requires dependence between both personnel and organisations at a local, regional, national, and even international level (Coppola, 2006; Hannigan, 2012). However, what the literature does not currently address are which risks local and international disaster managers consider significant in a developing nation setting, and how their conception of these risks impacts trust formation.
Trust and distrust Trust is also contextual and task specific; that is, a person can be trusted to do one task, but not another (Mayer et al., 1995; Rousseau et al., 1998). In disaster management, a local disaster manager might be willing to trust an international disaster manager to help plan the response strategy but may not trust them to translate the local communications received from the damaged communities (as they may not speak the native language). As such, trust is not absolute. Rarely does a person completely trust someone else with any task, regardless of the task. Rather, trust and distrust are separate, independent constructs (Saunders et al., 2014, p. 662) that are awarded based upon the type of task and the risks associated. Given this, it can be argued that “low distrust is not the same thing as high trust, and high distrust is not the same thing as low trust” (Lewicki et al., 1998, p. 444). Using the example above, the local disaster manager may have high trust in the international disaster manager to aid the response strategy,
30 whilst also having high distrust in their ability to translate the local communications accurately. Hence, the local disaster manager both trusts and distrusts the international disaster manager; depending on the task to be completed and the risks associated with that task.
Furthermore, just as there is a lack of knowledge concerning what risks local and international disaster managers consider significant that require trust, there is also a lack of knowledge as to what factors impact trust formation for disaster management personnel working in an EOC. This is an important question for disaster management to answer given that, “trust is associated with enhanced cooperation, information sharing, and problem solving” (Lewicki & Tomlinson, 2003, p. 305); all of which are essential skills required within an EOC when responding to a natural disaster.
Theories of trust Lewicki and Bunker (1995, p. 134) write that the term trust, “has received attention from several of the social sciences… [and] each field has approached [trust] with its own disciplinary lens and filters.” As such, this section reviews these theories under three main headings (sociological, economic and organisational theories of trust) and shows that none of these theories provide an ideal fit for the unique situation of international and local disaster managers working together within an EOC. It then introduces the theory of swift trust in temporary systems as a significant conceptual framework, as it correlates most closely with how trust could develop for local and international disaster managers who have not worked together before, operating out of an EOC.
2.4.1. Sociological theories of trust Many sociological theories of trust are concerned with the relationships between people within a society (termed a social system), how those people trust each other, and how that social system trusts other social systems. The following section addresses each of the major sociological theories of trust and their applicability to a disaster management context.
Parsons’ theoretical model of trust Parsons (1963) theoretical model of trust is built upon systems theory, with a focus on action theory and understanding how motives influence the decisions a person makes. Parsons (1963) argued that a society is governed by patterns of culture, and those patterns are formed by the consensual sum of the individuals within that social system. Put simply, the values of a society will be those of the people within the society. In such a system, Parsons (1963) argued that trust was necessary to ensure that each person would comply with these standardised values. Applying Parsons’ (1963) theory to the context of the disaster management, the values of the
31 EOC would be produced by both the local and international disaster management personnel, and trust would be necessary to ensure that all personnel complied with such values.
Parsons’ (1963) use of action theory led to him providing some further detail around what allows someone to trust. He stated that trustworthiness is based upon reputation, and reputation can be based upon competency. In particular, he argued that trust has the greatest chance of developing when there is no intention for deception. Parsons (1963, p. 49) also stated that there is a relationship between a shared purpose or “common belongingness” and trust. As such, his theoretical framework would suggest that local and international disaster managers would develop trust under three conditions: when they share a common purpose, where they have no reason to deceive each other, and when they have a reputation for being competent. He argued that holding a common purpose and being identified with a group can lead to the members of that group perceiving each other as more trustworthy than people not in their group (also argued by M Williams (2001)). Similarly, the influence of reputation and competency on trust has now been generally accepted within the literature (Coleman, 1990; Lewicki et al., 1998; Mayer et al., 1995; McAllister, 1995; McKnight & Chervany, 1996; McKnight et al., 1998; Mishra, 1996; Rousseau et al., 1998; Schoorman, Mayer, & Davis, 2007; Wiethoff & Lewicki, 2005). However, although such factors would seem applicable for trust formation in an EOC, the foundation of Parsons’ theoretical framework is time. He argues that trust is dependent upon an established set of consensual values determined by the members of a social system. Yet, the temporary nature of the EOC means that there is little time for a consensual sum of values to be agreed upon (consciously or not) against which trust can be evaluated. Secondly, the EOC can also be subject to competing priorities of the included personnel concerning the dynamic requirements of a national response. Therefore, Parsons’ model of trust is deemed inappropriate to model trust formation in the EOC, given its lack of adaptability to a temporary system that is subject to continually changing priorities.
Luhmann’s theoretical model of trust Luhmann (1979) provides a sociological analysis of trust through a combination of functionalism and systems theory. He believes that each person lives within a social system that is familiar to them; they know, and are familiar with, the people and places it contains. However, outside of their system is the ‘wider system’ of general society which is unfamiliar as it contains: new people, new relationships, and new places. In someone’s life, it is near impossible to avoid interacting with this wider system; simply meeting new people is going outside a person’s system of familiarity and the number of ways a person could interact with the wider society is near limitless. Luhmann (1979) calls this concept ‘complexity’ and goes on to argue that trust is
32 necessary to help manage complexity. A person must exhibit a certain level of trust when forming a potentially new relationship, or venturing outside their own social system, or they would never be able to leave their system of familiarity. Luhmann states:
“The world is being dissipated into an uncontrollable complexity; so much so that at any given time people are able to choose freely between very different actions…[however] trust, by the reduction of complexity, discloses possibilities for action which would have remained improbable and unattractive.” (Luhmann, 1979, p. 24)
Luhmann (1979) argues that if people were not able to trust, then every new interaction would be based entirely on hope; hoping that another person is trustworthy or hoping that a new location is safe. However, trust allows people to interact in new and unfamiliar situations with a confidence greater than that which can ever be provided from hope alone. He writes that, “trust reduces social complexity by going beyond available information and generalising expectations of behaviour in that it replaces missing information with an internally guaranteed security” (Luhmann, 1979, p. 103). In other words, trust allows a person to confidently engage with another person even when they do not know everything about the other person. In a disaster management context, applying Luhmann’s (1979) theory, trust would enable local and international disaster management personnel to collaborate and manage the response even though they know little about their colleague other than that they were selected to be included within the EOC due to the technical expertise or political authority. Trust is necessary to manage the complexity inherent within this environment, allowing disaster management personnel to meet and work together within the EOC.
However, although a person does not require complete information, Luhmann (1979) argues that a person will determine whether to trust based upon:
the consistency of the other person’s behaviour,
their personal familiarity with the other person, and,
their own predisposition to trust.
Considering the consistency of the other person’s behaviour, Luhmann (1979) argues that if someone does as they say, they are more likely to be trusted. Furthermore, the literature widely supports the idea that a person’s predisposition to trust will influence their initial decision to trust someone else (B. Barber, 1983; Gambetta, 2000; Hung et al., 2004; Lewicki & Tomlinson, 2003; McKnight et al., 1998; Tatham & Kovács, 2010). Interestingly, it is argued that people more
33 readily start a relationship by trusting the other as opposed to distrusting (B. Barber, 1983; Luhmann, 1979; Wiethoff & Lewicki, 2005). Similarly, familiarity influences trust (Lewicki & Tomlinson, 2003).
Specifically in the context of research question 2, ‘what factors influence trust formation and whether these are different for local and international disaster management personnel?’; Luhmann (1979) suggests three factors. However, to assume that these three factors are all that influence trust in every situation seems simplistic for a number of reasons. Firstly, given that local and international disaster management personnel in an EOC have probably not met and have only just started working together (Smith & Dowell, 2000; Törnqvist et al., 2009), there is no substantial information available for the first two arguments in Luhmann’s (1979) theoretical model. That is, there has been little time for an assessment of the consistency of the other person’s behaviour or to become personally familiar. Therefore, Luhmann’s (1979) theoretical model would either predict that trust would always be low in an EOC because two of the three criteria cannot be met, or suggest that the decision to trust is based entirely on a person’s predisposition.
Secondly, Luhmann’s (1979) theoretical model lacks specificity. He does not provide sufficient detail about what type of consistent behaviour the other person has to display, or how personally familiar someone must be with another person in order to trust them. For example, contrary to Luhmann (1979), the literature generally agrees that trust can be awarded to someone based on their reputation, without the trustor ever personally meeting the trustee (Mishra, 1996; Rousseau et al., 1998).
Thirdly, he does not provide a weighting to each of these factors, such that it could be assumed that they are all equal in determining trust. This seems improbable, given these trust factors do not consider the context in which trust forms. Given the EOC is a unique, temporary system formed specifically for the disaster response and recovery, the influence of this context on trust formation must be considered; as per research question 1. As such, it does not seem appropriate to use Luhmann’s (1979) theoretical model as a means to investigate trust formation for local and international disaster management personnel given its lack of applicability to new relationships and adaptability to unique contexts; both of which are aspects pertinent in natural disaster management.
Barber’s theoretical model of trust Another widely used theory of trust in sociology is proposed by Barber (1983). Building off the theoretical model proposed by Luhmann (1979), he provides an analysis of the role of experts
34 in society, such as disaster managers, and trust. He concluded that, given the complexity of society, individuals developing expertise would be common, and therefore, trust in experts would be a necessary part of any society. He wrote, “where there is such an accumulation of knowledge and technical expertise, expectations of trust in this sense are very common” (B. Barber, 1983, p. 14). He argued that trust is necessary in social systems for two reasons:
Individuals within a society must trust that those with knowledge and expertise are not going to abuse this power; and,
The presence of trust between members shows that they have shared values and common goals.
Although Barber (1983) states that expertise and trust are related, this remains only one factor that is applicable to trust formation for local and international disaster management personnel in the EOC. Barber’s (1983) theoretical approach does not provide a holistic analysis of trust formation; nor does it consider how a disaster management context may influence this process. Barber (1983, p. 14) summarises this best when he writes that, “trust…has the general function of social ordering…[and provides] cognitive and moral expectational maps for actors and systems as they continuously interact” (B. Barber, 1983, p. 14). However, in terms of trust in a disaster management context, what these cognitive and moral expectations are remains unclear and requires investigation.
Lewis and Weigert’s theoretical model of trust It is important to reiterate that international and local disaster management personnel are more likely to be establishing a new relationship rather than building upon a previous one where trust may already exist. However, sociologists Lewis and Weigert (1985), drawing on Parsons (1963) and Luhmann (1979), state that trust requires some previous interaction to be cognitively possible. They write, “some prior experience with the object of trust is a necessary condition for establishing the cognitive element in trust” (Lewis & Weigert, 1985, p. 970). They argue that the decision to trust is made through the joint evaluation of cognitive, emotional and behavioural elements. Conversely, much of the present literature argues that trust does not require prior interaction. If using the model proposed by Lewis and Weigert (1985), local and international disaster management personnel would require some form of interaction before trusting each other. This is not to say that ongoing interaction does not influence trust; indeed, Gambetta (2000) provides an explanation of the formation of trust over time. However, disaster personnel are often under time pressure to provide the response and recovery (Paton & Flin, 1999; Smith & Dowell, 2000) with little (if any) prior personal interaction with their colleagues. As such, what
35 remains to be understood is if trust exists between disaster personnel during the response and recovery, and if so, how can this trust form so quickly? Although Lewis and Weigert (1985) allude to competency and integrity as factors that influence trust, they offer little in terms of how local and international disaster personnel might develop trust. This is because their theoretical model does not account for the time sensitive working environment that disaster management personnel operate within, where trust could occur without prior interaction.
Coleman’s theoretical model of trust Coleman (1990), in his work ‘Foundations of Social Theory’, provides a mathematical model of trust, based upon the theory of action in social systems. He stated that trust involves three parameters; p (the probability that the trustee is trustworthy), L (the potential loss), and G (the potential gain). Coleman (1990) argued that the probability of trust (p) can be influenced by the attainment of information, and the degree to which someone may seek information about another person is dependent upon the potential loss ‘L’ and gain ‘G’ from the interaction. Coleman’s (1990) calculative framework goes on to state that trust can be dependent upon: the potential of future interaction, sanctions for breaking such trust, and the competency and integrity of the other person (termed performance capability).
In a disaster management setting, given that disaster personnel may not ever meet again, it is unclear from Coleman’s (1990) model how this will affect trust. If disaster management personnel may not work together again, is the probability of trust low or high according to his theoretical framework? Conversely, the literature does recognise the role sanctions play in establishing trust; particularly the organisational literature, which considers the role of contracts and terms in establishing trust (McKnight & Chervany, 1996). What remains to be answered is whether such sanctions exist and are enforced during the response and recovery from a disaster, and if so, how these sanctions impact trust formation for local and international disaster personnel.
Apart from these contributions, Coleman’s (1990) most applicable addition to trust in disaster management is his concept of an ‘advisor’: an intermediary between a trustor and trustee. As previously stated, many disaster personnel may not have worked together before, and their inclusion into a disaster management team, and subsequent trust in other team members, may occur through an intermediary. In other words, the trustor trusts the judgement of the intermediary, who trusts the performance capability of the trustee. Therefore, the trustor can trust the performance capability of the trustee because of the trust the trustor has in the intermediary’s judgement. It is unknown, however, whether the model of an intermediary
36 proposed by Coleman (1990) is valid for disaster management within an EOC, and how significant a role an intermediary would play in this emergency environment.
Gambetta’s theoretical model of trust Similarly to Coleman (1990), Gambetta (2000) used systems theory to focus on the probability of trust forming between two people. His theoretical framework argues that trust can be determined using probability, where trust is considered as a value between two thresholds: 0, which is complete distrust; and 1, which is complete trust. Gambetta (2000) argues that trust is the necessary response to having limited knowledge of the intentions of another person. As such, the degree of knowledge possessed by the trustor directly influences the probability that they will trust the trustee. That is, if the trustor had complete information, then they could decide to completely trust or distrust the trustee based on this information. For example, in a disaster management context, a disaster manager would use any available information, such as a colleague’s parent organisation, to determine the intentions of their colleague and whether or not to trust them.
This then raises the question of, ‘what probabilistic value of trust is sufficient for the trustor to trust the trustee and how is this determined?’ Gambetta (2000, p. 219) provides brief examples of what influences trust, which can be summarised as one’s disposition to trust, as well as the context in which the relationship exists. However, he states that, “we still know little…about how a certain level of trust is or can be achieved and promoted…[Furthermore] the optimal threshold of the probability of believing we trust someone enough…will not be the same in all circumstances” (Gambetta, 2000, pp. 219-220). As such, the model presented by Gambetta (2000) in systems theory does not provide sufficient detail regarding what factors promote trust for local and international disaster management personnel, nor does it detail what risks are involved in a relationship that requires trust to ensure cooperation. However, his theoretical framework does validate the necessity for further research into trust in an EOC, as it acknowledges trust to be contextual and situationally specific.
2.4.2. Economic theories of trust Theories of trust have also been applied in the field of economics. Knack and Keefer (1997) measured stronger levels of trust in countries with higher and more equal incomes, higher levels of education, and ethnically homogenous populations. Similarly, Porta, Lopez-De-Silane, Shleifer, and Vishny (1996) argued that trust improves the growth of gross national product (GNP) and improves wider national economics. They summarised their findings by stating that, “trust enhances economic performance across countries” (Porta et al., 1996, p. 336). However,
37 neither of these authors have provided specific details concerning trust formation between local and international disaster management personnel within the EOC. However, within economics, there is a common method of data collection used to understand economic theories of trust, called game theory (Osborne, 2004).
Trust using Game theory Game theory is broadly defined as, “the logical analysis of situations of conflict and cooperation” (Straffin, 1993, p. 3). More specifically, the games that investigate trust generally require at least two players, who use strategies to make decisions which determine the outcome of the game and the associated consequences for each player (Straffin, 1993). The most widely used game investigating trust is an investment game proposed by Berg et al. (1995). It operates as follows:
Person A starts with a monetary sum. They can choose how much of this they want to invest with person B. Any money sent to Person B will be tripled. Person B can then choose how much money to send back to Person A. Trust and trustworthiness are measured by observing how much money is sent and/or kept by each party.
The dynamics of such a game can be altered in certain ways to measure different factors of trust and trustworthiness. Examples include providing further information about person A and B to the other to help inform their decisions. Likewise, anonymity can be ensured so that factors such as reputation do not play a role. Furthermore, the game can be played more than once to see the effects of repeated interactions on trust development.
Considering the investment game and the assumption of self-interested behaviour in economics (Force, 2003), it would be predicted that no money would be sent by person A to person B. However, this was not the case (Berg et al., 1995). As such, it was seen that trust in person B by person A enabled cooperation, and as Berg et al. (1995) showed, a better return on investment than if no money was sent. Similar investigations into trust and modes of cooperation using game theory include games such as the Prisoner’s Dilemma (Kreps, 1990; Straffin, 1993) or the Peasant-Dictator game (Van Huyck, Battalio, & Walters, 1995). Interestingly, the latter found that self-interest and zero investment was the optimal solution; contrary to Berg et al. (1995).
As highlighted in Chapter 1, the works by Cassar et al. (2011), Castillo and Carter (2011), Veszteg et al. (2014), Fleming et al. (2014) and Conzo (2014) all utilised game theory and the method provided by Berg et al. (1995) to understand trust in natural disasters. However, each of these studies was only able to show that trust changed in their respective circumstances due to a natural disaster. Their use of game theory did not explore the factors that influenced trust and
38 trustworthiness, nor did they consider trust for local and international disaster management personnel. Furthermore, game theory measures trust and trustworthiness between two people without providing sufficient detail as to why the decisions taken by person A and B were made. As Fehr (2009, p. 253) states, the trust game, “provides a suitable behavioural measure of trust and it provides the opportunity to elicit a clean - preference-free - measure of beliefs about the trustworthiness of anonymous strangers.” However, it does not provide any detail into what influences these beliefs.
Considering local and international disaster personnel, the use of game theory would require ongoing monitoring of how the decisions of one party affects how the other reacts, and so on. This would enable a successful measurement of whether the local disaster personnel trust the international disaster personnel, and vice versa. However, as already argued, there is a gap in the literature regarding what factors influence this trust. There is a requirement to understand what the risks each of these parties consider prevalent while working together in a disaster management context; which ultimately influence their trust in one another. Game theory is insufficient in providing the depth necessary to answer these questions concerning trust formation.
2.4.3. Organisational theories of trust There has been much research in recent decades to understand the role of trust in organisations, with the aim of improving organisational cooperation and efficiency. This research has utilised stakeholder theory as well as management theory, although the primary theory used has been organisational theory. Organisational theory attempts to, “explain and predict how organisations and the people in them will behave in varying organisational structures, cultures, and circumstances” (Shafritz, Ott, & Jang, 2015). As organisations rely on cooperation, Hosmer (1995, p. 399) highlights the benefits of trust to cooperation when he writes that, “trust in organisational theory is the expectation of a similar behaviour that recognises and protects the interests of other people in order to increase willing cooperation and expand ultimate benefits within a joint endeavour or economic exchange.” Hosmer’s (1995) comment provides the foundation for conceptualising local and international disaster management personnel cooperating on a ‘joint endeavour’ to produce the benefit of a successful response and recovery. However, Hosmer’s (1995) work has been developed for permanent business organisations rather than the temporary organisational setting of an EOC. As explained in Section 1.5, the EOC is a unique organisation type, defined by its focus on collaboration and cooperation, its temporary nature, and its new multi-national workforce. Unlike contemporary organisations, the EOC is not focussed on economic exchanges for profit or on-going growth and expansion.
39 Therefore, when considering the organisational trust literature, it is necessary to remember the unique qualities of an EOC and how these can impact trust formation for local and international disaster management personnel.
This said, unlike the sociological theories of trust, the organisational literature is much more concerned with what factors influence trust and how an intention to trust can develop. By learning how trust can be formed between two people within an organisation, it may then be possible to adapt this concept to how trust forms for local and international disaster management personnel within the context of the EOC and its organisational attributes.
Antecedents of trust from organisational theory The intention to trust, according to Schoorman et al. (2007), forms through the trustor’s judgement of the trustee’s ability (or competency), benevolence, and integrity. These three antecedent conditions of trust are common to much of the literature considering trust between two people either within an organisation, or from different organisations (Ferrin & Gillespie, 2010; Lewicki & Tomlinson, 2003; Pirson & Malhotra, 2011; Rousseau et al., 1998; Saunders et al., 2014; Wiethoff & Lewicki, 2005). Given the widespread acceptance of these three terms as antecedents for trust, they are defined as follows (Mayer et al., 1995):
Ability – the group of skills, competencies, and characteristics that enable a party to have influence within some specific domain. This is also referred to as competency.
Benevolence - the extent to which a trustee is believed to want to do good to the trustor; the perception of a positive orientation of the trustee toward the trustor.
Integrity – the trustee’s adherence to a set of principles that the trustor finds acceptable.
Although these three terms are the most common to the organisational literature, there are also references to other antecedent conditions for trust between two people. McAllister (1995) states the importance of responsibility, reliability and dependability. Rousseau et al. (1998) similarly emphasise reliability and they, with Mishra (1996), also include openness/honesty and reputation. Lewicki et al. (1998) also consider the importance of reputation, and Wiethoff and Lewicki (2005) state that reputation can be destructive by creating distrust between two parties. Pirson and Malhotra (2011) believe that transparency and having similar values are antecedents for trust. McKnight and Chervany (1996) recognise the importance of predictability, whilst Lewicki et al. (1998) approach the antecedents for trust differently, emphasising the importance of personality, social similarities and the wider context. Although there are some differences in
40 what the organisational theorists emphasise as important antecedents for trust, the results of these studies outlined above represent a significant step toward a unified understanding of what causes trust, and, “whilst it is premature to suggest that there is total agreement on a conceptualisation of the various forms of trust, a significant result of this interest has been a clearer understanding of their antecedents and outcomes” (Tatham & Kovács, 2010, p. 2).
However, considering the three primary antecedents of ability, integrity and benevolence, several authors state that some are more important than others, depending on the stage of the relationship between the trustee and trustor. That is, early in a relationship, ability and integrity take more precedence in determining trust, whilst as a relationship develops, benevolence becomes more important (Lewicki & Tomlinson, 2003; Mayer et al., 1995; Pirson & Malhotra, 2011). As such, for international and local disaster personnel who may not have met, yet are working together, the organisational literature would suggest that ability and integrity would be the most important antecedents in establishing trust.
Trust, however, is not just based upon the qualities of the trustee; rather, it also depends on the trustor. For example, Mayer et al. (1995) highlight the importance of the trustor’s disposition to trust in enabling trust to develop. Lewicki and Tomlinson (2003, p. 2) argue that someone’s disposition to trust is a, “function of the degree to which trust has been honoured in that individual’s history of prior interactions, and may have its most profound effect in novel or ambiguous situations”; such as in managing the response to a natural disaster. In other words, a person’s propensity to trust provides an assumed list of the characteristics they expect in someone else (Lewicki & Tomlinson, 2003), which then influences their intention to trust.
Types of trust in organisations The organisational literature shows that three different types of trust can develop between two people in an organisational setting: deterrence based trust, situational based trust, and rule based trust.
‘Deterrence based’ trust is defined as trust in another, “because the costly sanctions in place for breach of trust exceeds any potential benefits from opportunistic behaviour” (Rousseau et al., 1998, p. 398). This type of trust is built upon contracts and agreements which the trustor can rely upon to ensure the trustee acts as expected (sometimes also called ‘systems based’ trust (McKnight & Chervany, 1996)). Although, as Rousseau et al. (1998) comment, complete reliance upon such a type of trust may not be trust at all, as trust is built upon costly sanction rather than the relationship. Furthermore, given the time pressures of disaster management, a contract may
41 not exist, or may be formed much later after a relationship (and subsequently trust), has already formed.
A second type of trust that may form in an organisational setting is ‘situational based’ trust. This type of trust forms when, “one has formed an intention to trust every time a particular situation arises, irrespective of one's beliefs about the attributes of the other party in the situation.” (McKnight & Chervany, 1996, p. 38) For example, an international disaster manager has chosen to always trust the local leadership within the EOC, even though they do not know their qualifications or competency. Kramer and Lewicki (2010) extended situational based trust to develop the third type of trust called ‘rule based’ trust. Rule based trust is the trustor’s trust in the trustee, due to the trustor’s knowledge of how the trustee fits within the organisation (Kramer & Lewicki, 2010). Considering disaster management and the fluid grouping of personnel responding, it seems that rule based trust is less applicable than situational trust in a disaster management context.
The importance of ‘role’ for trust within organisations Organisational theory also specifically highlights the importance of role for trust development. As Kramer and Lewicki (2010, p. 263) state, “individuals can trust presumptively on the basis of their knowledge of role occupancy and the system of role relations, even in the absence of personalised knowledge regarding the individual in the role.” In other words, the authors agree that if the trustor knows how the trustee fits within an organisation and what their role is, then it becomes easier to trust them; even though they do not have a direct relationship with that person or a concept of how that person will inhabit that role. Therefore, within the EOC, organisational theory argues that local and international disaster management personnel may form trust based on each other’s role in the operation; even if they do not know each other personally.
In addition to the significance of knowing the trustee’s role when forming trust, Kramer and Lewicki (2010), MacDuffie (2011) and McKnight et al. (1998) state that people are likely to extend trust to people who are in the same ‘in-group’ as them; i.e., if people are linked by social similarities or are in the same organisation. This is particularly significant for trust in a multi- national response, given the differences between local and international disaster management personnel in: culture, parent organisation and reporting lines, language, priorities of the response and recovery, or technical expertise. Building on this, McKnight et al. (1998) emphasise how stereotyping can impact trust formation for new relationships, such as those within an EOC.
42 Lastly, organisational theory also argues that trust in an organisational setting can be influenced by the organisation’s leadership and whether a leader creates an environment which is conducive to trust (Kramer & Lewicki, 2010). However, for an EOC, although disaster personnel may work under the head of the EOC, they may still be required to report to their parent organisation (Kendra & Wachtendorf, 2003) and act as conduits between the EOC and the wider network of responding agencies (Curnin & Owen, 2013). Therefore, there remains a gap in understanding how temporary authority, such as that of the head of the EOC, can influence trust formation for local and international disaster management personnel brought in to work in the EOC.
Organisational theories of trust applied to disaster management Organisational theories of trust state that trust develops over time and through repeated interactions, moving from cognitive appraisal of the other person into a more emotional bond. The three primary antecedents to this trust formation are ability, integrity, and benevolence. Ability and integrity play a more significant role during the cognitive appraisal stages of trust, with benevolence becoming more significant once a deeper, emotional trusting relationship has developed. Considering the previous analysis, organisational theory argues that:
Trust may be affected by the structure of the organisation and the perceived expectations of how people should operate within such an environment.
Trust may formally or informally be affected by sanctions.
An organisation can enable trust, especially within the early stages of a relationship, to be formed based upon the trustee’s role and performance within the organisation, and not on their personal characteristics. Similarly, stereotyping can also influence trust formation.
Trust can be promoted between employees due to their inclusion within an organisation.
Leaders within an organisation have a role to play in facilitating trust and creating an environment which fosters trust.
As such, the question remains regarding how organisational theory helps to consider trust formation for local and international disaster management personnel working together in a developing nation within an EOC during and after a natural disaster?
43 Firstly, unlike the sociological and economic theories of trust, organisational theories focus on the antecedent conditions of trust: primarily ability, integrity, and benevolence. Secondly, as an EOC has some similarities to an organisation in structure and mode of operation, the factors influencing trust in an organisational setting (such as leadership, ‘in-group’ membership, sanctions, role, expected performance, and competency) should all be considered as potential influencers in trust forming for local and international disaster management personnel.
However, there is one main idea that the organisational literature has not focussed on significantly in the formation of trust in organisations such as an EOC: the EOC as an ‘organisation’ is, as Militello et al. (2007, p. 25) states, “ad hoc” and temporary. That is, the team will come together, with many meeting for the first time, complete the response and recovery, and then be disbanded. What organisational theory has not adequately addressed is this type of ‘temporal workplace’, with first-time encounters as the primary relationship type. There remains a gap in the organisational literature to consider trust formation between local and international disaster management personnel within the ‘temporary organisation’ of an EOC.
The theory of swift trust in temporary systems Unlike the organisational trust literature, Meyerson et al. (1996) developed the theory of swift trust in temporary systems to address formation of trust between personnel working together in a temporary setting. This theory is particularly valuable to understanding trust formation in a disaster management context, because it accounts for the temporary nature of an organisation such as an EOC and considers the subsequent impacts of this unique characteristic on trust formation.
Swift trust is an impersonal trust that is necessary between members of a temporary system who have come together in order to complete a task, but who have not had the chance to develop trust over time (Kramer & Cook, 2004). These are two defining characteristics of an EOC, although swift trust theory has never been comprehensively used in this context before. An example of the application of swift trust theory outside a disaster management would be the trust formed between the cockpit crew of an aeroplane who have not met before, yet have to trust each other to complete their respective roles in flying the plane (Meyerson et al., 1996). The theory indicates that the key differential for swift trust in temporary systems compared to other types of trust is that it requires people to, “wade in on trust rather than wait while experience gradually shows who can be trusted and with what: Trust must be conferred presumptively or ex ante” (Meyerson et al., 1996, p. 170). However, in order to understand swift
44 trust in temporary systems and how it applies to natural disaster management, it is necessary to first understand what constitutes a temporary system.
2.5.1. Defining temporary systems According to Goodman and Goodman (1976, p. 494), temporary systems are defined as, “a set of diversely skilled people working together on a complex task over a limited time period.” A temporary system displays four key characteristics (Goodman & Goodman, 1976):
a high degree of interdependence is required,
the task is non-routine for the organisation,
the task is deemed important enough that a temporary group is needed to address it, and,
clear goals are needed to define the task and provide a time limit for completion.
In addition, Packendorff (1995, p. 327) writes that, “other attributes [of temporary systems] are that goals are usually more specified, [and] the personnel more likely to be recruited because of task-relevant competence...”. Furthermore, Meyerson et al. (1996) state that temporary systems generally consist of personnel:
from a small labour pool and overlapping networks, due to the need for specialised expertise;
who have a limited prior working history but the potential for working together again in the future; and,
who have been brought together by some leader (which they define as a ‘contractor’).
Piecing these authors’ work together, it can be concluded that a temporary system has the following attributes:
1. Formed to address a task that is non-routine and requires interdependence;
2. A task that is defined by clear and specific goals and has a time limit on its duration;
3. Involves diversely skilled people from a small labour pool and overlapping networks;
4. Members have a limited work history but the potential to work together again; and,
5. Assembled by the contractor.
45 2.5.2. Disaster management and the EOC as a temporary system The following section aims to provide an original contribution to the literature by showing that an EOC can be classified as a temporary system. Pérez-Nordtvedt, O’Brien, and Rasheed (2013) have argued that temporary systems are valuable in responding to natural disasters. In particular, Ryan (2013) has classified an EOC as a temporary system. However, neither of these authors has justified their conclusions with a critical appraisal, which requires comparing the criteria for a temporary system with that of an EOC. Thus, the criteria for a temporary system has been summarised in Table 2.5.a, and the literature concerning natural disaster management and EOCs has been collated to show how an EOC theoretically qualifies as a temporary system in response to each of these criteria.
46 Table 2.5.a Qualification of an EOC as a Temporary System
Criteria of a Temporary System How an EOC qualifies as a Temporary System An EOC is an ad hoc, co-located group of people assembled in a central location tasked with coordinating the response to a natural disaster (Kendra & Wachtendorf, 2003; Militello et al., 2007; R. W. Perry, 1995; Wildman et al., 2012). This requires interdependence and communication between political leaders, response agencies, the general public, and external sources of assistance, (R. W. Perry, 2003; Stephens & Grant, 2001) such as international disaster personnel. “The EOC concept allows for interpersonal 1. Formed to address a task that is communication, technically supported information exchange, and decision-making among the non-routine and requires representatives of different agencies” (Kendra & Wachtendorf, 2003, p. 39). interdependence The consequences of a natural disaster are complex (Alexander, 2008), with non-routine challenges arising from coordinating all the responding agencies and governmental departments (E. L. Quarantelli, 1988) during, “an evolving dynamic situation” (Kendra & Wachtendorf, 2003, p. 39). Furthermore, a national response may be considered non-routine due to the inclusion of international staff and other responding agencies (Curnin & Owen, 2013) coming together to help the permanent EOC staff with the demands of the response. The clear goal of the EOC is to successfully coordinate the response to a natural disaster. In order to achieve this specific goal, an EOC has many smaller, specific goals. According to R. W. Perry (1995), the primary goals are to: create and implement policies to guide the response and recovery, oversee the management of disaster operations, and, 2. A task that is defined by clear and collate and disseminate information related to the response and recovery. specific goals and has a time limit on Recovery from a natural disaster, and subsequent EOC operations, generally last between 12 and 36 its duration months, “a period defined by the birth and death of the organisations charged with managing the effort” (Anglin, 2011, p. 77). Furthermore, each task within the EOC may be time limited. As R. W. Perry (1995, p. 39) writes, “typically the constellation of disaster agent-generated demands changes as time passes from the time of impact; demands associated with initial impact may decline while new demands arise from secondary threats.”
47 Criteria of a Temporary System How an EOC qualifies as a Temporary System The members of an EOC are selected due to their specific competence in response and recovery. An EOC, “is the place where technical emergency management directly interfaces with elected political authorities to form legitimate emergency authority and expertise” (R. W. Perry, 2003, p. 151). The labour pools and networks from which people join an EOC are small and consistent, with personnel being chosen from the 3. Involves diversely skilled people from primary organisations that usually respond to a natural disaster (R. W. Perry, 1995). As such, after a natural a small labour pool and overlapping disaster, the members of an EOC would anticipate working again with the same organisations and their networks respective representatives (Kendra & Wachtendorf, 2003, p. 37). These expected organisations generally include: fire, police, and ambulance services; primary industries and critical infrastructure (such as water and energy); governments (Kendra & Wachtendorf, 2003, p. 39; R. W. Perry, 2003); and other NGOs and INGOs. Given that, “large incidents that absolutely require an EOC are infrequent” (R. W. Perry, 2003, p. 151), the 4. Members have a limited working change that members of an EOC have previously worked together is limited. However, given their level of history but the potential to work expertise, if required for another disaster response, the same personnel are likely to be called upon again together again to work together. Simulated responses to natural disasters are one way of building a working relationship between those who would typically be involved in the EOC (R. W. Perry, 2004). Members of an EOC may be asked to be a part of the ad-hoc team by a leader called the ‘contractor’. Once part of the EOC, the members report to this EOC leader. This leader may assume the title of disaster 5. Assembled by the contractor recovery manager (Anglin, 2011) or EOC commander (R. W. Perry, 2003). The EOC leader may also be supported by a committee (R. W. Perry, 2003): an example is the Mayor’s Office of Emergency Management (OEM) in New York City (Kendra & Wachtendorf, 2003).
48 2.5.3. The theoretical foundations of swift trust The theory of swift trust in temporary systems has been developed by Meyerson et al. (1996) through a culmination of the research conducted by Baier (1986), Gambetta (2000) and Luhmann (1979); the latter two having already had their theories of trust discussed under Sections 2.4.1.1 and 2.4.1.3 respectively. Baier (1986, p. 235) defines trust as, “accepted vulnerability to another’s possible but not expected ill will (or lack of good will).” She states, “intentional trusting does require awareness of one’s confidence that the trusted will not harm one, although they could harm one” (Baier, 1986, p. 235). Meyerson et al’s (1996) concept of swift trust in temporary systems draws upon this idea of untrustworthy behaviour (ill will) and vulnerability.
One condition influencing vulnerability within a disaster management team is a small labour pool. Most disaster management personnel remain on a roster until being deployed to respond: “Whilst some disaster response staff within a humanitarian organisation are permanent employees, many are drawn from a wider network of logisticians who form an ‘on call’ roster such as those held by Oxfam, RedR and the Red Cross movement in many countries” (Tatham & Kovács, 2012, p. 180). As such, a small labour pool increases the importance of a person’s reputation, due to the lower degree of separation between participants within that pool of expertise (Meyerson et al., 1996). Put simply, people within the labour pool of disaster management are more likely to know other people in the same network, and therefore reputation is important for continued inclusion within that network.
Although disaster management personnel may feel this reputational vulnerability, Meyerson et al. (1996) argue that they can reasonably expect to not be undermined by the other members of the temporary network. This is due to three primary reasons:
1. Untrustworthy behaviour can jeopardise the whole temporary system, which also affects the perpetrator.
2. The chance of potential future interactions due to a small labour pool.
3. The importance of role – that is, people are judged more by what they do than their personality simply because there is not enough time to make appropriate judgements. Also, roles attract stronger trust because they generally require qualifications which people can rely upon (e.g. degrees, accreditation). Therefore, if roles are well defined and clear to people within the temporary system, expected ill will decreases and trust is more likely.
49 Furthermore, it is important to add that when people consider untrustworthy behaviour, their expectations, “tend to be imported from other settings and are imposed quickly in categorical forms” (Meyerson et al., 1996, p. 174). In a temporary system, the trustor does not have the chance to continuously monitor the trustee’s behaviour, and as such, the trustor relies upon their predisposition and categorical assumptions in deciding whether to trust. Examples of categorical assumptions include: the institutional context of the temporary system, industry norms, occupations, or even identity-based stereotypes (amongst others). Conversely, given that these categorical assumptions can be influential in the building of swift trust, they can also damage swift trust if the trustee does not behave in accordance with the assumptions the trustor has made (Meyerson et al., 1996).
The use of categorical assumptions in influencing trust is also mentioned by Gambetta (2000, p. 223), and Meyerson et al. (1996) argue that these assumptions can lead to a polarising decision to trust (or distrust) in a temporary system without substantial evidence:
“People fall back on predispositions, categorical assumptions, and implicit theories to move them toward the greater certainty of clear trust or clear distrust…An individual’s associates in a temporary system function essentially as a pretext to access over-learned tendencies and cognitive structures that provide guidelines for trust or mistrust.” (Meyerson et al., 1996, p. 178)
Furthermore, drawing on Baier (1986), Meyerson et al. (1996) argue that a trustor within a temporary system should only moderately depend on any trustee. They argue that this degree of dependence is ideal, as it allows the trustor to maintain a degree of control whilst still being vulnerable; given their access to only categorical information about the trustee.
The final major contribution to swift trust is from Luhmann’s (1979) theoretical model, which argues that trust is necessary in order to engage with situations that are unfamiliar to a person, and therefore pose risk. Temporary systems generally comprise a higher level of risk because there is potentially more to be lost than gained from an individual’s involvement (Meyerson et al., 1996). Therefore, as Luhmann (1979) would argue, one must choose to trust in order to overcome this level of risk and partake in a temporary system. As Meyerson et al. (1996, p. 180) synthesises, “The temporary system must comprise trust because it faces a future of potential disappointments and unstable collaboration among near strangers.”
50 2.5.4. The trust criteria of swift trust Meyerson et al. (1996) combine these conceptual insights to develop several trust criteria distinct to the theory of swift trust in temporary systems. These criteria have not yet been tested within the temporary system of an EOC in a disaster management context. The criteria can be summarised as follows:
1. Reputation – With the temporary system being drawn from a small labour pool and having overlapping networks, the chances increase that the members of this network will work together in the future. Consequently, a respondent will consider their reputation when deciding how to behave when in the EOC. Similarly, a respondent will also expect their colleagues to consider their own reputations when deciding how to behave. As such, members of the EOC do not expect their colleagues to undermine them; which allows swift trust to develop.
2. Role – Swift trust develops faster when respondents interact around roles, as opposed to personalities. However, if a colleague’s role is not clear, or they do not act in accordance with their role, a respondent may be hesitant to trust them; slowing the development of swift trust.
3. Categorical assumptions - As temporary systems are under time pressure, members rely more on categorical assumptions as opposed to evidence-based information when determining who to trust. This is because the time pressure imposed on the temporary system means that the respondents do not have adequate time to accurately determine whether their colleagues are trustworthy based on evidence. Examples of categorical assumptions include: their colleague’s role; culture; work industry norms; and identity stereotypes, such as appearance. Examples of evidence-based information include competency, personality, and work ethic. As such, swift trust is characterised by a reliance on categorical assumptions, which encourages speed over accuracy when developing trust.
4. Confirmation of assumptions – As respondents rely on categorical assumptions when determining who to trust, they will look for their colleagues to behave in accordance with the assumptions that they have made. Additionally, the respondent may behave in a certain way so as to affirm the assumptions they have made about a colleague. However, if a colleague does not act in accordance with the assumptions that the respondent has made, formation of swift trust may be impeded.
51 5. Moderate interdependence – Swift trust is more likely to form when interdependence between members of the temporary system is moderate; as opposed to high. High interdependence means that a respondent is more vulnerable to untrustworthy behaviour by a colleague. Therefore, there is too much at risk for the respondent, and they subsequently require more than categorical assumptions to develop swift trust.
6. The contractor: Meyerson et al. (1996, p. 183) argue that swift trust in temporary systems is also a function of the ‘contractor’ or leader who brings the members together: “The reputation of the contractor and the expectation of good will on his or her part may be all that is necessary to create the general background expectation of good will, independent of information about the other participants.”
2.5.5. Factors influencing the resilience of swift trust It is important to note, however, that swift trust is not guaranteed within temporary systems. Rather, like trust that forms over time, swift trust is subject to factors which make it resilient or fragile. Meyerson et al. (1996, p. 184) define swift trust to be resilient if it can, “survive those moments and incidents that occur during the life of a temporary group… [that] call into question or threaten to disrupt trust.” The factors that can influence the resilience of swift trust in temporary systems are described (Meyerson et al., 1996) below: 1. Reputation of the contractor: the reputation of the contractor for bringing together successful temporary systems can substitute for a lack of interpersonal history between members. Furthermore, it is assumed that the contractor has used certain evaluative criteria in order to select who to include in the temporary system, which helps members of the system develop swift trust. Therefore, if the contractor has a trustworthy reputation and makes clear to the temporary system what criteria were used for inclusion in that system, swift trust is more resilient.
2. Optimism: People within a temporary system expect, “better things to happen to them compared to the average group member.” (Meyerson et al., 1996, p. 186) They assume that, if something were to go wrong, it would happen to someone else in the group. Therefore, members are optimistic, which in turn makes swift trust more resilient.
3. A presumption of trust: Within a temporary system, even though trust is presumed, members can act as if trust were present. When a member observes a colleague within the temporary system acting as if swift trust were present, this affirms that member’s original assumption that swift trust is appropriate. This personal confirmation of swift
52 trust being appropriate, initiated by their colleague’s behaviour, increases the resilience of swift trust.
4. Assumed qualifications: It is assumed that each member’s inclusion in the temporary system is based upon their expertise and experience; which has been gained elsewhere and does not require verification. If each member of the temporary system believes the other members to be qualified in their roles, then swift trust may be more resilient.
5. Hedges: The resiliency of swift trust can be altered due to members of the temporary system creating hedges. A hedge is defined as a ‘way out’ in order to, “reduce the perceived risks and vulnerabilities of trust by reducing interdependence and thus its perceived costs” (Meyerson et al., 1996, p. 187). Simply, a hedge in a temporary system reflects an attitude where one member trusts another, to an extent. Hedges can also lead to distrust if the other member(s) of the temporary system find out that those hedges exist. They can also lead to overconfidence by the person who created the hedge, and to reduced group commitment.
6. Improvisation: Given temporary systems are formed to specifically address complex tasks in a set time, it is expected that things, “almost inevitably must lapse or break down on occasion.” (Meyerson et al., 1996, p. 189) Consequently, swift trust may be more resilient in temporary systems that improvise solutions to problems because the members feel that the system is able to surmount inevitable setbacks.
7. Focussing on the task: As temporary systems operate with time constraints, they simply do not have time to develop harmful interpersonal and group dynamics that can erode trust. Rather, members are focussed on the task and ensuring its completion, which can make swift trust more resilient.
2.5.6. Studies considering swift trust The current literature provides limited research regarding swift trust in temporary systems. Hyllengren et al. (2011) found that in the military, the reputation of a leader (deemed the contractor by Meyerson et al. (1996)) and their ability to improvise influences the development of swift trust in temporary military groups. Several researchers have also considered swift trust in virtual temporary systems. Jarvenpaa and Leidner (1998) state that for virtual teams, swift trust develops when members of the temporary system are focussed on the task at hand. Crisp and Jarvenpaa (2013) found that if the virtual temporary system complies with a set of agreed upon goals and actions, this can make swift trust more resilient.
53 Popa (2005) considered swift trust in a university context, where he investigated swift trust of a member of a temporary system toward the whole system, breaking the convention of considering swift trust just between individuals. He claims that, “people trust others with whom they work (temporarily) because they perceive their group members as trustworthy, they have a propensity to trust others in general, and they like their group members” (Popa, 2005, p. 79) and that a person judges trustworthiness based upon the other person’s ability, integrity and benevolence. This is interesting, given that Meyerson et al. (1996) argue that ability (in competent role behaviour) is a factor influencing the development of swift trust; although they do not specify integrity and benevolence. Furthermore, the impact that ‘liking group members’ has on swift trust is also not included in the theoretical model proposed by Meyerson et al. (1996). As such, emperical evidence is required to determine whether the conclusions by Popa (2005) are valid, both between individuals within a temporary system, and within natural disaster management and the EOC as a temporary system.
2.5.7. Swift trust and natural disaster management As Hyllengren et al. (2011, p. 357) comment, “empirical research on swift trust in temporary groups is still in its infancy and there is a great lack of empirical studies.” There is even less research investigating swift trust in temporary systems in relation to natural disasters, although some researchers have recognised the potential value of this conceptual approach.
For example, Papadopoulos et al. (2016) has stated the importance of swift trust in improving supply chain networks after a natural disaster, focusing on the Nepal earthquake in 2015. Drnevich, Ramanujam, Mehta, and Chaturvedi (2009) has considered how swift trust influences decision making in a simulated EOC. Currao (2009) claims that swift trust is important for collaboration in emergency management, and similarly W. R. O'Brien (2011) has stated that swift trust is significant in relief organisations’ ability to form temporary systems in order to respond to a natural disaster. However, none of these authors have investigated whether swift trust exists after a natural disaster within the temporary system of responders, nor provided a comprehensive study of the trusting criteria of swift trust in a disaster management context, in accordance with the theoretical model proposed by Meyerson et al. (1996).
A more detailed investigation into the trust criteria proposed by Meyerson et al. (1996) has been conducted by Zolin (2002), who investigated how swift trust enhances cooperation and coordination between the US military, multinational forces, NGOs and foreign government agencies when partaking in stability, security, transition and reconstruction (SSTR) operations like those during and after a disaster. The author found that swift trust was unlikely to occur in
54 SSTR operations because there were not clear roles for members, the labour pools that members were selected from were too large, and members did not feel like they belonged to the temporary network. She writes that, “international temporary teams are not always so successful in developing swift trust, due to…differences in culture, goals, experiences, stereotypes and ideologies” (Zolin, 2002, p. 5). Although Zolin’s (2002) contribution does not provide any detail contributing to the development of swift trust within an EOC as a temporary system, it does point to the fact that not all of the swift trust criteria may be applicable in assessing how trust forms for local and international disaster management personnel.
Similarly, Majchrzak, Jarvenpaa, and Hollingshead (2007) found that swift trust between emergent response groups (such as spontaneous search and rescue teams) developed through action and getting things done as opposed to the conditions Meyerson et al. (1996) propose; such as clear roles and firm deadlines. It is important to note that the conclusions provided by Majchrzak et al. (2007) have not been empirically tested, nor are the temporary systems of emergent groups the same as a structured EOC. Considering roles, Curnin, Owen, Paton, Trist, and Parsons (2015) argued that role clarity is important for coordination in disaster management. However, the authors only validate the importance of role clarity in temporary systems, without empirically considering the impact of role clarity on swift trust formation. Furthermore, they do not consider the other swift trust criteria.
Fahy (2012) has investigated the relationship between swift trust and interagency collaboration, focusing on the police (NYPD) and fire (FDNY) departments of New York City when responding to a crisis. They concluded that for first responders in New York City, swift trust was affected by initial interactions and communications, role, team identity, and organisation culture. Considering organisation culture, Fahy (2012, p. 21) writes, “individuals tend to distrust those outside the organisation and its culture because uncertainty exists as to how they will act in a given circumstance.” Although the author is only considering two organisations (FDNY and NYPD), it is worth considering whether this ‘disposition to not trust’ may also apply within an EOC; which features many organisations coming together. Lastly, in order to understand the factors that affect swift trust, Fahy (2012) produced a swift trust matrix which summarised all the factors that the literature proposes influence swift trust in military, business, and virtual contexts. However, the matrix provides no consideration of swift trust in natural disasters. Therefore, an original contribution that this research can provide is a consideration of the factors that affect swift trust formation in the temporary system of an EOC.
55 In contrast to Fahy (2012), instead of considering multiple factors that affect swift trust, several authors have considered only one factor of swift trust in a specific context. Wishart (2008) argues that face to face meetings in a disaster situation (such as that between representatives from an NGO and the military) can help swift trust develop. Zolin and Gibbons (2014) also state that face to face meetings are important for developing swift trust in temporary systems. Alvinius, Danielsson, and Larsson (2010) found that for ‘on-the-ground’ disaster response, swift trust was dependent upon the organisation that a member of the temporary system came from. Tatham and Kovács (2012) argue that, due to the high influx and turnover of responders in a natural disaster, having a clear set of rules assists the development of swift trust; as this enables expectations about the temporary system and subsequent roles of the members to be readily established. Finally, considering the resilience of swift trust, Stephenson Jr (2005, p. 344) states that for natural disaster management, “while participants may trust a participant at the start of their relationship in a relief situation, a single miscue can foul this opportunity for the remainder of the emergency and certainly for organisational relations in the next.”
2.5.8. The need to investigate swift trust in an EOC Despite the aforementioned research, there remains a lack of empirical evidence, and a gap in the analysis, regarding whether swift trust exists within the temporary system of an EOC during the response and recovery from a natural disaster. Furthermore, there is a need to explore whether all the trust criteria proposed by Meyerson et al. (1996) are applicable to this temporary system:
“[It is necessary to] investigate the phenomenon of swift trust in further detail, looking at the characteristics of people who are more likely to engage in this form of trust, [and] at the contexts in which swift trust is more likely to appear.” (Popa, 2005, p. 96)
To this end, building on the crucial literature review presented above, this study will investigate three research propositions to test the applicability of the theory of swift trust in temporary systems, and to help better understand the process of trust formation for disaster management personnel working together on the response and recovery from a natural disaster in an EOC.
1. Research Proposition 1: During disaster response and recovery in a developing nation, an EOC exhibits the properties of a temporary system.
2. Research Proposition 2: The trust that develops between disaster management personnel within an EOC is in accordance with the trust criteria proposed by swift trust theory.
56 3. Research Proposition 3: The resilience of trust in a disaster management EOC is determined by the resilience criteria proposed by swift trust theory.
57
METHODOLOGY & RESEARCH DESIGN
This chapter starts by describing the ontological and epistemological position adopted by this research. It then justifies the adoption of a single case study research design and the selection of the Emergency Operations Centre of Tropical Cyclone Winston in Fiji as the case study. Finally, it presents a three-phase research design. Phase one considers the EOC as a temporary system. Phase two explains the data collection through semi-structured interviews, document analysis and participant observation, whilst recognising the importance of reflexivity. Phase three outlines how content analysis and the subsequently derived coding framework were used to analyse the data.
The philosophical position The philosophical position of this research provides a framework for understanding the reality of trust within an EOC and how it can be made known. It subsequently directs the research design toward data collection methods that complement this philosophy. As Miller and Brewer (2003, p. 231) write, “it is possible to define the philosophy of social research as the study of the theories of knowledge which validate particular research practices.” In order to determine an appropriate research design, it is first necessary to define the ontological and epistemological position of this research project.
3.1.1. Ontology and epistemology The ontological position of this research must be defined in order to provide a rubric for understanding the conceptions of reality that impact trust and its existence within an EOC (Creswell, 1994). Within the realm of social science ontologies, five main philosophical positions could be used to perceive and model the reality of the phenomenon of trust. These are:
Positivism,
58 Realism,
Pragmatism,
Naturalism, and,
Social constructionism.
Each of these philosophical positions is coupled with a subsequent epistemology. The epistemology of this research complements its ontological position, and frames how knowledge about trust within an EOC is constructed. Each of these philosophical positions, and how they relate to trust, is outlined in Table 3.1.a.
59 Table 3.1.a The possible philosophical positions for this research
Philosophical Ontology Epistemology The philosophy’s approach to the phenomenon of trust position Positivists believe that reality is Knowledge is based upon “The primary goal of positivistic inquiry is an explanation “independent of people’s perception determinism, reductionism and that (ultimately) leads to prediction and control of of it: the social world is revealed to theory building: that is, cause and phenomena” (Ponterotto, 2005, p. 128). In reference to us, not constructed by us” (Miller & effect thinking through the the previous chapter describing the many antecedents of Brewer, 2003, p. 236). As such, reality interrelation of select variables, to trust, coupled with the fact that a single, accepted is singular (Creswell & Clark, 2007). provide a theory that understands definition of trust has not been developed, it has been the singular reality (Creswell & shown that trust is not a binary concept. A positivistic Clark, 2007). Therefore, what can philosophical approach to this research project would aim Positivism be known of the social world is to show whether someone trusts someone else without that which is revealed to us by our understanding why. This element of ‘why’ is key to senses through observation, in answering the research propositions; especially the focus order to produce deductive on the theory of swift trust. It requires interpretation, as explanations using objective (or opposed to reductionism and determinism, in order to ‘hard’) data (Miller & Brewer, appreciate the multifaceted nature of the phenomenon 2003). trust. Put simply, a positivistic reduction of trust devalues its complexity. “Realism involves the assertion of the Knowledge is gained by A realist could conceive of trust in two ways. The first existence of a social reality over and understanding the, “social would be to consider trust as a social structure that is above the existence of individuals” structures that determine or independent of individuals. This would mean that the (Miller & Brewer, 2003, p. 256). As influence the actions of research project attempts to measure the impact that trust such, realism considers reality as the individuals” (Miller & Brewer, (as a social structure) has on people, and vice versa. Realism relationship between social 2003, p. 256). Knowledge is also Secondly, a realist approach could investigate trust by structures and individuals; that is, the gained through understanding the considering the role that it plays as a medium between impacts these structures have on reciprocal influence that society and a different, specific social structure and its, individuals and vice versa. individuals have on these social “influence [on] the actions of individuals” (Miller & Brewer, structures. 2003, p. 256).
60 Philosophical Ontology Epistemology The philosophy’s approach to the phenomenon of trust position Reality is both singular and multiple; The question of the research Pragmatism would approach trust with less consideration that is, there could be an overarching becomes the focus as opposed to as to whether it is a subjective or objective phenomenon, theory to explain a phenomenon, or, the philosophical position. and seek to answer the specific research questions by individual interpretation is needed. Therefore, knowledge is gained whatever means possible. Using this philosophical Reality can be both objective and through all means necessary to framework for trust could result in an approach that does Pragmatism subjective. (Creswell & Clark, 2007) answer the research question. not reward uncovering the complex nature of trust Pragmatism, “is pluralistic and because it is more focussed upon a pluralistic approach orientated toward “what works” towards answering the research propositions. and practice” (Creswell & Clark, 2007, p. 41). According to naturalism, reality is Knowledge in naturalism is based Naturalism would argue that trust is a phenomenon that is partially constructed by individuals. upon, “understanding the freely created between people within their natural context. Trust This is due to their ability to describe constructed character of human would be influenced not only by relationships, but also the their experiences and communicate actions and institutions in the situation and social structures that surround it. As such, with each other. Reality, however, is natural settings and context that naturalism is more aligned with the current literature of Naturalism not a completely social invention, as influence and shape people’s trust, where trust is context and relationally dependent. it requires people to interact with meanings” (Miller & Brewer, and interpret social structures and 2003, p. 211). situations (Miller & Brewer, 2003) which they have not created or conceived.
61 Philosophical Ontology Epistemology The philosophy’s approach to the phenomenon of trust position Social constructionism argues that Given that reality is constructed Social constructionism considers the phenomenon of trust reality is subjective to the individual’s through society and culture and to be subjective. Trust is given meaning due to an interpretation and experience. the interactions individuals have individual’s personal history and their previous Therefore, meaning is prescribed to with each other (Kim, 2001), interactions with other people. Put simply, social something based upon a person’s knowledge concerning a constructionism is appropriate for the study of, “the social Social history and interactions with others phenomenon is obtained through practices engaged in by people, and their interactions with Constructionism (Creswell & Clark, 2007). As such, learning about these personal each other” (Burr, 1995, p. 7). As such, given that the meaning is historically and culturally interactions and experiences literature highlights the subjective nature of trust and that relative (Burr, 1995); therefore, there directly from the respondents trust is contextually dependent and relies on relationships, are multiple realities (Creswell & (Creswell & Clark, 2007). a philosophical approach of social constructionism would Clark, 2007). capture this reality; where meaning and knowledge are created through relationships.
62 In order to adopt a feasible philosophical approach for this research, one must consider which position most appropriately considers the phenomenon of trust; in particular, the formation of swift trust within a temporary system. Each philosophical approach will identify and measure trust in some manner, however some ontologies and epistemologies do this better than others for the phenomenon at hand.
The argument from positivism, in which a single reality can be objectively measured, is inappropriate for this research. As shown in chapter two, the diverse antecedents of trust, the fact that there is no agreed definition of trust, and the many different contextual forms of trust, shows that trust is diverse and not ‘objectively’ understood nor “independent of people’s perception” (Miller & Brewer, 2003, p. 236). Therefore a deterministic and reductionist approach (Creswell & Clark, 2007) would not sufficiently capture either the complexities of trust, or the factors that cause a person to trust.
Realism argues that, “the environment consists of entities or beings with objective properties, that is, with characteristics that are independent of social processes and human understandings” (Cudworth, 2003, p. 25). However, as argued with positivism, the complexity of trust, coupled with its contextually antecedent conditions, cannot be neatly captured within the realist paradigm. The literature (Cudworth, 2003) shows that there is no ‘one’ set of objective properties to characterise trust and that trust is not a singular social structure of reality that individuals interact with. Rather, it is formed through social interactions among individuals; not imposed upon them as a pre-existing social structure, as a realist model may suggest.
A pragmatic philosophical approach to trust would consider the research questions of highest priority, being less concerned with the ontological argument regarding the objective or subjective nature of trust. However, the literature claims that trust is relational, complex, contextually dependent, and influenced by many different antecedent conditions, and this research espouses that trust is subjective to the individual. Furthermore, this research aims to not just measure the existence of trust, but to understand why it occurs within an EOC and if the theory of swift trust is relevant to that context. Therefore, because this research aims to focus on understanding the subjective nature of trust, it is necessary to select an ontology that is solely dedicated to this philosophy rather than the pluralistic approach of pragmatism.
Naturalism adopts a position used frequently with ethnography and grounded theory, and aims to understand a phenomenon using an inductive ‘bottom up’ approach (Miller & Brewer, 2003). According to Miller and Brewer (2003, p. 212), for naturalists, “knowledge of the social world…is inadequate if we do not also document, observe, describe and analyse the meanings of the
63 people who live in it.” Naturalism is closely associated with gaining an understanding of the motivations behind, and actions of, people in their natural context. However, this research is not focussing on the meaning that people themselves prescribe to trust in order to develop a ‘grounded theory’ for trust within an EOC. Rather, this research is seeking to determine if the theory of swift trust is an appropriate theory on which to model trust between local and international disaster management personnel in the unique context of an EOC.
Conversely, the ontology and epistemology of social constructionism is based upon social interactions (Burr, 1995); and considers how these interactions shape trust. As the literature has shown that trust is dependent upon relationships, such as those between local and international disaster management personnel, it seems appropriate to choose a philosophical position that promotes social interaction: a criterion which is core to the existence of trust. Each person within the temporary system of the EOC will have unique experiences that affect how they trust other people they are working with during the response. As such, this research project aims to understand these, “subjective motivations that lead to the social actions of individuals” (Miller & Brewer, 2003, p. 256) in order to learn the applicability of the theory of swift trust to an EOC. Given these arguments, this research project has chosen to adopt the philosophical position of social constructionism.
This choice relies on an interpretivist epistemological approach which treats people as meaning makers (David & Sutton, 2011, p. 33), and investigates the issue of trust in its natural setting and from the perspectives of the individuals who operate in that setting (Malcolm Williams, 2000). That is, “it is only through the research participants’ understanding that we can understand their actions” (David & Sutton, 2011). Interpretivism emphasises that the researcher should concentrate on why those within the EOC form trust and understand their personal justification for such behaviour (Punch, 2013).
Single case study research design The methodology of this research, which adopts social constructionism and interpretivism, necessitates a case study research design and the use of qualitative methods (Creswell, 2013). The case study approach involves an in-depth investigation of a contemporary phenomenon within its real life context, which is bound by time and the activity that is occurring (Creswell, 2013; Yin, 2017). In the context of this research, the value of the case study approach is in allowing the researcher to investigate how and why the specific phenomenon of trust occurs within an EOC (Yin, 2017).
64 Flyvbjerg (2006, p. 223) provides further justification for the use of a case study in the context of this kind of research. He argues that case studies are valuable because they provide, “a nuanced view of reality…[with] concrete context-dependent knowledge.” That is, a case study satisfies the requirements of an interpretivist epistemological approach and can provide the insights into understanding the reality of trust within the specific context of an EOC. Furthermore, with particular consideration to the fast paced context of an EOC, Noor (2008, p. 1603) states that a case study, “enables the researcher to gain an holistic view of a certain phenomenon…capturing the emergent and immanent properties of life in organisations and the ebb and flow of organisational activity, especially where it is changing very fast.”
3.2.1. Case study type There are many different types of case studies that can be undertaken. Yin (2017) states that case studies can be:
Exploratory – a case study designed to identify hypotheses and propositions that can be further investigated.
Descriptive – a case study designed to describe an event or situation; appropriate for research questions that ask ‘what’, ‘where’, ‘how many’ and ‘who’.
Explanatory – a case study designed to understand the causal links behind a phenomenon; most appropriate for research questions that ask ‘why’ or ‘how’.
The most appropriate case study type to undertake is best determined by the research question and propositions. An exploratory case study is not appropriate in this thesis, as this research does not seek to identify new hypotheses or propositions; but rather test the theory of swift trust within an EOC context. Similarly, a descriptive case study is not appropriate for this research, because this would be more suited to understanding the quantitative aspects of an EOC and emergency response as opposed to the social construct of trust. Rather, this research project seeks to understand ‘why’ trust is awarded, and ‘how’ it forms between disaster management personnel in an EOC in a developing nation. This approach leads towards the adoption of an explanatory case study. Additionally, an explanatory case study is well suited to a social constructionism and interpretivist philosophical position as it promotes an investigation into the subjective nature of trust forming within an EOC, treating actors within the EOC as creators of meaning. In adopting an explanatory approach, the case study chosen will seek to understand how the phenomenon of trust is developed for local and international disaster
65 management personnel and whether swift trust in temporary systems is an appropriate model to predict how such trust forms.
3.2.2. Single case study design When conducting a case study, the researcher can choose whether to adopt a single, or multiple case study approach. This project adopts a single case study approach for the following reasons:
After a natural disaster, the number of disaster personnel involved in the response is extensive. To conduct a case study of swift trust amongst all these respondents and their respective temporary systems would be an immense task. As Stake (1995) argues, it is necessary to define the boundary of the case study. Therefore, this research proposes that narrowing in on the EOC as the unit of analysis (Yin, 2009) will allow the project to obtain a deeper understanding of the development of swift trust for local and international disaster personnel in this context; as opposed to taking the entire disaster response as the chosen case.
This research aims to understand the socially constructed phenomenon of trust for disaster management personnel within an EOC. To achieve this requires gaining deep, personal insights from these individuals, whilst also understanding the natural context in which they operate and the impact this context has upon them (Creswell, 2012). Conducting a single case study allows this level of depth to be achieved. Pursuing several cases may limit the richness of data gleaned due to field research time being spread across multiple EOCs. Simons (2009, p. 12) reiterates this argument when she writes that, “the primary purpose for undertaking a case study is to explore the particularity, the uniqueness, of the single case.” Similarly, both Baxter and Jack (2008) and Flyvbjerg (2011) argue that a case study allows depth and rich analysis when considering a phenomenon, such as trust. This research project needs to understand each disaster management person within the temporary system, the broader context of the EOC, and how these factors fit together to influence the formation of trust. As opposed to obtaining a broad statistical analysis across multiple cases, a single case study approach enables the researcher to obtain these detailed accounts whilst in the real-life environment of an EOC (Zainal, 2007). As such, this research espouses that to conduct multiple case studies would come at the cost of a more detailed understanding of the formation of trust within an EOC for local and international disaster management personnel.
66 Each natural disaster, and its response and recovery, are rare and unique, making it what Yin (2009, p. 48) calls a ‘revelatory’ case. This type of case study is defined as a case where, “an investigator has an opportunity to observe and analyse a phenomenon previously inaccessible to social science enquiry.” Given this, focusing on one case allows the researcher to develop a deeper understanding of the study and provide a more thorough analysis of the development of trust from this ‘one of a kind’ occasion.
It is worth noting here that one primary argument against conducting a single case study is the concept of generalisability; that is, ‘the more case studies conducted, the more generalisable the results can be, and therefore a more applicable research project’. However, this is not necessarily the case. Stake (1995, p. 4) argues that, “case study research is not sampling research” and that there should be no requirement that a case be studied in order to understand how it relates to other cases. This research project does not aim to understand the phenomenon of trust through large scale sampling and experimental repetition in order to draw a generalisable conclusion. Rather, it seeks to understand the formation of trust within an EOC in great depth. This depth enables the results of this research to be highly ‘valid’, in that they accurately represent the phenomenon under investigation, but not necessarily ‘reliable’, in terms of generalisability to other settings (Creswell, 2013).
“The real business of case study is particularisation, not generalisation. We take a particular case and come to know it well, not primarily as to how it is different from others but what it is, what it does. There is emphasis on uniqueness, and that implies knowledge of others that the case is different from, but the first emphasis is on understanding the case itself.” (Stake, 1995, p. 8)
Similarly, Yin (2009) agrees that there should be no expectation that generalisability is the end goal of case study research. He believes that case studies can be generalisable to the theoretical propositions being tested, however the specific conclusions drawn are not expected to be transferrable to a wider population. That is, obtaining a deep knowledge about the formation of swift trust within one EOC is greatly beneficial in helping to develop the theory; however it is not expected that the conclusions will be transferrable to every EOC responding to a natural disaster. It celebrates the particular. This argument is well summarised by Flyvbjerg (2006, p. 228), who writes, “formal generalisation is overvalued as a source of scientific development, whereas “the force of example” is underestimated.” Each natural disaster is unique, and each nation that is affected has its own culture and systems of responding. As such, the antecedents to trust can be different across each developing nation, due to factors such as culture and work
67 expectations. It is essential to not under-appreciate the influence of these factors on the formation of trust, and it is appropriate to deeply investigate what makes any single case study unique. Therefore, this project seeks to gain a thorough understanding of a particular formation of trust for local and international disaster management personnel within one EOC as a single case study.
Case study selection criteria It is necessary that the choice of case study for this research led to a better understanding of the formation of swift trust within an active EOC. Therefore, the natural disaster chosen was required to satisfy the following two criteria:
Criterion 1: An established EOC – The response to the natural disaster had to have been run out of an EOC. This was essential, so that a temporary system could be established in one centralised location. It would house the disaster management personnel; some who will not have worked with each other prior to the response.
Criterion 2: The EOC consisted of local and international disaster management personnel – Firstly, the natural disaster chosen must have occurred in a developing nation. That developing nation then needed to request international assistance. This was necessary so that there would be both local and international disaster management personnel working together on the response in the EOC.
On the 20th February 2016, a devastating category 5 cyclone, Tropical Cyclone (TC) Winston, struck the Republic of Fiji (Fiji), causing large scale damage to the nation. This natural disaster met all the selection criteria described above:
Fulfilled criterion 1: An established EOC – Fiji has a permanent EOC which acts as the centre for coordination for all pre and post natural disaster related matters. As such, this EOC was used for the response to TC Winston.
Fulfilled criterion 2: The EOC consisted of local and international disaster management personnel – After TC Winston struck, the government of Fiji declared a national state of emergency on the 20th February 2016, and made an appeal to the international community for assistance. This was in line with the National Disaster Management Plan of Fiji, where it states that:
“An appeal for international assistance, either to specific countries or a general appeal, is made by the Prime Minister on the basis of advice by the National
68 Disaster Controller; international assistance will be sought when the impacts of the disaster go beyond the capabilities of the local and national resources to cope.” (NDMP, 1995, p. 111)
This declaration allows TC Winston to fulfil the second case study criterion listed above. With the request for international help, the response to TC Winston now involved international disaster management personnel who worked alongside the local disaster management personnel. Without such a declaration and request, TC Winston would not have been a viable case study for researching the formation of trust for local and international disaster management personnel in an EOC.
Furthermore, it is necessary to add that many natural disasters may receive international aid, either in the form of relief funding or personnel. However, when selecting a case study, the more international personnel who respond to the natural disaster, the more opportunities there are for swift trust to develop between locals and internationals. In light of this, TC Winston was further positioned as an ideal choice for the following two reasons:
International personnel were sent from multiple national governments to aid the response and recovery, including Australia, New Zealand and France (National- Emergency-Operation-Center, 2016a).
Fiji is the geographical centre for many of the international NGOs who are working on development projects within the Pacific. As such, in addition to the international NGOs who flew in to specifically aid the response and recovery, many international NGOs who were already in-country sent additional staff into their Fiji offices to also support the response effort. Examples of these NGOs include: UNOCHA, UN Women, Red Cross, World Health Organisation (WHO), World Food Programme (WFP), and UNICEF.
Other potential natural disasters for the case study Apart from TC Winston, other natural disasters occurred around the same time, which could have been considered for this research. However, they were deemed less appropriate when considered against the selection criteria, or infeasible, when compared against TC Winston. These natural disasters were in:
Taiwan - A 6.4 magnitude earthquake struck Taiwan on 6th February 2016, killing 117 people (Aljazeera, 2016; Ssu-Jui & Huang, 2016). This natural disaster, however, was dealt with by the Taiwanese Government without a request for international assistance.
69 As such, international personnel were not included in the response; thus making this case study inappropriate as it did not fulfil the second criterion.
India - A 6.8 magnitude earthquake struck Imphal, India, and was also felt in Myanmar, Bangladesh and Nepal; killing at least 11 people and injuring nearly 200 (Fernandez, 2016). Like in Taiwan, this natural disaster was responded to by the national government, who sent two teams from the National Disaster Response Force to aid the rescue efforts (Fernandez, 2016). As the ongoing response and recovery efforts were managed by national governments and lacked international assistance, this disaster was less appropriate than TC Winston when examining criterion two. Furthermore, with the disaster striking multiple countries (primarily India and Bangladesh), conducting research across two emergency operation centres would have been logistically difficult for one researcher.
Tonga - Category 3 Tropical Cyclone Ula caused Tonga to declare a state of emergency, however this was only for preventative reasons. Although there was damage recorded, fortunately there were no casualties from the natural disaster (ABC-News, 2016). Oxfam New Zealand were sent to respond to the event (Oxfam-New-Zealand, 2016), however the degree of response and the size of the disaster when compared against TC Winston made the latter a more feasible and appropriate case study.
Therefore, although other natural disasters had occurred around the same time, TC Winston proved to meet all the necessary criteria to be selected as a case study for this research. Additionally, the scale and local geographical concentration of the response and recovery effort in Fiji made it a feasible and realistic case study option.
Case study description – Tropical Cyclone Winston in the Republic of Fiji The Republic of Fiji (Fiji) is an island nation consisting of 332 islands, approximately 110 of which are inhabited (The-World-Factbook, 2016). It has two main islands, Viti Levu (south) and Vanua Levu (north). The nation is divided into four administrative divisions; western, central, northern, and eastern. The Capital is Suva, located in the central division on the south of Viti Levu (refer to Figure 3.2.1). As of 2015, Fiji had a population of 892,145 (WorldBank, 2016).
Across the 20th and 21st of February 2016, category 5 Tropical Cyclone (TC) Winston moved through the islands of Fiji (UNOCHA, 2016). Winds reached speeds of 325km/h and TC Winston was recorded as one of the most severe cyclones ever within the South Pacific (Shelter-Cluster,
70 2016). It was estimated that 350,000 people (approximately 40% of the population) were affected, with 31,000 houses damaged or destroyed; primarily in the eastern, northern and western divisions (UNOCHA, 2016). The damages from TC Winston were estimated to be approximately US $1.4 Billion (WorldBank, 2016); which is equivalent to 32% of the nation’s GDP. As discussed in Chapter 1, losses that are proportionally high when compared to the national GDP can be devastating for a developing nation’s economy (Charvériat, 2000). The Fijian Government requested international assistance and declared a national state of emergency on the 21st February 2016 (Shelter-Cluster, 2016), which lasted for 30 days. This was then extended for another 30 days (until 19th April) for the areas most severely affected by the cyclone (Reliefweb, 2016), and again for an additional 14 days; making the state of national emergency 74 days in total. TC Winston killed 44 people (National-Emergency-Operation- Center, 2016c) and up to 56,000 evacuees required the use of emergency shelters (the majority of whom were in the western division) (National-Emergency-Operation-Center, 2016b).
71 VANUA LEVU
NORTHERN DIVISION
WESTERN DIVISION
TC WINSTON PATH
EASTERN DIVISION
CENTRAL DIVISION VITI LEVU
Suva
Figure 3.2.1 The path of TC Winston across Fiji
72 TC Winston Emergency Operations Centre (EOC) As Fiji is at risk of annual tropical cyclones, tsunamis, drought, and other natural disasters, the government has established the National Disaster Management Office as a permanent EOC to prepare and mitigate against these potential threats. This office is staffed by a small skeleton staff outside of a natural disaster. In response to TC Winston, this EOC increased in size to also include the National Emergency Operations Centre, Current Operations, and Future Operations teams. All of these teams were housed within the one main government building in Suva, which became the EOC for TC Winston. All teams reported to the Director, who coordinated the EOC. The purpose of each group was as follows:
National Emergency Operations Centre – this office was specifically staffed to deal with incoming data from the on-the-ground response. It was the centre of communications, receiving constant updates about what was happening in the divisions in Fiji. The divisional EOCs communicated to the National Emergency Operations Centre information such as supply needs, aid requirements, the number of people affected by TC Winston, and how many people were in emergency shelters. The National Emergency Operations Centre would then write a situation report (every eight hours) which would be handed over to Current Operations. During the first few days of the response to TC Winston, the National Emergency Operations Centre was manned at all times, with teams taking eight-hour shifts. Personnel in the National Emergency Operations Centre included: government staff; primary services, such as telecommunications, gas and water providers; and police, fire and healthcare services.
Current Operations – Current Operations was a small team of five personnel, including the Director. They were charged with planning the next 24 hours of the operation, based off the situation reports that came out of the National Emergency Operations Centre. This group coordinated the immediate response and was responsible for the strategic direction of the operation.
National Disaster Management Office – the role of this team was to act upon any directives provided from Current Operations. This would include: distributing resources and aid to the districts; organising transport; procurement; and disseminating information to the media, government and citizens of Fiji.
Future Operations – the role of Future Operations was to plan for the next 48-72 hours of the response. This team was primarily run by the Fijian military and became paired with the Australian and New Zealand military forces that were sent to help coordinate 73 the response. Future Operations planned the distribution of resources by air, land and sea. They were also responsible for providing aerial surveys, which were critical to damage assessments, and providing ongoing support in managing the response.
Table 3.2.a provides a description of each respondent interviewed for this research, as well as additional information including the specific group they worked within during the response.
74 Table 3.2.a Information concerning each respondent interviewed within the EOC
Permanent or Respondent Gender Nationality Role Team during the operation Original office surge staff Ministry of Rural & Maritime National Disaster EOC 1 Male Fijian Surge Media Officer Development & National Management Office Disaster Management
National Disaster Australian Civilian Corp – EOC 2 Male International Surge Logistics Advisor Management Office stationed in EOC
Ministry of Rural & Maritime Senior Media National Disaster EOC 3 Female Fijian Surge Development & National Officer Management Office Disaster Management
Administration National Disaster EOC 4 Male Fijian Permanent EOC Assistant Management Office
Assistant – Current National Disaster EOC 5 Female Fijian Surge Ministry of Women Operations Management Office
National Emergency EOC 6 Male Fijian Permanent Deputy Director Operations Centre/National EOC Disaster Management Office Team Leader (Current Current Operations/ National Secretary of the Pacific EOC 7 Female Fijian Surge Operations) & Emergency Operations Centre Community Assistant to the Director
75 Permanent or Respondent Gender Nationality Role Team during the operation Original office surge staff GIS/Information Future Operations/ National Secretary of the Pacific EOC 8 Female Fijian Surge Management Disaster Management Office Community Specialist
National Disaster EOC 9 Male Fijian Permanent Deputy Director EOC Management Office
Cluster Coordinator National Disaster Australian Civilian Corp – EOC 10 Female International Surge in the EOC Management Office stationed in EOC Team Leader (Current Current Operations/ National EOC 11 Female Fijian Surge Operations) & Public Service Commission Emergency Operations Centre Assistant to the Director Senior Disaster National Disaster EOC 12 Female Fijian Permanent Management EOC Management Office Officer Current Operations Liaison Officer UN OCHA – liaised with the EOC 13 Male International Surge National Disaster (OCHA) EOC Management Office
Coordinator - Secretary of the Pacific EOC 14 Female Fijian Surge Current Operations Current Operations Community
Head of Accounts National Disaster EOC 15 Male Fijian Permanent EOC and Procurement Management Office
76 Permanent or Respondent Gender Nationality Role Team during the operation Original office surge staff
Coordinator - Secretary of the Pacific EOC 16 Male Fijian Surge Current Operations Current Operations Community
EOC 17 Male Fijian Permanent EOC Director Current Operations EOC
Accounts and National Disaster EOC 18 Female Fijian Permanent EOC Procurement Management Office
77 TC Winston Disaster Management Response Structure The TC Winston EOC can be thought of as the conduit for information flowing from the response on the ground, all the way up to the highest levels of government (the cabinet and Prime Minister). This flow of information is outlined in Fiji’s National Disaster Management Plan (NDMP, 1995), which documents the roles of personnel from Cabinet through to the divisional managers in the field. It is summarised in Figure 3.2.2.
78 Tropical Cyclone Winston Emergency Operation Structure Cabinet The purpose of the National Disaster Management Council is to formulate policies for the emergency operation, as far as possible during the emergency stage. Sitting inside the National Disaster Management Council is the Emergency Committee, which is chaired by the National National Disaster Management Council Disaster Controller. The role of the Emergency Committee is to: provide central coordination for the emergency operation, review preparedness and make-ready the response, provide guidance for the operation, and direct agencies of what to do during the emergency. After a Emergency Committee state of emergency is declared, the National Disaster Controller: assumes control, is able to direct government resources to where they are most needed during the operation, and informs National Disaster Controller the Disaster Management Minister of the progress of the emergency operation. In simple terms, the National Disaster Controller is the conduit of information from the ground operations to the Disaster Management Minister, and has government resources at their disposal to enable the operation to occur. The EOC includes the National Disaster Management Office; which is the permanent office that Emergency Operations Centre (EOC) cares for all forms of disaster management for Fiji. When a natural disaster occurs, the EOC houses the National Emergency Operations Centre; which acts as the hub for all disaster related (National Emergency Operations Centre and information during the emergency operation, including operational and logistical support. They National Disaster Management Office) also prepare briefings for the National Disaster Controller, prepare damage and relief needs assessments, and provide media and public information and warnings. The EOC also includes Current Operations Future Operations Current Operations and Future Operations; which respectively plan and coordinate the next 24 and 72 hours of the operation. EOC daily routine during a declared state of emergency: Clusters Situation report—0800H Briefing to the EOC Director—0830H Divisions Briefing to the National Disaster Controller—0900H Briefing to Prime Minister/Government Ministers/Disaster Council—1000H Debriefing—1800H Districts, Communities, Villages NDMC 1995. Fiji - National Disaster Management Plan. In: RMDNDM (ed.). Fiji. Figure 3.2.2 The TC Winston Emergency Operation Structure 79 The challenges of conducting a case study after TC Winston Although it has been shown that TC Winston was a viable case study; there were many challenges in conducting research concerning trust after a natural disaster; some of which were exacerbated as the case occurred outside the researcher’s usual culture and context. These can be summarised as follows:
Access to the EOC – As an EOC is generally the headquarters for the response after a natural disaster, it is important to manage who receives permission to work there. Members generally include political authorities, technical emergency managers, and representatives from emergency management organisations (R. W. Perry, 2003). Therefore, without having a designated role within the EOC, it is challenging to gain permission to research inside such a centre. Therefore, after multiple emails and ‘cold calls’, I obtained permission to conduct research in the Fijian EOC by the Director of the EOC. They expressed an interest in the project and were intrigued to learn how the results could be used to help Fiji’s disaster management capabilities in the future. Upon arriving to the EOC, I had a meeting with the Director to explain the research project in more detail and they then provided me with a space in the EOC to conduct my research.
(Note: If this project was unable to obtain access to an EOC, or a natural disaster had not occurred, then the research would have been conducted in an EOC in a developing nation during a training exercise or similar simulation which involved both local and international disaster management personnel. This would have still allowed the topic of trust between the respondents to be investigated under circumstances similar to that of a real natural disaster response.)
Access to sensitive data – As I was not employed by the EOC, I did not have access to any internal emails or private correspondence between disaster management personnel. As will be elaborated on during the results, the EOC, and ultimately the Fijian government, placed a high value on the security of information concerning the response to TC Winston. As such, information was not to be released to any external source without internal permission first being given.
Access to disaster management personnel – Although I had obtained permission to research the EOC, the challenge still remained of actually gathering data from the disaster management personnel, who were concurrently dealing with the complex consequences from the natural disaster (Alexander, 2008). After a natural disaster, disaster personnel are generally time poor and convincing them to prioritise partaking
80 in research can be difficult. However, there were three things that made approaching respondents easier:
1. Having the approval of the Director of the EOC to conduct research enabled me to approach each member in the EOC and ask them to partake in the project, reassuring them that to do so was in line with the Director’s approval. Each member of the EOC partaking in the research was told that the research was investigating trust formation between natural disaster respondents. (Refer to Section 3.2.2.7 for a discussion on the sensitive nature of trust and how the results were obtained in light of the respondents knowing that trust was being researched.)
2. On my arrival to the EOC, as well as meeting the Director, I was introduced to the personnel in the EOC by one of their local colleagues. This helped to ‘break the ice’ for starting conversations over the coming weeks with members of the EOC.
3. Being stationed in the EOC every day for several weeks made my presence a familiarity to the disaster personnel. This made it easier to approach them to partake in the research, as they were aware of who I was and why I was there.
Timing – As this research is investigating swift trust in temporary systems after a natural disaster, it was important that the data was obtained during the immediate response and recovery efforts. Therefore, the project had to wait for a natural disaster to occur in a developing nation, which then proceeded to ask for international assistance. These set of circumstances, albeit not impossible, remained a challenge for conducting such a research project. It is worth noting here that, given the pressure on the Director in responding to the disaster, it took a few weeks for me to gain permission and land in Fiji to research the EOC. Although this may seem to be quite a long gap between the event and obtaining access, it also must be considered that to conduct any form of research (particularly interviews) during the first weeks after a natural disaster the size of TC Winston would be unfeasible and potentially insensitive (given disaster personnel are working consecutive long days and might themselves be affected by the disaster). Landing after the emergency period and arriving during the early stages of the response meant that I had access to the personnel I needed to interview at the stage when they were working shorter days; and therefore had more time and willingness to cooperate. Furthermore, it was obviously not possible to physically observe the behaviour of disaster management personnel during the first days of the response. Although this would have been beneficial, to obtain such access would have meant being on the
81 ground as the disaster occurred, in addition to holding a disaster management role that would have warranted the researcher working within the EOC during those first critical days.
Expectation to contribute to the response – Given that TC Winston had just occurred for Fiji, it might have been expected that I was to contribute to the recovery effort. Similarly, if I had been perceived as a latent observer within the EOC, this may have affected peoples’ perception of me, and in turn, impacted their interview answers (if they had contempt toward my lack of involvement). Therefore, it was important that my role within the EOC was clearly defined from the outset so that disaster personnel within the EOC were aware of my purpose in being there. Therefore, during my first meeting with the Director, it was agreed that I would not work on the response and recovery but rather just focus on the research project. This was beneficial as I was able to purely conduct research.
Conducting cross-cultural research outside of the researcher’s home nation – there are many challenges that arise from conducting research in another country. Firstly, being able to speak the language is very important if conducting interviews and making observations about how disaster management personnel interact and converse. Similarly, cultural factors can impact how disaster management personnel interact and converse, as well as how those personnel respond to the research. Therefore, the researcher must take this into account and gain a level of cross-cultural competency where possible. Another major consideration after a natural disaster is to be sensitive to whether respondents are grieving from the damage. Finally, logistically, obtaining accommodation, funding, and potentially a visa (depending on the nation), are challenges that can threaten the viability of the research project.
Operating in Fiji brought with it many of these challenges. Fortunately, the nation predominately spoke English in addition to the native Fijian dialects. As such, interviews and observations were all possible without the need for a translator. Considering the impact of grief, most of the disaster management personnel working in the EOC were not directly affected by TC Winston. This was because it struck the outer islands and not Suva (where the EOC was located) or the major southern cities directly.
In terms of logistics, the research project was entirely self-funded, as the EOC was not able to provide any financial support. Fortunately, Fiji does not require Australians to hold a visa to conduct research. However, obtaining accommodation in Fiji after the natural disaster was very
82 difficult. Given that the government had requested international assistance, all the major accommodation options in Suva were booked out because of the influx of respondents. Therefore, I indefinitely boarded with a local family until finding permanent accommodation close to the EOC three days later. Although logistically difficult, the initial days in Fiji provided valuable insights into Fijian culture, such as: how to greet people and other forms of etiquette; how some Fijians live and the socio-economic status of a typical Fijian family; the importance of family and the family unit; and an understanding of cultural ceremonies, such as funeral planning (which I was a part of), or sharing a common ceremonial drink called kava. These insights were valuable in shaping the researcher’s understanding of Fijian culture and enabling greater sensitivity to the influence of culture on the research project, and on the interactions of EOC members.
The sensitivity of asking about trust One particularly pertinent challenge with conducting this research was determining a way to measure whether someone trusted someone else within the temporary system. It could not be assumed that the respondents would tell me who they trusted and, particularly, who they did not trust. And even if they did, there was no means to test whether they were telling the truth. Continued observation of the respondent within the temporary system could potentially help validate (or challenge) any answers they might provide (Krefting, 1991); however, this was not a definitive method of assurance. Furthermore, conducting an in-depth pilot interview would not be able to affirm whether the questions asked about swift trust were valid; simply because people could lie when asked about such a sensitive topic (Corstange, 2009). Stone, Bachrach, Jobe, Kurtzman, and Cain (1999) argue that if a respondent deems a question to be potentially threatening to them, they might answer in a socially desirable way as opposed to telling the truth. Simply, an individual’s responses to the sensitive issue of trust cannot be controlled to ensure truth (Weiss, 1995). Therefore, the approach taken had to anticipate this by creating an interview environment that made the respondent comfortable, and by writing questions that encouraged the respondent to tell the truth.
As such, prior to the fieldwork in Fiji, this research considered some of the questions concerning trust that would be ideal to ask; if a truthful answer could be guaranteed. These included:
Do you think that you are trustworthy?
Is there a difference to how you trust a local colleague compared to an international colleague?
83 Has anyone done anything in the EOC that has affected your trust in them?
Given that such sensitive questions could not be asked as, ironically, a trustworthy answer could not be guaranteed, it was decided that questions concerning trust could be asked either:
In an indirect way so as to not alert the respondent that the topic of their trust in a colleague was being considered. An example from the interview included:
‘If you have a problem that you needed help with, who would you go to?’
This answer could then be compared against their recorded working relationships in order to see who they potentially trust and what role and relationship that other person has in the temporary system.
In a way so as to be concerning a ‘generic’ person and not a specific colleague. An example from the interview included:
‘Of the local/international colleagues you had not met or worked with before, how did you judge whether to trust them?’
Another approach that was adopted to help get truthful answers was to start the interview with non-threatening questions that were easy for the respondent to answer (Gubrium & Holstein, 2002). Hence, the start of the interview was designed to not only gain information about the temporary system and swift trust, but also help develop a rapport between the interviewer and respondent in a short period of time. It is important to note that the researcher, too, must have been regarded as an ‘international colleague’ and outsider to the respondents. Subsequently, the level of trust that they had with me was unknown. However, as Rasinski, Willis, Baldwin, Yeh, and Lee (1999) found, conducting a face to face interview, as opposed to an automated survey, can help provide more truthful answers to sensitive questions.
Finally, one additional technique employed to help obtain truthful answers concerning sensitive information was to provide anonymity to the respondent (Aquilino, 1994). However, for this research project, complete anonymity was not desirable. This research ultimately sought to understand the formation of trust for local and international colleagues in a disaster management context, and therefore, it was essential to know whether the respondent was a ‘local’ or ‘international’ colleague. Fortunately, each respondent agreed to being categorised as either a local or international colleague in the results. This partial degree of anonymity, albeit unusual, was necessary for this research and provided another mechanism by which truthful answers could be encouraged from the respondents. 84 Lastly, in addition to the aforementioned arguments, to help encourage truthful answers:
The interview was conducted in a space familiar to the respondents. Specifically, the interview was conducted in a private office away from other colleagues.
Only one interview was conducted with each respondent, which was beneficial when answering questions on sensitive topics, like trust, as the interviewer and respondent will not meet again in the future (Lee, 1993).
Clear instructions and the estimated interview duration were provided to the respondent before the interview. Each respondent was approached in person and their permission sought to partake in the research project. Before starting the interview, the respondent was informed that the interview would last approximately forty minutes to one hour.
The respondent was informed that they could ask questions at any time; such as to clarify a question.
Taking this approach maximised the potential for creating an environment in which the respondents were willing to talk about the sensitive topic of trust.
Three-phase research design Having selected TC Winston’s EOC as the single case study, this research project then implemented a three-phase research design in order to obtain data and analyse the results. Each of these phases are described briefly below and expanded upon in the remainder of the chapter (refer to Figure 3.3.1).
Phase 1 – Upon arriving in the EOC, there was a period of contextualisation. This was used to understand which disaster management personnel were working within the EOC and who could be included within the temporary system to be studied. Respondents who were part of the temporary system were chosen using specific criteria (refer to Section 3.4) and through ongoing participant observation and document analysis. Similarly, informal conversations with disaster management personnel helped in understanding the roles of each person within the EOC and whether they should also be interviewed.
Phase 2 – Once an initial temporary system boundary had been decided, interviews of the included personnel commenced. Through the interviews, if it was learnt that
85 additional disaster management personnel were part of the temporary system, they were then added to the interview schedule. At the completion of this stage, all interviews were conducted with the disaster management personnel. This phase continued to be supported by ongoing participant observation and document analysis.
Phase 3 – The final phase of the research design was an analysis of all the data collected. This was supported by all documentation and recorded observations as well as any analytical memos and personal reflexivity notes documented. The analysed data was then applied to the three research propositions, in order to better understand how trust forms for local and international disaster management personnel responding to a natural disaster.
86 Three-phase research design Refine the boundary of the Ongoing temporary system and which Compare personnel to interview analysed data Refinement Analysis of on trust against Active Participant Initial selection Semi-structured Semi-structured interview research Observation of temporary interviews of interviews of any data and all propositions to system personnel included new personnel other better boundary within the included within the supporting understand within the case Document analysis temporary system temporary system data how trust study forms in an EOC To complement
Field notes, transcribed conversations
Reflexivity
PHASE 1 – Contextualisation and the PHASE 2 – Data collection methods PHASE 3 – Data analysis temporary system boundary Personal contextualisation to the natural Refine the boundary of the temporary system as Investigation of all three research disaster zone data is collected and interpreted. propositions using the collected data and Establish the initial boundary of the temporary Conduct all semi-structured interviews of content analysis. system by identifying which respondents are selected personnel within the temporary system. active disaster management personnel. Figure 3.3.1 Research design - single case study approach
87 Phase 1 – Contextualisation and the temporary system boundary The purpose of Phase 1 of the research design was twofold:
Gain an understanding of the research project context. This meant continued learning about:
o the TC Winston EOC,
o work place culture,
o national culture,
o the physical environment,
o the disaster management personnel working on the response,
o the management and reporting structure,
o Fiji’s disaster management response procedures,
o the extent of damage of TC Winston to Fiji, and,
o the geography of Fiji.
Establish the boundaries of the temporary system within the EOC; that is, which disaster management personnel to include and exclude from the temporary system.
The process of determining which disaster management personnel to include within the temporary system involved three stages (refer to Figure 3.4.1):
Stage 1 – Determining a respondent’s original organisation.
Stage 2 – Determining when a respondent arrived in the EOC.
Stage 3 – Determining a respondent’s role within the EOC.
88 Stage 1 Stage 2 Stage 3 Permanent EOC Personnel
Arrival Role Inclusion in the into the within Temporary EOC the EOC System International and local surge staff
Figure 3.4.1 The process for selecting personnel from the EOC to be included within the temporary system
3.4.1. Stage 1 – Determining a respondent’s original organisation The purpose of this first stage was to identify where personnel within the EOC had come from. This was important, because it was necessary to know if a person was a local or international colleague in order to assess whether swift trust formed differently for these parties. As such, this meant determining if personnel were:
Permanent staff – who were part of the EOC prior to TC Winston; or
Surge staff – who were brought into the operation from another Fijian government department or international organisation. These personnel joined the response to bolster the capacity of the EOC to meet the requirements of the response.
To determine where personnel came from, this research started with an investigation of the organisation chart of the permanent staff of the EOC. However the chart was out of date and had not been updated to reflect recent staffing changes in the EOC (see Appendix A for the initial org chart). As such, informal conversations were conducted with disaster management personnel in the office in order to understand where each person had come from and why they were in the EOC. These conversations were validated through ongoing observations and analysis of other documentation (such as the staff movement board – Appendix A). From these data sources, a purposive sample of potential members of the temporary system was determined.
3.4.2. Stage 2 – Determining when a respondent arrived in the EOC This purposive sample of respondents was then screened through the second criterion: their arrival time into the EOC. Swift trust requires each member of the temporary system to be
89 working together since the chosen task’s inception and throughout its limited duration. As such, for a person to be included within the temporary system, it was decided that they had to be an active member of the EOC during the first declared national state of emergency. This timing was chosen so that those who were to be a part of the temporary system had all been involved in the same emergency response. This meant that those within the temporary system would have all had the same objective, approximately the same timings, and been under the same leadership; which is important in the formation of swift trust.
3.4.3. Stage 3 – Determining a respondent’s role within the EOC The final parameter that determined whether someone would be included within the temporary system was their role within the EOC. Many people within the purposive sample had roles that directly influenced the disaster response. These included logisticians, planners, military personnel, accountants, the Director, and section managers. As expected, the EOC also housed many supporting roles, such as executive administrators, kitchen assistants, and office clerks. It was decided that those with a role that had direct influence on the response were to be included within the temporary system (as opposed to those with supporting roles). This was because those with a direct role were more involved and unified by the task of responding to the damage of TC Winston. They also were the personnel more likely to have to establish new relationships based on the response, which is critical to determining the formation of swift trust within the EOC.
After applying these selection criteria, the temporary system consisted of a purposive sample of 21 personnel; the majority of which were local Fijians. [It is worth noting that until arriving in the EOC, it was not possible for the researcher to determine how many disaster management personnel would actually be working there, and what would be an acceptable amount of disaster management personnel to consider a part of the temporary system. As such, it was fortunate that there remained many of the original personnel who had been a part of the response during the declared (and still on-going) national state of emergency.]
Phase 2 – Data collection methods The purpose of phase 2 of the research design was to conduct interviews with the local and international disaster management personnel included within the EOC. A second critical component of phase 2 was being able to refine the boundary of the temporary system, to include any respondents not captured during the first pass at defining the boundary.
Therefore, this section begins by describing the use of semi-structured interviews as the most appropriate interview type for the research. It then explains how a brief pilot interview was 90 undertaken; which led to the introduction of the Network Mapping Worksheet, and changes in some interview questions. It then describes the refined data collection strategy: use of a single semi-structured interview, accompanied by addition data collection methods including document analysis, participant observation, field notes, and ad-hoc conversations. Finally, it considers the sensitivities in conducting research concerning trust, as well as the importance of reflexivity.
3.5.1. The use of semi-structured interviews To understand how members of the temporary system form trust requires an awareness of the experience of working in the EOC during the response to TC Winston. The benefit of using an interview is that it provides the researcher such an insight into, “the lived experience of other people and the meaning they make of that experience” (Seidman, 2013, p. 9). However, many forms of interviewing are available to the researcher, dependent upon the intended research outcome. The three primary approaches are structured, unstructured, or semi-structured (Dunn, 2005). They are described as follows:
Structured – an interview with a predetermined, standardised list of questions which are asked in the same manner and same order so as to avoid interviewer bias (Dunn, 2005). This type of interview is used for hypothesis testing, generally using closed questions (yes/no answers) such that the data obtained can be meaningfully quantified (King, Cassell, & Symon, 1994).
Unstructured - sometimes referred to as in-depth interviews, an unstructured interview is comparable to an open conversation with few questions. Most questions are open ended (Berry, 1999) such that the interviewee recounts or explains an event or phenomenon from their perspective (Dunn, 2005). This type of interview has its foundations in ethnography and anthropology (Whiting, 2008) and is particularly useful for oral histories or obtaining life stories from individuals.
Semi-structured – A semi-structured interview naturally falls in-between the prior two interview types. Semi-structured interviews utilise a predetermined order of questions with a degree of flexibility built in, such that any issue that needs clarification or further depth can be addressed (Clifford, Cope, Gillespie, & French, 2016; Whiting, 2008). The purpose of this interview type is to focus on the meaning a respondent ascribes to a phenomenon or event (King et al., 1994).
This research chose to conduct semi-structured interviews for the following reasons:
91 Understanding the formation of swift trust within a temporary system requires a strong grasp of the personal experiences of disaster management personnel within this context, and what impacts this environment may have on the formation of trust. Obtaining this depth, which relates to subjective perceptions and experiences, requires discourse between the interviewee and interviewer, as opposed to only using a strict set of questions; (as in a structured interview, which would not allow an interviewee’s experience to be fully explored or grasped by the researcher).
On the other hand, this research is primarily testing the applicability of the theory of swift trust as a model of how trust can form after a natural disaster. As described in Chapter 2, the theory has several propositions that are required to be true in order for swift trust to exist and be a valid model for use in a disaster management context. As such, this research requires these propositions to be tested through asking set questions to each respondent in an interview.
Therefore, the interview type required for this research must have a set structure which is able to test each of the propositions of swift trust, as well as sufficient freedom for the researcher to delve deeper into the answers provided by the interviewees in order to further understand their responses. Thus, a semi-structured approach was deemed to be the most appropriate to achieve this outcome given its, “attention to lived experience while also addressing theoretically driven variables of interest” (Galletta, 2013, p. 24).
3.5.2. Pilot study Before arriving in the EOC, it was planned that each member who was included within the temporary system would be asked to undertake two interviews. These were:
Interview one – the purpose of this initial interview was for the interviewer to ask the respondent about which colleagues they worked with in the operation, and what their colleagues’ roles were. This was done to help shape the temporary system boundary. From this initial interview, it was planned that any additional members of the temporary system that had not been included would then be incorporated into the research; assuming that they satisfied the aforementioned selection criteria. Respondents may not have been included in the initial purposive sample because they were:
o not included on the (outdated) EOC organisation chart or staff movement board,
o not mentioned by the colleagues who provided informal conversations, or,
92 o away from the EOC during the time the temporary system boundary was being established.
Therefore, the first interview would enable the inclusion of all relevant disaster management personnel within the purposive sample; including those who may have worked in a location outside the EOC but were still a necessary part of the temporary system.
The preliminary interview consisted of the following five questions:
1. Which of your colleagues have you most commonly interacted with in the past 30 days?
2. When with them, what did you interact about?
3. How many of your interactions were related to your work? How did you interact with them, phone, email, person to person?
4. Before coming into the EOC, which of these people had you not worked with before or known well?
5. Would you consider any of your colleagues a personal or close friend?
The purpose of these questions was to gain a preliminary understanding of the type of trust that each person had with their colleagues. Swift trust theory predicts that people trust others based on their colleague’s role within the team, as opposed to personally knowing such a person. It suggests that a person will rely more on categorical assumptions (as opposed to evidence-based information from the relationship) in determining whether to trust them or not. Hence, these questions were trying to develop an initial map of the relationships between each respondent and their colleagues, to see what these interactions were based upon. That is, were conversations entirely work related or did they include ‘personal life’ as well?
Interview two – the purpose of the second interview was to ask questions concerning the research propositions and the applicability of swift trust. This was done in light of the answers from the first interview so as to be able to probe deeper into specific relationships and gain a thorough understanding of the formation of trust within the EOC.
93 After arriving at the EOC, it became clear that this ‘two interview’ approach was not feasible given the limited time of the respondents. Furthermore, some disaster personnel, although agreeing, seemed hesitant to partake in the research. Therefore, it was decided (whilst in Fiji) that to conduct two interviews was not viable and that only one interview would be undertaken. However, it was not practical for all the questions from interview one to be transferred across to interview two. Question one and question four from the first interview required the respondent to list the names of those they interacted with in the EOC, and whether they had met for the first time during the response. Instead of asking and annotating notes on these important questions during the interview, it was decided that the respondent would be given a Network Mapping Worksheet to fill out prior to the interview.
The Network Mapping Worksheet This worksheet was used to understand each respondent’s connections within the temporary system. The Network Mapping Worksheets consisted of a table which the respondents filled out prior to being interviewed with the names of the colleagues they interacted with during the response (refer to Figure 3.5.1). However, after being in the EOC for some time, it became apparent that many disaster personnel were from external governmental departments or other agencies and had been asked into the EOC as surge staff. Therefore, additional columns were added to identify the respondent’s original department/organisation, their role in the operation, their colleagues’ roles in the EOC, and who had asked them into the response. These additional fields of information were necessary because:
Knowing the respondent’s original department helped in determining who they knew or did not know in the temporary system, and therefore which relationships could develop swift trust.
Swift trust is heavily dependent upon the roles of members in the temporary system and therefore it was valuable to know the function of each respondent in the EOC.
Learning who asked the respondent into the EOC allowed the researcher to identify, according to the theory of swift trust, the contractor.
94 Your name:
Your role in the EOC:
Your original office:
Who do you report to in the EOC?
Is this the same person who asked you into the EOC. If not, who is that person?
Please fill out that table below with the colleagues you interacted with most during your first month when you came into the EOC
Position/Role of your colleague in the Average number of times per week you Had you met this colleague before Name of colleague EOC communicate (about anything) working in the EOC?
1 <2 2-5 >5 Yes No
2 <2 2-5 >5 Yes No
3 <2 2-5 >5 Yes No
4 <2 2-5 >5 Yes No
5 <2 2-5 >5 Yes No
6 <2 2-5 >5 Yes No
7 <2 2-5 >5 Yes No
8 <2 2-5 >5 Yes No
9 <2 2-5 >5 Yes No
10 <2 2-5 >5 Yes No
Figure 3.5.1 The Network Mapping Worksheet
95 Another benefit of the worksheet was that it enabled the boundary of the temporary system to be refined. If a name which had not been included within the initial temporary system appeared on a respondent’s Network Mapping Worksheet, then that new individual would undergo the same selection criteria as all other members of the temporary system, to determine if their inclusion was necessary. This approach supported the initial observations and conversations used to establish the temporary system boundary and helped ensure that important members of the system were not excluded.
Secondly, by having a hard copy worksheet showing who a respondent interacted with, it was possible to look at this data prior to (and during) the interview. This meant that any relationships on the respondent’s worksheet could be directly referred to during the interview; therefore allowing the interview to discuss trust not in a general sense, but (if appropriate) by referring to specific relationships between the respondent and their colleagues within the temporary system.
In summary, the purpose of this worksheet was to:
understand which colleagues the respondent interacted with in the temporary system, the frequency at which they met, and whether they had met prior to, or during, the response to TC Winston;
understand the roles of both the respondent and their colleagues within the temporary system;
learn how the respondent became a part of the temporary system;
learn, according the theory of swift trust, whether a contractor existed in the temporary system;
help refine the temporary system boundary; and,
enable the interview to ask about trust in specific relationships, if deemed appropriate.
Pilot test Before conducting any interview, it is ideal that the questions to be asked undergo a pilot test within a context similar to that of the real interview. The purpose of pilot studies is to, “assist the research in determining if there are flaws, limitations, or other weaknesses within the interview design and [allow]…necessary revisions prior to the implementation of the study” (Turner III, 2010, p. 757). Furthermore, a pilot test should be undertaken with research
96 participants who represent the intended, final sample of respondents (Turner III, 2010). Pilot tests are beneficial as they enable the researcher to consider early on: whether the sampling technique appears to be accurate, any logistical problems involved in the research, any preliminary data, the feasibility of the research design, initial feedback on the interview, and whether the interview requires refining (Van Teijlingen & Hundley, 2001).
However, as Van Teijlingen and Hundley (2001, p. 2) argues, “completing a pilot study successfully is not a guarantee of…success”. Furthermore, Hay (2000) argues that a pilot test is less critical for semi-structured interviews as the interviewer has the ability to clarify any questions as the interview progresses. Regardless, to conduct a pilot test during a disaster response is extremely difficult, and opportunities for such a test are sporadic. Furthermore, given that swift trust is contextually dependent and time sensitive, and that a case study of a natural disaster is rare, it became necessary to use a one interview approach; and make any adjustments to the questions as necessary: a pilot test was, in the context of this research, simply unfeasible.
Altering interview questions once in the field is not uncommon (Hay, 2000; Taylor, Bogdan, & DeVault, 2015; Tremblay, 1957). In this research, there were several reasons why the interview questions were adjusted in situ:
Due to some difficulties in spoken English for the respondents, some questions were re- worded using simpler phrases. One technique used was to replace a more complex word(s) with more common words. These subtle changes made the questions easier to understand. If there was still difficultly in understanding the question, the question would be worded in yet another way and the interviewer would collaborate with the interviewee to ensure they understood what was being asked. This approach meant that no questions were left unanswered due to language barriers.
After being in the EOC for some time, it became apparent that there could be factors influencing how trust formed for the respondents in the EOC related to the involvement of the Fijian military. Therefore, a question was added to understand the effect (if any) that the military were having on trust formation in the temporary system.
Similarly, it became apparent that Fijian culture was influencing relationships between local personnel in the EOC. Having conducted informal conversations with local disaster management personnel about their culture, it became clear that a question had to be
97 included around the influence of culture on trust formation. (Refer to Appendix B for an elaboration on Fijian culture).
3.5.3. Refined final data collection strategy The following sections provide more detail about the specific data collected and the approach that was taken once on the ground in Fiji.
Semi-structured interviews The interview (refer to Appendix C) was conducted to utilise the benefits of a semi-structured approach. Although there were set questions, if an answer or comment provided by the respondent was of particular relevance to the variable being asked, or raised an interesting idea that warranted further investigation, a follow up question was asked to obtain this data.
The final interview consisted of two sections:
Section one – introductory questions helping the researcher gain an understanding of the EOC, followed by questions concerning each respondent’s Network Mapping Worksheet. This part of the interview helped map the temporary system boundary. It also provided clarity around; how the temporary system operated and who corresponded with who, and helped gauge the number of new relationships within the temporary system.
Section two – questions testing the applicability of swift trust to a disaster management context.
All interviews were audio recorded for transcribing purposes. This was disclosed to the respondent before the interview commenced to ensure they consented to being recorded (Gubrium & Holstein, 2002). Furthermore, prior to conducting any interviews, the research project went through the appropriate ethics clearance process set by the University of New South Wales. As part of the ethics process, prior to the interview, each respondent was provided with a:
Project Information Statement – which described the research project, as well as information concerning the degree of confidentiality and disclosure of information.
Project Consent Form – which each respondent signed to consent their inclusion within the research and their desired level of anonymity.
98 With the EOC, it was possible to obtain 18 interviews from a potential 21 in total. Two of the missing three were high ranking Fijian military officers who operated occasionally out of Future Operations and thus were generally not present during the time that the research was conducted. The last was a high ranking Australian Airforce Officer who was willing to provide some comments off the record. These comments were specifically used in understanding the relationship between the Australian and Fijian military.
3.5.4. Additional data collection methods Combining multiple data collection techniques in case study research can strengthen and validate results (Noor, 2008). This process, called triangulation (Guion, Diehl, & McDonald, 2001; Yoder-Bontrager, 2014), is beneficial as each data collection technique’s weaknesses are offset by the other techniques’ strengths (Gubrium & Holstein, 2002). The sensitivity of respondents in talking about trust meant that using additional data sources through triangulation was imperative; as it allowed for increased validity (Golafshani, 2003) of the interpretations made concerning swift trust in this disaster management context. A pertinent example includes Flick’s (1992) study of trust in counselling relationships that utilised triangulation to increase credibility in the results analysed.
In addition to the semi-structured interview, the data collection framework also included the additional data collection methods of document analysis and participant observation. This was done so that each document and observation could be directly linked back to a specific research proposition. The following sections outlines the documentation used from the EOC and how participant observation was recorded. It also explains the use of ad-hoc conversations and field notes as additional data sources of evidence, used to validate the results.
Document analysis Within the EOC, there were several documents that helped in understanding; how trust might form, as well as who to include within the temporary system. These were as follows:
EOC organisation chart and staff movement board – these two documents identified who was included within the EOC and which disaster management personnel were outside of this network. The EOC organisation chart was a record of all the people who were permanent staff within the EOC outside of an emergency period. The EOC staff movement board was a record of the main disaster management personnel who worked in EOC during the response. A comparison of these two documents helped to establish the preliminary temporary system boundary for the EOC.
99 EOC and Government ‘vision and values’ signage – upon the walls of the EOC were signs outlining the mission, vision and values that each member of the EOC was expected to uphold. These signs were used in the research in the following ways:
o To better understand the mission and vision of the EOC as goals of the temporary system.
o Considering the values signage, one value outlined on both the EOC and government signs around the office was trust. The definition provided was compared against that of swift trust and the wider literature. A question was also added to the interview to capture the influence this signage had on the formation of swift trust within the EOC.
Signs concerning stealing – upon arrival at the EOC, there were signs stating that property had been stolen from inside the EOC. I asked some respondents about what had been stolen, and used this information to assess the impact of stealing on the formation of trust within the EOC. This was particularly in reference to trust criteria one and four of swift trust theory; which state that swift trust can be dependent upon a person’s reputation, and that they behave in accordance with how other people in the temporary system think they should behave. However, because the culprit was not known, the impact of stealing was unknown; yet worth noting.
Fiji National Disaster Management Plan – this plan was essential in understanding how the EOC fit into the national response and what the purpose and function of the EOC was in terms of collating, reporting, and disseminating information to the government and public: these are critical components of determining how the EOC was a temporary system.
Participant observation DeWalt and DeWalt (2011, p. 260) define participant observation as, “a way to collect data in a relatively unstructured manner in naturalistic settings by [researchers]…who observe and/or take part in the common and uncommon activities of the people being studied.” Within this definition, it is important to recognise that there are different levels of involvement that a researcher can take when conducting their research; ranging from passive observation to complete involvement. Given the nature of the EOC and the requirements of the response, this research chose to conduct ‘moderate participant observation’. Moderate participant observation is where the researcher is present at the site of the case study, is identifiable as a
100 researcher, but does not actively participate in the tasks at hand or only occasionally interacts with the respondents (DeWalt & DeWalt, 2011). (For simplicity, moderate participant observation will be referred to as participant observation from now on). According to Spradley (2016), good participant observation means seeing the things that people naturally ignore (the mundane), and having a wide-lens perspective when looking at the environment. As such, being stationed within the EOC enabled the research to gather data through observation, which would complement the interviews and documentation.
Participant observation was used to understand how the respondents’ behaviour within the temporary system, as well as the physical and emotional context, could influence the formation of swift trust. Furthermore, any observations could be used to validate answers respondents provided during the interviews. Given the sensitivity of asking about trust, the use of participant observation was beneficial in helping to confirm the factors concerning trust that people might have been unwilling to talk about during the interview (Becker & Geer, 1957). Additionally, the use of participant observation whilst in the EOC was also critical in helping the respondents become familiar with my presence in the operation; and ultimately increasing the perception that I was trustworthy as a researcher.
Participant observation occurred throughout the entire time the research was being conducted in Fiji. Observations were conducted to test each of the three research propositions. Specifically, whilst in the EOC, this research observed (as examples):
which respondents had established relationships and who had not worked together before;
which respondents were locals and which were internationals, and which team they worked in whilst in the EOC;
the level of interdependence required for colleagues to complete their tasks as part of the operation;
whether there was any segregation within the EOC or if respondents were avoiding each other;
who the contractor of the EOC was;
the behaviour of the respondents and how they interacted with their colleagues whilst in the operation;
101 the nature of the work within the EOC, and whether tasks were discreet, or sporadic and based on changing priorities;
if the respondents interacted only about work or whether there were social interactions too;
how many colleagues a respondent interacted with as part of their role in the operation;
what each respondent’s role entailed and who they worked with; and,
any evidence of safeguards and hedges existing.
All relevant observations were recorded as field notes either directly on the data collection framework against the relevant research proposition, or in a field log for later analysis. Thus, the use of participant observation, in conjunction with document analysis and interviews, provided a triangulated approach with which to address the research propositions.
Ad-hoc conversations It is also worth noting that whilst in the EOC, additional ad-hoc conversations were held with disaster management personnel. These conversations were unstructured and proved to be additional sources of information, which allowed the researcher to further understand specific concepts which could influence the formation of swift trust. The significant conversations are listed in Table 3.5.a:
Table 3.5.a Ad-hoc conversations with specific respondents within the EOC in addition to the interviews
Conversation with Purpose of conversation respondent
EOC 11 Many of the impacts of Fijian culture go unseen. I specifically sat with these local colleagues to learn about Fijian culture. This information was critical to the research because if local colleagues had an EOC 9 additional tool for building trust based on culture, which international disaster personnel did not, then this could directly affect how readily EOC 12 swift trust forms and between whom. To ascertain the boundary of the temporary systems and which EOC 4 respondents to include in the EOC. To understand the structure of the EOC, how the EOC operated, and EOC 2 who to include within the temporary system boundary.
Field notes Throughout the operation, field notes were used to help shape, interpret and understand the research process. The purpose of the field notes was to:
102 Provide a record of observations and their potential influence on the research propositions.
Record the justification behind altering an interview question, or including a new question due to additional information becoming available (such as the involvement of the military or the influence of Fijian culture on trust formation).
Provide a medium where ideas could be elaborated on that might impact the research propositions.
Record any challenges in conducting the field work, such as obtaining access to respondents.
Provide a means by which the research process could be remembered after the field work had been completed.
Provide a space where personal reflections and notes on reflexivity could be recorded.
3.5.5. Reflexivity A final methodological concept that required consideration was reflexivity. Reflexivity asks the researcher to consider their role in the data collection process and how their intervention generated the data (Gibbs, 2008, p. 47). This involved regularly reflecting on the research process because data interpretation is subjective to a researcher’s personal experiences (Creswell, 2012). That is, in order to understand trust formation within the EOC, it was necessary to evaluate and mitigate any personal or external sources of influence that might skew the results and provide a misrepresentation of the phenomenon of trust within a disaster management context.
Consequently, below are specific reflexive factors which may have impacted the data. First, there is discussion of the factors concerning the respondents’ perception of me as a researcher. Secondly, there is consideration of how my own perspective might influence the results. For both of these points of view, the subsequent steps taken to minimise their impact on the results and analysis are explained.
The influence of the respondents’ perspective of the researcher:
o Not working on the response whilst in the EOC: As explained in Section 3.2.2.6 the respondents within the EOC may have expected that I would contribute to the recovery effort given the magnitude of TC Winston. If this was the case, the
103 respondents may have had contempt toward my lack of involvement; and this could subsequently impact their interviews. To alleviate this, I ensured that it was made known that I was in the EOC for research purposes only and that this was discussed and agreed with the Director.
o Being an international person: Simply not being a local may have impacted the respondents’ perceptions of me and my research. It is difficult to know to what degree this factor played a part; however, it is considerable that, in particular, local respondents may have withheld discussing local issues or their colleagues because they did not trust me or believe that I would use the data only for research purposes. I tried to limit the influence of this factor by offering the respondents anonymity and by using the Network Mapping Worksheets; however it is a bias that is hard to overcome completely. Similarly, when asking the local respondents about trusting their international colleagues, they may not tell the truth because they perceive me to also be an international; and so do not want to be seen to be prejudiced or inconsiderate.
o Being new to the EOC: Throughout the field work, I was unable to go into Future Operations because I was not known or trusted by the military personnel within that office. Simply being an unknown, international presence meant that there was data that I could not access because of who I am. This also emphasises the reality that local researchers may have better opportunities at obtaining data from their local colleagues than an international researcher.
The influence of the researcher’s perspective on the respondents’ answers and, subsequently the results:
o A lack of understanding of Fijian culture and the military: Not being Fijian meant that I could never fully comprehend the influence of Fijian culture on trust formation whilst in the EOC, even though it was found to influence how the locals trusted each other. Similarly, I could not understand by experience how trust may be affected by the Fijian military and living in a country with recent coups. To address this, I conducted multiple ad-hoc conversations and did wider reading in order to better understand Fijian culture and the role of the military. This also helped ensure these topics were captured appropriately in the interview. Furthermore, it is worth noting that in the analysis, as an international, I did not have any directly relevant personal convictions or
104 opinions concerning the influence of culture or the military on trust formation; and so the analysis is unbiased in this regard. Therefore, there are also benefits of being an international when conducting such qualitative research.
o Weighting international respondents’ opinions as more valid and reliable: Throughout the field work, I recognised that it was easier to conduct the international respondents’ interviews. This was because we were from the same/similar cultures, spoke English proficiently, and so could discuss pertinent topics in a more nuanced way. It was important to recognise that depth of conversation does not correlate to more valid and reliable responses. To counter this natural bias, the research used the transcribed interviews as the basis for analysis; which removed any assigned privilege due to the complexity or nuance in the delivery of an answer. Furthermore, testing the applicability of a specific theory also helped control the weight assigned to the internationals’ responses. This was because a respondent’s answers were being tested against a specific set of criteria developed from the theory; and whether they supported or countered it. That is, although language can help develop a sophisticated answer and provide extra depth (which can be beneficial), is not essential when determining whether a criterion of swift trust was true or not; nor does it devalue a simple response.
o Language: Occasionally throughout the interviews, some of the local respondents struggled to express their responses in English. However, I came to recognise that slow or simpler responses, due to a language barrier, are not always signs of uncertainty (which may unjustly devalue these responses). As above, this factor was alleviated by using the transcribed interviews for the analysis; which removed the audible influence of a respondent’s difficulties with English. It is important to note, however, that other factors of language, such as corrections, pauses, emphasis, were still included in the analysis.
Phase 3 – Data analysis Phase 3 of the research project involved the analysis of the interviews, documents, observations and field notes. This section first justifies the use of content analysis as a method of qualitative data analysis. It then explains the importance of the theoretical and broader situational context when analysing the results (called the analytical construct) and provides these specific
105 contextual factors for TC Winston and the EOC. Finally, it describes the process of content analysis, focussing on unitising the data and the coding framework adopted.
3.6.1. The use of content analysis as a method of qualitative analysis Considering qualitative data analysis, two main approaches were considered most applicable: narrative analysis and content analysis. Narrative analysis, as described by Riessman (2005), is commonly used in case studies and is defined as, “a family of methods for interpreting texts that have in common a storied form” (Riessman, 2005, p. 11). It is an approach that seeks to understand how and why things were said (Riessman, 2005); which is pertinent when considering the socially constructed phenomenon of trust. Narrative analysis can provide an insight into the mind and culture of those being researched (Cortazzi, 2014). However, the primary purpose of using narrative analysis is to provide continuity and flow of an account and use large sections of a narrative (or the whole account) to deduce meaning. A researcher would use narrative analysis if they were trying to unify multiple accounts into an overarching story. This lends itself to open ended interviews, without structure, in order to allow the respondent to tell their account of the events as they see fit. Therefore, given that this research is specifically testing trust formation for each respondent rather than reproducing a single phenomenon, it was decided that narrative analysis was not appropriate.
Conversely, content analysis is defined as, “a research method for the subjective interpretation of the content of text data through the systematic classification process of coding and identifying themes or patterns” (Hsieh & Shannon, 2005, p. 1278). Unlike narrative analysis, content analysis does not necessarily require the formation of a story from open-ended interviews, but rather is well suited to theory testing. This approach, called directed qualitative content analysis, is used, “to validate or extend conceptually a theoretical framework or theory” (Hsieh & Shannon, 2005, p. 1281). Hsieh and Shannon (2005) state that there are two strategies that can be used when conducting directed content analysis:
Inductive approach: The researcher goes through all the text and highlights what appears to be related to the theory. The content is then grouped under coding schemes and anything that does not fit into one of these schemes is grouped into a new category.
Deductive approach: The researcher uses predetermined codes to code the data. Any data that does not fall under one of the predetermined codes is classed as a new code or a sub-category of a pre-existing code.
106 Elo and Kyngäs (2008, p. 109) state, “deductive content analysis is used when the structure of analysis is operationalised on the basis of previous knowledge and the purpose of the study is theory testing.” Given that the main purpose of this research was to test the applicability of swift trust theory to a disaster management context, it was decided that a deductive content analysis approach would be most appropriate.
However, deductive content analysis can have several limitations, especially when it comes to the predetermined coding framework. Coding can sometimes be perceived as a methodological approach that seeks to reduce information and produce objective ‘truths’. However, coding does not always result in objective truths. Rather, building on the philosophical positions of this research, a social constructionism and interpretivist paradigm means, “since we each most likely perceive the social world differently, we will therefore experience it differently, interpret it differently, document it differently, code it differently, analyse it differently, and write about it differently” (Saldaña, 2013, p. 39). As this research aims to consider the applicability of the theory of swift trust to a disaster management context, there will be instances where the data needs to be reduced in order to test the theory. However, if this research is to understand a phenomenon as personal and subjective as trust, it must seek to draw out and explain in depth the respondent’s voice; not simply reduce it to objective ‘truths’. The coding framework adopted in this research aims to incorporate space for these nuanced views and, in doing so, seeks to provide new insights into swift trust within an environment where it has not been tested.
Following this point, deductive coding has also drawn criticism as it can become the sole method for analysing qualitative data (Saldaña, 2013). For this reason, content analysis encourages the use of other sources of information to ensure validity in the interpretations made from the data. This research has adopted data triangulation through using participant observation and document analysis to support any conclusions drawn from the coding of the interview data; which is considered important for trustworthiness of the analysis (Elo et al., 2014).
3.6.2. Context – the analytical construct A possible mistake when conducting content analysis is to analyse the data without reference to the theoretical and broader context in which it was recorded: what Krippendorff (2013) calls the analytical construct. Context is critical when analysing qualitative data (Kohlbacher, 2006). As Krippendorff (2013, p. 39) writes, “context specifies the world in which texts can be related to the analyst’s research questions.” That is, the particular theoretical lens through which a text is analysed must be made explicit to the reader, in order that they can understand the conclusions drawn by the analyst (Krippendorff, 2013). Apart from a theoretical context, it is
107 also important that the reader understands the broader context in which the data was obtained. Mayring (2014) argues that a text must always be analysed in terms of its circumstances of origin. For this research project, the most significant broader contextual elements were:
Cyclone Pam – in 2015, category 5 Tropical Cyclone Pam struck Vanuatu, causing unprecedented destruction to the region. As with TC Winston, Vanuatu’s government requested international assistance to help with the response and recovery. The international humanitarian community were highly responsive, but the subsequent response and recovery was not well received by the national government. The international respondents were criticised by the Vanuatuan government of not cooperating with their requests and, in essence, ‘running the show’ themselves. As Benjamin Shing, deputy chair of Vanuatu’s national disaster committee stated, “In nearly every country in the world where they [the international community] go in they have their own operations systems, they have their own networks and they refuse to conform to government directives” (ABC/AFP, 2015). R. Barber (2015, p. 2) also wrote, as part of a collaborative INGO report on Cyclone Pam, that there were “tensions between national and international responders”. She argues that the tension was on both sides, however,
“the international community must show much greater readiness to move away from ‘one-size-fits-all’ systems and procedures and understand its core role as providing surge capacity, technical advice and expertise to national actors to enable them to lead and coordinate disaster response in their own countries.” (R. Barber, 2015, p. 3)
Contextually, Cyclone Pam, just one year before TC Winston, tainted the reputation of the international community when responding to a national disaster in a developing nation in the Pacific. Significantly, whether this perception was warranted or not, the Fijian government did not intend to let the international community run the response and recovery in isolation from the EOC and wider government structure. Their hesitance pointed to a potential lack of trust in the international community’s response to requests for international assistance.
Information sharing – the Fijian government was very careful about how much information they shared with the international community. This concept is elaborated on further in the results; however, the sensitivities around the international response to Cyclone Pam, coupled with a perceived breach of trust that occurred early during the
108 response to TC Winston by the United Nations Office for the Coordination of Humanitarian Affairs (OCHA), the Fijian government were hesitant to share information with the wider humanitarian community. As a national government, they were determined to run the recovery themselves, with the international community supporting them; as opposed to the other way around.
Fijian Culture – the cultural history of Fiji plays a significant role in how Fijians relate to one another. All Fijians hold different types of cultural relationships with one another based on where they are born and their heritage. These cultural factors are significant as they can influence the development of trust between Fijians. Of particular importance is that international colleagues within the temporary system do not have these cultural relationships. The influence of culture on swift trust must be considered in the results. (Refer to Appendix B for an explanation of Fijian culture).
The Fijian Military and Australia – Fiji has been subject to several coups over the past decades; grounded partially in cultural tensions between the indigenous Fijian and Indo- Fijian populations (Tran, 2006). Most recently was the 2006 Fijian coup d’etat, where Commodore Frank Bainimarama, commander of the military, sought control of the nation from the then civilian government (BBC, 2016). A decade later, Bainimarama was officially elected as Prime Minister of Fiji. After TC Winston, the Fijian military were paramount to the response, having set up Future Operations and coordinating the next 48-72 hours of the response. However, it appeared that the 2006 coup influenced how some citizens perceived the military. As such, it was deemed that the involvement of the military in the response and recovery could affect swift trust formation between the civilians and military personnel in the temporary system.
Secondly, as part of the 2006 coup, Bainimarama removed the Fijian Police Commissioner Andrew Hughes (PIR, 2006); which was a contributing factor in Australia denouncing the actions of Bainimarama. However, tensions between Australia and Fiji go deeper. Over the coming years, sanctions were put in place by the Pacific Islands – Forum Island Countries (FIC) against Fiji, of which Australia is a member, in response to Fiji’s military government (Chapman, 2011). However, there have been calls for Australia to repair its regional relationship with Fiji and move past this tension (Herr & Bergin, 2011). As a result, Australia’s aid to Fiji after TC Winston could be considered both a humanitarian response as well as a political stabiliser for the relationship between the two nations. This complicated, political history may have an effect on the
109 formation of trust between certain personnel within the temporary system; especially when working alongside the Fijian military.
This broader situational context, coupled with the theoretical context, culminates in what Krippendorff (2013) calls the analytical construct. The analytical construct is the lens that is applied to the content analysis process and influences both the coding of data as well as interpretation of results:
“The analyst must, in effect, construct a world in which the texts make sense and can answer the analyst’s research questions. A context renders perceptual data into readable texts and serves as a conceptual justification for reasonable interpretations, including for the results of content analysis.” (Krippendorff, 2012, p. 24)
Thus, the application of context through the analytical construct is critical, as it orientates the reader to the analysis and its interpretations.
3.6.3. The process of content analysis The following diagram outlines the approach for the deductive content analysis undertaken for this research project.
Interview Content Analysis Answer research Interpretation data (including the question coding framework)
Validating Analytical construct evidence (triangulation)
Figure 3.6.1 Content analysis process - adapted from Krippendorff (2013)
Content analysis, as a method, has been used with relative freedom of late, such that, “content analysis is used to refer to any qualitative data reduction and sense-making effort that takes a volume of qualitative material and attempts to identify core consistencies and meanings” (Quinn Patton, 2002). However, Krippendorff (2012) provides a framework when undertaking content analysis. Figure 3.6.2 shows Krippendorff’s (2012) three stage content analysis process: requiring the unitisation of data, the development of the coding framework, and then the interpretation of data. This process is done in unison with the analytical construct, to contextually answer the
110 research propositions. The processes of unitising and coding the interview transcripts (using the coding framework developed) are described below:
Content Analysis
Interview Coding Unitising data framework
Answer research Interpretation questions
Analytical Construct
Figure 3.6.2 Coding framework - adapted from Krippendorff (2012, 2013)
Unitising Unitising is the process of systematically isolating portions of the interview that are to be coded (Krippendorff, 2013). Unitising provides a structured approach to coding a text and uses three different unit types. These are; sampling, recording, and context units (Krippendorff, 2013):
Sampling units – this refers to the texts that are chosen to be coded and analysed. Given that this research has used purposive sampling (as opposed to probabilistic or random sampling) (Riff, Lacy, & Fico, 2014), the sampling units remain as all the individual interviews.
Context units – are sub-units that group text within an interview into smaller sections that can then be analysed. For this research, a context unit is a portion of the interview transcript that addresses one of the theoretical conditions of swift trust.
Recording units – are the sub-units of content (within each context unit) that undergo coding for later interpretation. In this research, the boundaries for what would constitute a recording unit were determined by the dialogue for each question. This
111 would start by the asking of the question, and generally conclude with the answer to the question by the respondent. If any follow up questions were asked, then these would be included within the recording unit; if they were deemed to address the same context unit. If not, the additional information provided would be coded within another context unit if possible. However, if the data provided did not fit under any context unit, it would then be classified into a new context unit for later analysis, to support or amend the applicability of swift trust in a disaster management context.
Coding Once the interview data had been unitised, in order to address the research propositions, methods of coding were chosen, and a coding framework developed. This was done to extract the relevant data, so that it could be examined in a structured way (Gibbs, 2008).
Coding can be defined as, “identifying and recording one or more passages of text…[to] exemplify the same theoretical or descriptive idea” (Gibbs, 2008, p. 38). Different coding methods are available when conducting deductive content analysis. Saldaña (2013, p. 76) states that it is necessary to, “consider which coding method(s) may generate the types of answers…need[ed], based on the…research questions.” There are many ways to approach coding the data, and more than one approach may be chosen if deemed appropriate. However, whichever coding method(s) is chosen, it must be able to extract information from the data that answers the research question.
As previously explained, in order to analyse swift trust theory, this research project needed to balance reducing the interview data with maintaining the voice of the respondents and their personal perspective toward the formation of trust. To achieve this, it was necessary to use two separate coding methods; a process called simultaneous coding (Saldaña, 2013). While Saldaña (2013) outlines many coding methods that could be applied to interview data, this research has adopted hypothesis coding and descriptive coding.
Hypothesis coding Hypothesis coding is defined as,
“the application of a researcher-generated, predetermined list of codes onto qualitative data specifically to assess a researcher-generated hypothesis. The codes are developed from a theory/prediction about what will be found in the data before they have been collected or analysed.” (Miles & Huberman, 1994, p. 78)
112 Hypothesis coding compliments the deductive content analysis approach of this research, as it enables the researcher to use a pre-determined list of codes that can help determine whether the underlying propositions of swift trust theory are true, and thus whether the theory is applicable to a natural disaster management setting. If data falls outside one of these pre-determined codes, and it is deemed a new code is necessary to capture the respondents’ answers, then it would be developed, defined and utilised alongside the pre-determined codes. This is one of the benefits of hypothesis coding for this research project. It allows the researcher to test the applicability of swift trust in a disaster management context using pre-determined codes grounded in the theory, whilst also being adaptable to any unexpected data which may be captured through the introduction of a new coding term. Furthermore, any new coding term will become a marker to highlight data that does not fit the pre-determined codes, and thus, could challenge and extend the theory of swift trust. As Hsieh and Shannon (2005, p. 1283) write, “newly identified categories either offer a contradictory view of the phenomenon or might further refine, extend, and enrich the theory.”
Descriptive coding Descriptive coding is the use of a word or short phrase to summarise the core topic in a section of qualitative data (Saldaña, 2013, p. 70). However, the words or phrases used should not be summaries of the content, but rather act as identifiers of the specific topic being addressed. The benefit of descriptive coding is that it can be used to identify key topics within the research that complement the pre-determined codes, extrapolated from swift trust theory by hypothesis coding. Using both coding methods simultaneously can result in a deeper level of analysis, and a richer and clearer understanding of the data when considering each research proposition. An example of descriptive coding is as follows:
113
EOC 7: “Because in this operation, in the disaster management act and
disaster management plan, there is only one person authorised to talk Contractor – Director only to the media, not everybody, and addresses media that is the director. He’s the only
one…Other than that, none of us, so we must be very careful of the information that we release. Now Rapport – operational information we are talking about new people in affects trusting disposition the block; especially when you come from an outside organisation coming to work with government.
People are so eager to gather information, the number of people affected and all. But if it’s correct
(might do a hand action here of like, ‘that’s ok I guess’), if it’s not correct, then can get into trouble.”
Developing the coding framework In consideration of the two aforementioned coding methods, the coding framework was developed to provide a consistent approach to coding the data. The coding framework was produced using the following principles:
Each research proposition had specific criteria that required testing:
o Research proposition #1 tested whether the EOC fulfilled the criteria of a temporary system.
o Research proposition #2 tested the trust criteria of swift trust theory and their applicability to the EOC.
o Research proposition #3 tested whether the resilience criteria of swift trust theory were applicable to trust in the EOC.
114 Each proposition’s respective criteria had specific variables that could be tested to determine whether they were applicable/existed within the EOC. For example, for the EOC to be a temporary system, it must fulfil the criterion of being, ‘formed to address a task that is non-routine and requires interdependence’. As such, this criterion has three variables that required testing:
o that the EOC was formed;
o that it addressed a non-routine task; and,
o that the task required interdependence.
For each of the variables determined (from each swift trust criterion), a list of hypothesis codes was constructed (where applicable) to represent the theoretical answers that swift trust theory would predict. For example; swift trust theory proposes that swift trust will form based upon categorical assumptions (such as a colleague’s role) as opposed to evidence-based information (including personality and competency). As such, one of the interview questions asked was, “Of the local/international colleagues you had not met or worked with before, how did you judge whether to trust them?” The respondent’s answer was coded with one (or more) of the following hypothesis codes:
o Role (categorical assumption) – the respondent stated that they used their colleague’s role in the operation when deciding who to trust.
o Competency (evidence-based information) – the respondent stated that they trusted their colleague if they were proficient and competent in their role.
o Personality (evidence-based information) – the respondent stated that trust formed based on their colleague’s personality; such as their integrity, openness, kindness, living up to their word.
o Other – the respondent stated other trusting criteria in addition to the main criteria of role, competency, and personality. These additional criteria are then coded and grouped accordingly.
Descriptive coding was then used to complement the hypothesis codes by providing descriptive insights from the data. This included both further depth to support the hypothesis codes, or new codes that were not covered in the theoretically-derived hypothesis codes. For instance, considering the previous example about categorical
115 assumptions, the hypothesis code of ‘other’ was supported by descriptive codes describing additional trust criteria used by the respondents, such as ‘motivations’, ‘integrity’, and ‘cooperation’.
The coding framework, and subsequent codes, were applied to each interview to produce the analysis and subsequent interpretation of results.
Analytical memos It is important to note that coding and analysis are not successive processes but rather occur simultaneously; each process continually refining the other (Saldaña, 2013). As Weston et al. (2001) write, “coding is not what happens before analysis, but comes to constitute an important part of the analysis. There is a reciprocal relationship between the development of a coding system and the evolution of understanding a phenomenon.” An analytical memo provides an organised medium for recording any analysis that occurs throughout the coding process. They do not focus on summarising the data but reflecting on it. Therefore, analytical memos can be used to link codes, relate back to the theory investigated, and propose new questions to ask; anything that propagates analysis of the data.
This research used analytical memos throughout the coding process to record any insights learnt from applying the coding framework. This in turn refined the coding system, and became an essential resource for developing an understanding of the applicability of swift trust theory within a disaster management setting. An example is as follows (using the same comment from EOC 7 previously mentioned):
116
EOC 7: “Because in this operation, in the disaster management act and disaster management plan, there is only one person authorised to talk to the media, not everybody, and that is the director. He’s the only one…Other than that, none of us, so we must be very careful of the Analytical Memo (AM): It sounds like the information that we release. Now government are eager to protect their we are talking about new people in information and ensure it is all correct the block; especially when you before anything goes out. However, their come from an outside organisation perception is that other people and external coming to work with government. organisations are keen to produce People are so eager to gather information despite it maybe not being information, the number of people accurate. Compare this against their affected and all. But if it’s correct upcoming response about OCHA. (might do a hand action here of like, ‘that’s ok I guess’), if it’s not correct, then can get into trouble.”
117
ANALYSIS OF RESULTS RESEARCH PROPOSITION 1
The following three chapters provide an analysis and discussion of the results against each of the three research propositions. This chapter explores research proposition one, which determines whether the EOC was a temporary system. Firstly, it provides an analysis of the results against each of the five criteria used to determine a temporary system, in accordance with swift trust theory. It then discusses the implications of these finding in terms of both swift trust theory and the wider disaster management literature. It shows that the EOC can be classified as a temporary system and it provides several unique contributions to the disaster management literature. Specifically, the results showed that diversity in a disaster management temporary system includes taking on additional tasks outside of a member’s role and expertise. Also, interdependence within the system is contingent upon a member’s seniority and level of expertise. Finally, the data also showed that an EOC is a unique type of temporary system as it is not completely ‘new’; but rather formed by surge staff being incorporated alongside the pre-existing, permanent EOC members.
Introduction The purpose of this research is to understand how trust forms during the response to a natural disaster; and particularly to consider how trust forms for local and international disaster management personnel working in an EOC. As argued in Chapter 2, the theory of swift trust in temporary systems (Meyerson et al., 1996) presents the most applicable theoretical basis from which to understand how trust would form in an EOC. Subsequently, three research propositions
118 were formulated with reference to swift trust theory (Meyerson et al., 1996). They are restated below:
1. Research Proposition 1: During the response and recovery to a disaster in a developing nation, an EOC exhibits the properties of a temporary system.
2. Research Proposition 2: The trust that develops between disaster management personnel within an EOC is in accordance with the trust criteria proposed by swift trust theory (Chapter 5).
3. Research Proposition 3: The resilience of trust in a disaster management EOC is determined by the resilience criteria proposed by swift trust theory (Chapter 6).
This chapter explores research proposition one. First, it provides an analysis of the results against each of the five criteria that contribute to a temporary system in accordance with swift trust theory (as outlined in Chapter 2). Then, in relation to these five temporary system criteria, it provides a discussion of research proposition one, showing that the EOC was a temporary system. This discussion presents the implications of the EOC being a temporary system in relation to the theory of swift trust and the wider disaster management literature; as well as providing unique contributions to the literature.
Research Proposition 1: The EOC as a temporary system
To demonstrate the relevance of swift trust theory, it is necessary to first show that the EOC operated as a temporary system as the six trust criteria of swift trust rely on the characteristics of a temporary system. Therefore, if the EOC did not exhibit the qualities of a temporary system, then the development of trust within the system may not be characterised as ‘swift trust’. As such, the following chapter addresses each of the criteria of a temporary system (as outlined in Chapter 2). For the EOC to be a temporary system (Meyerson et al., 1996), it should:
1. Be formed to address a task that is non-routine and requires interdependence,
2. Have a task that is defined by clear and specific goals and have a time limit on its duration,
3. Involve diversely skilled people from a small labour pool and overlapping networks,
4. Include members that have a limited working history and the potential for working together again in the future, and,
119 5. Be assembled by the contractor.
The results relating to each of these criteria are discussed in turn below.
Formed to address a task that is non-routine and requires interdependence Considering the first criterion for a temporary system, this section will show that the EOC was formed to address the task of managing and coordinating the response and recovery, and that these tasks were non-routine and required interdependence.
4.2.1. Formed to address the task of managing and coordinating the response and recovery…………………………. Swift trust theory argues that a temporary system must be formed to addresses a task, rather than being a pre-existing team. The data showed that the EOC was a deliberately formed office that consisted of four teams: Current Operations, Future Operations, the National Emergency Operations Centre, and the National Disaster Management Office. This section will first explain the EOC prior to TC Winston. It will then discuss the incorporation of the additional teams after the disaster occurred. Finally, it will outline the tasks involved in managing and coordinating the response, and how each of the teams were formed to address these respective tasks.
Firstly, prior to TC Winston, the EOC was tasked (according to the National Disaster Management Plan),
“as the central agency to implement, and to coordinate the implementation of, policies of the NDMC [National Disaster Management Council] including disaster mitigation and preparedness.” (NDMP, 1995, p. 15)
That is, the EOC was tasked primarily with working on the pre-disaster stages of mitigation and preparation; as opposed to the post-disaster stages of response and recovery.
Once a state of national emergency was declared on 21st February 2016, the EOC became active for the operation, with Current Operations and the National Emergency Operations Centre forming, followed by Future Operations a few days later (after decisions were made on how the Fijian military were to be involved). Particularly, the National Emergency Operations Centre was the first team formed as per the National Disaster Management Plan, with the following purpose:
120 “During emergency operations, [the EOC] sets up within its precincts the National Emergency Operations Centre to coordinate the activities of disaster monitoring, warning and immediate post-disaster response including disaster relief work…The centre will normally be activated…when a disaster has occurred…[The] National Disaster Management Centre further provides operational and logistic support for the emergency operations.” (NDMP, 1995, p. 27)
The National Emergency Operations Centre was tasked with feeding information into Current Operations and Future Operations, which would plan the next stages of the response. However, through document analysis of the National Disaster Management Plan, which outlines the personnel and teams that are to respond to a national disaster, it became evident that Current Operations and Future Operations were not standard teams included in a response. Rather, it was found that Current Operations and Future Operations were two new teams that formed specifically for the TC Winston operation. Uniquely, their formation was a segmentation of the widespread responsibilities which fell to the EOC; given the EOC was charged with, “the implementation of disaster relief assistance” (NDMP, 1995, p. 57). This meant that the EOC incorporated two new teams into the response, both of which were formed specifically for the response to TC Winston and tasked with unique roles:
“The size of this disaster was unbelievable…so you had Current Operations, which is all the stuff coming in, and it was coming in thick and fast…And [the National Emergency Operations Centre], it was absolutely alive with people in there. Phones were running and stuff was going on in there the whole time…And then there’s all sorts of stuff happening here. And then you had Future Operations…that’s where all the military guys were. That room was chockers as well. Planning stuff going on and missions going out every day.” (EOC 10)
Furthermore, as outlined in the National Disaster Management Plan, the tasks of the response and recovery were divided into five categories: preparation for emergency operations, data collection and reporting, relief distribution and operations, rehabilitation, and post-disaster review. Each of these tasks had specific outcomes that were the responsibility of the particular team charged with implementing that task. Table 4.2.a shows each task category, the subsequent objectives and the team within the EOC responsible for implementation.
121 Table 4.2.a The tasks required for the response and recovery, their respective objectives, and the team(s) responsible for their implementation (NDMP, 1995)
Task Objectives Team(s) responsible
Organisation of the EOC (provision of equipment, staffing, reporting, etc.) National Disaster Management Office Preparation for Emergency Monitor any impending disaster’s progress National Emergency Operations Centre Operations Dissemination of warnings National Emergency Operations Centre
Obtain real-time information needed for running the operation National Emergency Operations Centre
Arrangement of air reconnaissance and air-ground surveys Future Operations National Emergency Operations Centre, National Preparation of daily briefings and situation reports Disaster Management Office, Current Operations Initial damage and relief needs assessment (within 48 hours) All teams Data collection Relief needs assessment report (within one week) All teams and reporting Damage assessment and outstanding relief needs report (within two weeks) All teams National Emergency Operations Centre, National Provision of relevant information to the divisions, public, agencies, and media Disaster Management Office, Current Operations Provision of Secretarial support to the Government and National Disaster Current Operations Controller Record all communications National Emergency Operations Centre
Mapping of relief, logistics, and impacted areas Future Operations
122 Task Objectives Team(s) responsible
Immediate planning of relief distribution Current Operations Relief distribution and Future planning of relief distribution Future Operations operations National Disaster Management Office, Current Coordination of relief distribution Operations, Future Operations Preparation of an on-going relief assessment National Disaster Management Office
Rehabilitation Coordination of on-going disaster relief assistance National Disaster Management Office Coordinate the preparation of a comprehensive rehabilitation programme (three National Disaster Management Office months after the emergency period) Conduct a post-disaster review within six months to formulate recommendations National Disaster Management Office for revision of procedures Post-disaster Provide recommendations for the revision of disaster plans National Disaster Management Office review Provide guidance to improve performance in future disasters National Disaster Management Office
123 The data showed that the EOC, in its fully operational form, was not a pre-existing structured team. Rather, the EOC, and subsequent teams of Future Operations, Current Operations, the National Emergency Operations Centre, and National Disaster Management Office, were formed to address the specific tasks required for the response to TC Winston. Thus, it meets the first element of criterion one required for the EOC to be classed a temporary system.
4.2.2. Non-routine Meyerson et al. (1996) defines a non-routine task as one which is unique relative to the established procedures within an organisation. The data showed that the task of responding to TC Winston was non-routine relative to the established procedures of the EOC outside of a disaster. This was demonstrated by the creation of new reporting lines within the EOC, the permanent EOC staff taking on new roles as part of the operation, and the tasks of the EOC shifting from pre-disaster mitigation and preparedness to post-disaster response and recovery activities.
Firstly, the EOC was not pre-established; rather it included multiple new teams (Current Operations and Future Operations) and was staffed with a high amount of surge staff (personnel brought into the EOC specifically for the operation). This led to a change to the established reporting structure adopted by the EOC outside of a disaster. Prior to a natural disaster, the EOC had a standard internal reporting structure (refer to Appendix A for the EOC organisation chart outside of a disaster), focussing on pre-disaster activities such as research, mitigation and preparation. The EOC was a hierarchical organisation led by the Director, who had two deputies underneath them, who, in turn, had staff reporting to them. However, the large amount of surge staff being included in the response, and the incorporation of Current Operations and Future Operations, meant that non-routine reporting channels had to be created between the teams. This was a unique requirement of the operation relative to the established procedures of the EOC prior to the response (refer to Figure 4.2.1).
124
Response data received from the The expanded EOC and four national divisional offices new reporting lines
National Emergency Operations Centre
Current Operations Future Operations
National Disaster Management Office
EOC prior to a natural disaster
Figure 4.2.1 The expanded EOC and new reporting structure adopted after TC Winston
(Field Note) “Prior to TC Winston, the EOC operated as a small team, all working under the Director as the head of the EOC. However, after the declared state of emergency, the EOC expanded to include the National Emergency Operations Centre, Current Operations and Future Operations. These teams all relied upon one another and so the EOC had to develop new ways to coordinate and share information between each team. As such, the National Emergency Operations Centre fed their information in Current Operations, who would then plan the next 24 hours of the response. Current Operations would liaise with Future Operations so they could plan the next 72 hours of the operation. After the declared state of emergency, the EOC changed from one small team into an interconnected network; with each team responsible for a separate aspect of the response.”
Secondly, this expansion of the EOC meant that the permanent staff from the National Disaster Management Office prior to the operation had to take on new and unique roles as part of the response.
125 Table 4.2.b The new roles adopted by the permanent EOC staff as part of the operation
Respondent Role prior to TC Winston Role in the EOC
Head of Current Operations and the EOC 17 Director EOC
Head of Emergency Planning and EOC 9 Head of Logistics Coordination Senior Advisor for Knowledge Disaster management support to the EOC 12 Management, Accounts and Permanent Secretary and senior Finance officials
EOC 4 Clerical Officer Support to the EOC
Procurement, purchase orders, and EOC 15 Accounts and finance timesheets
Procurement, purchase orders, and EOC 18 Accounts and finance timesheets
Head of Risk Management and National Emergency Operations EOC 6 Research Centre shift team leader
Thirdly, the regular tasks of the EOC are concerned with the pre-disaster stages of preparedness and mitigation. Given that natural disasters are rare, the EOC primarily coordinates and manages the following pre-disaster related activities.
Table 4.2.c Pre-disaster tasks and objectives for the EOC
Task Objectives
Coordinate disaster related projects implemented by the Fijian government, local and international agencies
Emergency Formulate, implement and monitor disaster management programs preparedness
Issue cyclone season warnings to relevant agencies
Prepare baseline data including; maps, population distribution, agricultural areas, locations of key facilities and infrastructure, Collation of emergency shelters, key personnel details. baseline data Maintain baseline data (reviewed annually)
126 Task Objectives
Assist the divisions and districts to prepare baseline data
Establish a simple but comprehensive database
Assist the divisions, districts and government agencies in the preparation Preparation of of their respective disaster management plans disaster management Ensure that the National Disaster Management Plan is updated and that plans district, division, government agency disaster management plans are viable Ensure that the EOC is operationally ready at all times for a national response Preparation of the EOC Assist the divisions and districts to ensure that their localised EOCs are operationally ready at all times for a national response
Formulate strategies on public education and awareness
Policy and Formulate policies for disaster related matters for the Fijian government strategy and other appropriate agencies development
Implement disaster policies put forward by the Fijian government
Organise an annual Disaster Awareness Week
Formulate, plan and organise in-country training programs
Training Identify overseas training programs and arrange participation
Arrange for international agencies to conduct in-country training
Formulate, plan and organisation disaster response simulation exercises
Maintain correspondence with the Fijian government, local and international agencies on disaster related activities and projects Correspondence Annually review and report on the divisions, districts and government agencies progress concerning disaster preparedness and mitigation
127 A comparison of the tasks and objectives post-disaster (Table 4.2.a) against the tasks and objectives pre-disaster (Table 4.2.c) show that the response and recovery involved entirely different (non-routine) tasks relative to the established procedures of the EOC outside of a disaster. There was a sudden change from updating disaster management plans and conducting training exercises, to actually upscaling and activating the operational EOC for data collection, reporting, and coordinating the distribution of relief.
Additionally, although the data showed that the response to TC Winston was a non-routine task, it also revealed two additional considerations concerning disaster management. Firstly, after a natural disaster occurs, an EOC moves from the pre-disaster stages of mitigation and preparation to the post-disaster stages of response and recovery. Subsequently, the roles of the permanent staff stationed within the EOC change in line with the needs of the response. Furthermore, the new post-disaster roles taken on by the permanent staff are similar to their previous pre-disaster roles, but emergency focussed, rather than concerned with preparation and mitigation (refer to the role comparisons in Table 4.2.b).
Additionally, the data also showed that the reporting structure of an EOC can be different pre-disaster to post-disaster, with the introduction of additional teams with their respective tasks and objectives (Table 4.2.a). This is significant to recognise because national responses are time pressured events that rely on the accurate communication of information between personnel. Therefore, it is important to consider the impact of introducing new reporting lines after a natural disaster, and what effect these might have on information dissemination and the efficiency of the response.
4.2.3. Required interdependence Meyerson et al. (1996) state that the task undertaken by a temporary system should, ‘require interdependence in order to arrive at a solution’. Swift trust theory defines interdependence as the continual interrelation between members of a temporary system when trying to find a viable solution to their problem.
To investigate interdependence within the EOC, each respondent was asked to record who they worked with throughout the response on the Network Mapping Worksheet. These interactions were discussed in more detail in the interview. This data was then summarised and tabulated to produce a network of the EOC; with the purpose of understanding whether the respondents required interdependence to complete their respective tasks as part of the operation. There were two steps used to code interdependence between a respondent and their colleagues:
128 If a respondent recorded on their Network Mapping Worksheet that they worked with a colleague as part of the response, then this was recorded as an initial source of interdependence.
These preliminary sources of interdependence were then followed up in the interview. An interaction was confirmed as a source of interdependence if the respondent stated that they worked with, or interacted with, that colleague as part of fulfilling their role in the operation. This coding process was utilised to produce the EOC network. Furthermore, the interview also gave a respondent the chance to include any colleagues that they may have missed on their Network Mapping Worksheet.
Given this approach, this section first explains interdependence between the respondents within the EOC. It then addresses interdependence between each team in the operation. Finally, it considers the impact that seniority and expertise had on interdependence.
129
Figure 4.2.2 The EOC network
(Node size is proportional to the number of connections recorded for the respondent as a percentage of the total recorded connections)
130 Table 4.2.d The number of connections for each respondent within the EOC as determined from their Network Mapping Worksheets
Percentage of Percentage of Number of total possible Respondent Team total recorded connections connections for a connections respondent EOC 17 Current Operations 22 65% 14%
EOC 7 Current Operations 16 47% 10%
EOC 9 Logistics 14 41% 9%
EOC 11 Current Operations 12 35% 7%
EOC 16 Current Operations 12 35% 7%
EOC 2 Logistics 10 29% 6%
EOC 18 Procurement 10 29% 6%
EOC 14 Current Operations 10 29% 6%
EOC 12 Support 9 26% 6%
EOC 13 OCHA 9 26% 6%
EOC 5 Support 8 24% 5%
EOC 4 Support 7 21% 4%
EOC 15 Procurement 5 15% 3%
EOC 10 Clusters 4 12% 2%
EOC 3 Media 4 12% 2%
EOC 8 Support 4 12% 2%
EOC 1 Media 3 9% 2%
EOC 6 Support 3 9% 2% 162 Total 100% (Average of 9)
Interdependence between the respondents Firstly, the aforementioned analysis of the data using the Network Mapping Worksheets and subsequent interview codes revealed that every respondent had some degree of interdependence with their colleagues during the response, and that a respondent had, on average, nine points of connection for the operation. Considering an analysis of each respondent’s Network Mapping Worksheet, coupled with observations in the EOC and the
131 coded interview data, the results showed that the respondents required interdependence for three main things within the operation:
Making and approving decisions (especially from the Director in the EOC):
“…[I would get] strategic guidance from the Director, then I’d convert that into what I needed to do, and then I’d talk to [my colleagues]…” (EOC 2)
“…before I released any situation report, it has to be endorsed by [EOC 7] and the Director, [so I talk to them about this].” (EOC 6)
“I talk to [the Director] by giving him guidance on decisions. There are cases that come up that he needs to know. I advise him of that.” (EOC 7)
“…[I would talk with] Current Operations [about] things that I could advise the PS [Permanent Secretary] on, what to do, what not to do…I would go to them daily just to get feedback from them.” (EOC 12)
Sharing of information:
“…[I would talk to the Director about] functional day to day sort of logistics operations type things, what he was doing, what I wanted him to do, what we needed to do…[my role] is an advisory role…” (EOC 2)
“OCHA collects 3 W information. You know 3 W’s? Who, what, where. It’s basically a table of what all the assistance that has been distributed or is planning to be distributed that is reported through the clusters. So, we collect that information…[and] then distribute [it] to…[the National Emergency Operations Centre] and Current Operations, and Future Operations…as well as specifically to [the Director]...” (EOC 13)
“…[talking with the Director] it would be on just on information and feedback that he would give on messages that he would like present out in the public. And we would also give him feedback on things that can go out, should not go out.” (EOC 1)
“[With the Director,] we talk about issues that need to go out to the public. Warnings, advisories, I talk about organising media briefings, interview with the local and international media and all this other stuff.” (EOC 3)
132 “[My colleagues would] update me about the situation reports that were coming from the divisions.” (EOC 6)
“…we [in Current Operations] receive information from all the four divisions and we get information from all the agencies…[and] we were the ones actually collating this information. We would present it to the Director...” (EOC 11)
“So under operations we had Current Operations, the [National Emergency Operations Centre] and Future Operations. So basically, [my role]…was to be the gel between the three. And to keep a way of ensuring there was networking so that the Director was informed of everything that was happening.” (EOC 16)
Producing outputs and deliverables for the response:
“We normally talk to [EOC 2] about supplies because he is a logistics specialist…He does all our stock, sharing to the divisions…He collates the whole numbers and then, he spits them out to the divisions and that’s how we [distribute relief].” (EOC 4)
“I talk to [EOC 14] about…how to calculate rations…[for] the sitreps.” (EOC 5)
“…I started doing mapping for [the EOC, and]…the communication from [Future Operations] to me was just mapping out response in stages. So ‘where’s this at?’, ‘where’s it coming to?’, you know, ‘how we going to get there?’.” (EOC 8)
“…[I would talk to my colleagues about] the pushing of items that need to go out, relief rations and relief items, logistics team and cluster registry….” (EOC 17)
“Ah, [EOC 2] was the main line I was talking to on the logistics, like the purchasing of [supplies]…[The Director also] asked [me] about like, to pick things up on the buying of the rations, not to slow down.” (EOC 15)
These results show that interdependence was a necessary part of the response and recovery for the respondents. Additionally, the data highlights that in a disaster operation, interdependence between the respondents is built primarily around: information sharing, seeking advice and approval for decisions, and producing outcomes for the operation (such as situation reports, purchase orders, relief distribution, etc). This is also important to recognise, because swift trust theory argues that swift trust forms based upon a respondent’s use of categorical assumptions.
133 However, each of these forms of interdependence are examples of evidence-based information; not categorical assumptions. Therefore, contrary to swift trust theory, the results suggest that for a disaster management context, the formation of swift trust may be impacted more by evidence-based information such as, how a colleague makes a decision, shares information, and produces their work, than by any categorical assumptions.
Team interdependence Secondly, by grouping the above coded data concerning each respondent’s interdependence with their colleagues by their colleague’s team, it was possible to determine team interdependence in the operation.
Table 4.2.e The number of connections for each team within the EOC
Number of Percentage of Number of Interacted with these Team Respondents total recorded Connections other teams in Team connections Logistics, OCHA, Support, Current 5 72 44% Clusters, Procurement, Operations Media Procurement, Current Support 5 31 19% Operations, Logistics, Media, Clusters Procurement, Current Logistics 2 24 15% Operations, Support, OCHA, Clusters
Current Operations, Procurement 2 15 9% Support, Logistics
Current Operations, OCHA 1 9 6% Logistics, Clusters
Current Operations, Media 2 7 4% Support
Current Operations, Clusters 1 4 2% Support, OCHA, Logistics
Total 18 162 100%
The data showed that Current Operations was the team most called upon for the operation (having 44% of all recorded interactions within the EOC). This was primarily due to the purpose of Current Operations as the immediate planning and coordination team for the response. This
134 team was set up such that each of the four members (EOC 7, EOC 11, EOC 14, EOC 16) had specific responsibilities in managing a part of the EOC while reporting to the Director (EOC 17); who had the responsibility of overseeing the entire operation.
“We all have our different areas of expertise. [EOC 16] does logistics and [EOC 14] does mapping on IT and all that. [“GIS and all that”] Yep. And we [EOC 11 and EOC 7] do all the other coordination stuff.” (EOC 11)
“…we [Current Operations] all work directly for the Director. And part of my role too is to supervise the operations inside the [National Emergency Operations Centre]…To supervise and ensure things done inside that room are in accordance with proper procedures...And also clear daily sitreps.” (EOC 7)
“My role was, um, firstly to assist with operations…[and] to be the gel between the [EOC groups]. And to keep a way of ensuring there was networking so that the Director was informed of everything that was happening.” (EOC 16)
This set-up of Current Operations, with each respondent managing a particular section of the EOC, meant that this team became a bottleneck for the response. Furthermore, the data showed that every team had to liaise with Current Operations; as Current Operations was the planning centre for the immediate stages of the response. Therefore, although there may have been interdependence required between the respondents to deliver the response, most of these interactions required Current Operations; and particularly approvals obtained from the Director. These results show that interdependence within the operation was not limited to individuals, but also included team interdependence. Furthermore, the results also showed that for a natural disaster operation and swift trust, the team charged with planning and coordinating the immediate response will be critical to the progression of the operation; given they are relied upon by the wider EOC, particularly for the approval of decisions.
The impact of seniority and expertise on interdependence Finally, the results also showed that the density of the network was 26%; with 162 connections recorded out of a possible 612. However, although the network density was relatively low, the connections that were coded and recorded were held primarily by those in positions of leadership and expertise. That is, the data revealed a unique relationship between the level of expertise of a respondent and their degree of interdependence within the operation.
Focussing on Table 4.2.e, all the respondents within Current Operations and the Logistics team (EOC 2, EOC 7, EOC 9, EOC 11, EOC 14, EOC 16, EOC 17), as well as EOC 12 (as a Senior Disaster
135 Management Consultant) and EOC 13 (the OCHA liaison) were either in positions of leadership within the EOC or experts in disaster management. Significantly, they attributed for 114 of the 162 points of connection within the network (70%). These respondents also had a network density of 37%; 11% higher than the entire network density average. Furthermore, these specific respondents held nine of the top ten positions in terms of connectivity in the network. This was in direct contrast to some of the junior or less trained staff, who held support roles (EOC 6, EOC 8, EOC 4), or those not in positions of leadership, such as the media team (EOC 1, EOC 3). These respondents specifically accounted for only 21 connections in the EOC, with a network density of 12%: below half the entire network average.
The data suggested that having a position of leadership or being an expert in disaster management (and the associated responsibilities that came with this expertise, such as working in Current Operations) resulted in a higher degree of interdependence within the operation. Considering swift trust theory, this finding is significant because it shows that those in positions of leadership or who are experts may be required to form more swift trusting relationships than their junior or less experienced colleagues when working in a response. Additionally, this result highlights a gap in swift trust theory. Swift trust theory does not consider if the number of trusting relationships impacts on how swift trust forms. That is, is there a correlation between the formation of swift trust and the quantity of trusting relationships a person has?
In summary, the task of responding to TC Winston required interdependence. Individually, the respondents relied on each other to make decisions, seek approvals, share information, and collaborate to produce work outputs. The data also showed that teams who plan and coordinate the response are relied on more in an operation. This is significant as these teams make and approve the decisions of the wider EOC. Lastly, leaders and experts within the operation have a higher degree of interdependence when compared to their junior colleagues. This highlights a gap in swift trust theory; which does not consider whether the number of trusting relationships being made impacts how swift trust forms.
A task that is defined by clear and specific goals and has a time limit on its duration………………………………. According to swift trust theory, for the EOC to be a temporary system, it should also have a task that is defined by clear and specific goals and have a time limit on its duration. This section will first show that the goals for the EOC were specific and clear. Secondly, it will describe the EOC’s limited duration.
136 4.3.1. The goals of the EOC being specific and clear Firstly, the theory of swift trust argues that the operation should have clear and specific goals around what the EOC, as a temporary system, was formed to achieve. The National Disaster Management Plan provided three clear overarching goals for a national operation. It stated that during the response to a natural disaster, “the primary concern is to protect life and property, the restoration of essential public services and to provide immediate relief to victims of [the] disaster” (NDMP, 1995, p. 43).
Secondly, as shown in Section 4.2, the EOC was tasked with five, more specific sub-goals that would guide the TC Winston operation; from inception to completion. The first four of these sub-goals, and their specific objectives, were:
1. Preparation for emergency operations (leading up to the operation) – readying the EOC for a national operation, monitoring a disaster’s progress, and disseminating warnings.
2. Data collection and reporting (during the operation) – obtaining real-time information including damage assessments; arranging reconnaissance and surveys; producing a relief needs assessment and report; preparation of daily situation reports, and reporting to government, divisions, agencies, and the media.
3. Relief distribution and operations (during the operation) – mapping of relief, logistics, and impacted areas; immediate and future planning of relief distribution; and coordination of relief distribution.
4. Rehabilitation (during the operation) – preparation of an on-going relief assessment, coordination of on-going disaster relief assistance, and coordination of a comprehensive rehabilitation programme.
Lastly, as the operation was still happening whilst the research was conducted, it was not possible to determine the overall success of the response and recovery to TC Winston. However, the final sub-goal of the EOC was to conduct a post-disaster review, which would measure the success of the three overarching goals of the operation, as well as the four sub-goals mentioned above (and their respective objectives). This post-disaster review had two objectives: to conduct a review of the overall operation within six months of the response period, to provide recommendations for the revisions of plans and procedures; and provide further guidance to improve performance in future disasters.
137 The data showed that the task of responding to TC Winston was defined by clear and specific goals. Furthermore, each of these goals were specific to the stages of the operation; from pre- response through to the lessons learned during the post-disaster evaluation.
4.3.2. The EOC’s limited duration Considering the duration of the EOC, the office was to remain active during the declared state of emergency. The period for which the EOC operated under the declared state of emergency was 30 days from the 21st February. This was then extended for an additional 44 days given the severity of the damage by the cyclone. As the declared state of emergency ended, the EOC began the hand-over process of the response to the Fijian Government, who would manage the ongoing recovery. Therefore, the EOC had a limited duration of 74 days as a temporary system.
Diversely skilled people drawn from a labour pool with overlapping networks Meyerson et al. (1996, p. 168) state that a temporary system is characterised by, “members who represent a diversity of functions or skills” and who are drawn from a labour pool with overlapping networks. As such, the following section addresses; the diversity of roles and skills of the respondents in the EOC, the labour pool these respondents were drawn from, and how the network of the respondents’ organisations overlapped.
4.4.1. The diversity of roles and skills of the respondents in the EOC Meyerson et al. (1996) state that a temporary system is characterised by personnel with diverse skills who are brought together to enact their expertise. Therefore, it was necessary to understand the diversity of the roles within the EOC, and the skills and expertise required of respondents in those roles. As such, at the start of each interview, every respondent was asked to provide a description of their role and what they did during the response. Table 4.4.a details each respondent’s role in the operation, the unique role expertise required for that role, and any secondary skills that were also required.
138 Table 4.4.a The respondents’ roles, their unique role expertise, and secondary skills
Respondent Role in the EOC Unique role expertise Secondary skills required for the role Photography and graphics Media – reporting and public Press releases and reporting Information management EOC 1 correspondence Preparation of public correspondence and EOC correspondence statements National and international correspondence Movement of supplies Logistics – International relations and Resource coordination EOC 2 Tracking of relief planning Purchase orders Production of a logistics support plan Leadership Press releases and reporting Information management EOC 3 Head of Media Preparation of public correspondence and EOC correspondence statements Human resource administration EOC 4 Support – EOC administration Information management Personal assistance to EOC teams EOC 5 Support – Current Operations Personal support to Current Operations Information management Strategic advice Support – National Emergency Operations EOC 6 Response shift team leadership Information management Centre shift team leader EOC administration Overseeing the production of the situation National and international correspondence reports Leadership EOC 7 Current Operations – Situation reports Strategic guidance of the National Information management Emergency Operations Centre Government coordination Information management EOC 8 Support – GIS and mapping Mapping and GIS EOC correspondence Personal assistance to EOC teams
139 Respondent Role in the EOC Unique role expertise Secondary skills required for the role National and international correspondence Movement of supplies Resource coordination EOC 9 Head of Logistics Tracking of relief Purchase orders Production of a logistics support plan Leadership National and international correspondence International cluster coordination and Leadership EOC 10 Head of Cluster Management management Strategic advice Cluster meetings and reporting EOC correspondence Information management Overseeing the production of the daily National and international correspondence briefing to the Prime Minister Leadership EOC 11 Current Operations – Daily briefings Strategic guidance of the National Information management Emergency Operations Centre Government coordination Leadership EOC 12 Support – Senior Disaster Manager Strategic advice to the Fijian government Information management Production of the United Nations situation reports Information management Production of the EOC response report Strategic advice EOC 13 OCHA Liaison Officer International correspondence between the National and international correspondence EOC and OCHA Leadership Oversight of resource coordination National and international correspondence Current Operations – Information EOC 14 Mapping and GIS for Current Operations Leadership management Information management Procurement Production of purchase orders EOC correspondence EOC 15 Head of Procurement Finance and accounting, including the Leadership response budget
140 Respondent Role in the EOC Unique role expertise Secondary skills required for the role Leadership Coordination and liaison for Current Information management Operations, Future Operations, and the EOC 16 Current Operations – Logistics National and international correspondence National Emergency Operations Centre. Government coordination Coordination of logistics Strategic advice Leadership Leadership of the EOC Information management Approval of spending National and international correspondence EOC 17 Current Operations – Director Public briefings and speaking on behalf of Government coordination the operation Strategic advice Reporting Procurement Procurement – Purchase orders and EOC 18 Production of purchase orders and EOC correspondence accounting timesheets
141 Firstly, an analysis of the data in Table 4.4.a showed that the EOC was comprised of disaster management personnel with diverse and unique roles, with no two respondents undertaking exactly the same role within the operation. Secondly, the results also demonstrated diversity by showing that each respondent held a specific set of expertise required to fulfil their respective role in the operation. Furthermore, each role within the EOC also required the respondent to demonstrate secondary skills in addition to their specific expertise.
Furthermore, the data also showed that although a member of an EOC may have a specific role and expertise, disaster management can be characterised by personnel having to take on additional roles that are outside of their area of expertise or having to work long hours. This was also the case for the EOC, and therefore the diversity of the respondents’ roles and skills was not limited to a single area of expertise.
“There were times when we had to do double jobs, sometimes triple jobs… I just had to absorb some other functions temporarily so the work could go on.” (EOC 1)
“…given the stresses during the first 30 days here…We’ve all got an extra workload… the long hours people put in to work.” (EOC 14)
“Winston was a category five cyclone. If it was a category three, one, you know, the extent of the work will not be as heavy as a category five cyclone. So sometimes you have to work 24 hours shift...” (EOC 3)
The data showed that ‘diversity’ in a disaster management context extends beyond having diverse roles and a specific set of skills and expertise; as per swift trust theory’s definition. Rather, given the requirements of a national operation, diversity also includes: being able to adapt to long working hours, taking on additional tasks outside of a respondent’s role, and demonstrating secondary skills in addition to a respondent’s respective area of expertise. These findings are also significant to the discipline of disaster management as they provide a detailed list of roles and the respective skills for disaster management personnel when working in an EOC.
4.4.2. The labour pool and overlapping networks For the EOC to be considered a temporary system, its members should be drawn from a limited labour pool and an overlapping network. According to swift trust theory, formation of trust within a system is directly impacted by the size of the labour pool the respondents are drawn from and the degree to which these networks overlap. This is because as the size of the labour pool decreases, and the degree to which these networks overlap increases, a respondent
142 becomes more vulnerable to reputational damage if they act in an untrustworthy way, as, “information about [their] performance diffuses more effectively” (Meyerson et al., 1996, p. 171). This is because it is more likely that the respondents will know one another and have a higher chance of working together again in the future. Therefore, the fewer organisations involved, and the more connected they are, the more important swift trust becomes. As such, this section will first show the number of organisations involved in the response and the implications this had on swift trust. It will then describe the degree to which these organisations overlapped as part of the response network and how this impacted swift trust formation.
Firstly, every respondent within the EOC was asked to provide the organisation they worked for prior to the operation. Each of these are recorded in Table 4.4.b.
143 Table 4.4.b The respective organisations that each respondent worked for prior to the national operation
Respondent Organisation Ministry of Rural and Maritime Development and National Disaster EOC 1 Management EOC 2 Australian Civilian Corp - DFAT Ministry of Rural and Maritime Development and National Disaster EOC 3 Management EOC 4 Emergency Operations Centre (EOC)
EOC 5 Ministry for Women, Children and Poverty Alleviation
EOC 6 Emergency Operations Centre (EOC)
EOC 7 Secretariat to the Pacific Community
EOC 8 Secretariat to the Pacific Community
EOC 9 Emergency Operations Centre (EOC)
EOC 10 Australian Civilian Corp - DFAT
EOC 11 Public Service Commission
EOC 12 Emergency Operations Centre (EOC)
EOC 13 United Nations Office for the Coordination of Humanitarian Affairs (OCHA)
EOC 14 Secretariat to the Pacific Community
EOC 15 Emergency Operations Centre (EOC)
EOC 16 Secretariat to the Pacific Community
EOC 17 Emergency Operations Centre (EOC)
EOC 18 Emergency Operations Centre (EOC)
This data showed that the respondents of the EOC were drawn from only seven organisations. Comparing this data against the potential number of organisations that can be involved in a response (refer to Section 1.4), this was a small amount of organisations. Additionally, it is important to note that many of the organisations involved were Fijian. This demonstrated that the response for TC Winston was locally led, with assistance provided by international
144 organisations. This is contrary to the response to Cyclone Pam and other disasters, which can be internationally led. The (primarily) local composition of organisations highlighted the intentionality of the locals to coordinate the TC Winston response and recovery.
However, even though there were only a small number of organisations involved in the response, swift trust is affected by the degree to which these organisations overlap to create a network in which the reputation of a respondent can be disseminated. If these organisations did not overlap, then the need for swift trust decreases because the respondents are less likely to work together in the future and their reputations are less vulnerable should they act in an untrustworthy way.
Therefore, the following process was undertaken to measure the degree to which these organisations overlapped. Firstly, using the Network Mapping Worksheets, as shown in Section 4.2.3, a network of the EOC was created based on the interdependence between the respondents and their colleagues during the response. This data was then grouped in terms of a colleague’s organisation prior to coming into the EOC. This created a table showing which organisations each respondent was connected to through their colleagues in the EOC (refer to Table 4.4.c). Lastly, Table 4.4.d provides a summary of the number of organisations each respondent was connected to through the operation.
145 Table 4.4.c The organisations each respondent was connected to through their colleagues in the EOC
Number of colleagues a respondent interacted with (sorted by their colleague’s organisation) Ministry of Rural Ministry for United Nations Department of Emergency and Maritime Women, Office for the Secretariat to Foreign Affairs Public Service Respondent Operations Development and Children and Coordination of the Pacific and Trade Commission Centre (EOC) National Disaster Poverty Humanitarian Community (DFAT) Management Alleviation Affairs (OCHA) EOC 1 1 1 EOC 2 3 EOC 3 1 1 EOC 4 2 3 EOC 5 3 1 1 1 1 EOC 6 2 1 EOC 7 2 1 1 1 1 EOC 8 1 1 2 EOC 9 1 1 1 EOC 10 1 1 EOC 11 2 1 2 EOC 12 2 1 2 EOC 13 2 1 1 1 2 EOC 14 2 1 2 EOC 15 3 EOC 16 1 2 1 1 EOC 17 1 2 2 1 2 EOC 18 1 5 Total 9 37 3 6 2 7 17 Number of respondents 2 7 2 1 1 1 4 from each organisation
146 Table 4.4.d The number of organisations each respondent was connected to in the EOC
Number of organisations the Respondent Role respondent was connected to Media – reporting and public EOC 1 2 correspondence EOC 3 Head of Media 2
EOC 10 Head of Cluster Management 2 Logistics – International relations and EOC 2 3 planning EOC 6 Support – response shift team leader 3
EOC 9 Head of Logistics 3
EOC 15 Head of Procurement 3
EOC 8 Support – GIS and mapping 4
EOC 4 Support – EOC administration 5
EOC 11 Current Operations – Daily briefings 5
EOC 12 Support – Senior Disaster Manager 5 Current Operations – Information EOC 14 5 management EOC 16 Current Operations – Logistics 5
EOC 7 Current Operations – Situation reports 6 Procurement – Purchase orders and EOC 18 6 accounting EOC 5 Support – Current Operations 7
EOC 13 OCHA Liaison Officer 7
EOC 17 Current Operations – Director 8
Considering Table 4.4.c and Table 4.4.d, an analysis of the data showed that:
The more respondents included in the response from one organisation, the more likely that an interaction within the EOC would involve a colleague from that organisation (with the exception of the ‘Ministry of Rural and Maritime Development and National Disaster Management’). Consequently, this meant that the more respondents from a
147 particular organisation working in the response, the more reach that organisation had across the network.
Every respondent in the EOC was connected to every organisation involved in the operation by an average of between two and three degrees of separation. Fifteen of the respondents were connected to every organisation through only two degrees of separation; with the highest being five degrees of separation. The data presents two main reasons why the degree of separation within the EOC had a low average. Firstly, there were only a small number of organisations involved in the response, making it easier for the network to be tightly connected. Secondly, there were some respondents connected to multiple colleagues that were all from different organisations; rather than multiple respondents from only one or two organisations.
Considering this last point, the number of organisations a respondent was connected to through their colleagues ranged from two to eight. The data showed that those in positions of leadership or who were experts were not more connected than their junior colleagues, ruling out a correlation between seniority in the EOC and organisational connectivity. Rather, the data suggested that the role a respondent had in the operation was a more accurate predictor of the number of organisations they would interact with. EOC 18 (Procurement), EOC 5 (Support to Current Operations), and EOC 13 (OCHA liaison) all had roles which were reliant on regular engagement with their colleagues; and therefore interacted with colleagues from six, seven, and seven organisations respectively. Conversely, the media team, which primarily worked alone or with external stakeholders in the response, only interacted with colleagues from two organisations.
Overall, the data suggests that the EOC was comprised of a small number of organisations that formed an overlapping network. However, the primary reason that organisations were closely tied was because a few individuals within the EOC worked with colleagues from many of the seven organisations involved. As shown in Section 4.2.3.3, it would seem reasonable to expect that the members of Current Operations would be most influential in the organisational network overlapping because they recorded the most interactions as part of the response. However, the data showed that leadership and expertise did not correlate with the number of organisations an individual was connected to. Rather, the data showed that role (particularly roles that relied on interdependence) was a better predictor of the number of organisations a respondent would interact with.
148 Considering swift trust, the data presents three implications. Firstly, for this disaster, the small number of overlapping organisations within the EOC increased the importance of swift trust development; given the greater risk that reputational damage from untrustworthy behaviour could easily disseminate throughout the network. Secondly, some respondents had higher degrees of organisational interconnectivity than their colleagues. This meant that these personnel had more reach across different organisations to be able to share untrustworthy behaviour if it did occur. This is important, because swift trust theory only considers the impact of the whole network’s interconnectivity on trust formation. It does not consider the interconnectivity of specific individuals in the temporary system; particularly those who are connected to many of the organisations involved in the operation. Thirdly, given there were only seven organisations involved in the temporary system, and the majority of these were Fijian, the involvement of Australia (through DFAT) and the United Nations (through OCHA) is important to recognise. Unlike their local counterparts, these two organisations rely on maintaining relationships in order to be invited to assist in an international response. Consequently, maintaining a trustworthy reputation becomes essential for these organisations as any untrustworthy behaviour can be widely disseminated amongst many national organisations, limiting their opportunities to work in that region in the future.
In summary, the data showed that the EOC included personnel with diverse roles and skill sets. Also, the labour pool (seven organisations) from which the respondents originated was small. The respondents’ interactions throughout the response created an overlapping network for both the respondents and their organisations. The degree to which the network overlapped was not only a function of the number of personnel sent to the operation from an organisation, but also the role a respondent had in the response.
Members that have a limited working history but the potential to work together again……………………………….. For the EOC to be considered a temporary system, its members should, “lack history with respect to previous contact with each other…[but] have a higher chance or interacting with one another in the future” (Meyerson et al., 1996, p. 187). Therefore, this section outlines the work history of the respondents in the EOC and discusses the potential of them working together again in the future.
4.5.1. The respondents’ work history Using the Network Mapping Worksheets, each respondent recorded who they worked with as part of the response. They were then also asked to record whether they had a pre-existing
149 relationship with that colleague prior to the response, or whether they had formed a new relationship with that person through working in the EOC. This data is tabulated in Table 4.5.a alongside the respondent’s role, whether they were a local or international colleague, and whether they were a surge staff member or a permanent staff member for the EOC.
150 Table 4.5.a The amount of pre-existing and new relationships for the respondents within the EOC
International or Surge Staff or Pre-existing New Respondent Role Local Permanent relationships Relationships EOC 1 Media – reporting and public correspondence Local Surge 2 EOC 2 Logistics – International relations and planning International Surge 3 EOC 3 Head of Media Local Surge 2 EOC 4 Support – EOC administration Local Permanent 3 2 EOC 5 Support – Current Operations Local Surge 7 EOC 6 Support – response shift team leader Local Permanent 3 EOC 7 Current Operations – Situation reports Local Surge 5 1 EOC 8 Support – GIS and mapping Local Surge 3 1 EOC 9 Head of Logistics Local Permanent 2 1 EOC 10 Head of Cluster Management International Surge 2 EOC 11 Current Operations – Daily briefings Local Surge 4 1 EOC 12 Support – Senior Disaster Manager Local Permanent 5 EOC 13 OCHA Liaison Officer International Surge 7 EOC 14 Current Operations – Information management Local Surge 4 1 EOC 15 Head of Procurement Local Permanent 3 EOC 16 Current Operations – Logistics Local Surge 4 1 EOC 17 Current Operations – Director Local Permanent 6 2 EOC 18 Procurement – Purchase orders and accounting Local Permanent 4 2 Total (81) 52 (64%) 29 (36%)
151 Table 4.5.b shows a comparison between the new and pre-existing relationships in the EOC for the surge and permanent staff in the operation.
Table 4.5.b Comparison of new and pre-existing relationships for surge staff and permanent EOC staff
Pre-existing New relationships Total relationships Surge Staff 22 26 48 Permanent Staff 7 26 33 Total 29 52 81
The data revealed that 64% of respondents in the EOC had worked together previously, or knew some of their colleagues prior to the operation. There were three respondents (EOC 2, EOC 5, and EOC 13 – all surge staff) who had no pre-existing relationships prior to the response. Similarly, 12 of the 18 respondents still formed new relationships as part of the operation.
Considering disaster management and swift trust, an analysis of the data revealed two key factors:
An EOC can have a team of respondents who have worked together (in some capacity) prior to the present operation.
Pre-existing relationships are beneficial for those in positions of leadership and seniority.
Initially, the results showed that the EOC may consist of a team of personnel who have worked together prior to the current response. The data suggests that a contributing factor for this scenario is the way an EOC is assembled. Unlike other temporary systems that form to create a new team, an EOC brings in specialised surge staff to help the permanent EOC staff meet the requirements of a national emergency. Therefore, the permanent staff know one another prior to an operation occurring, and the majority of new relationships in the EOC are experienced by the surge staff (as was the case for the EOC, shown in Table 4.5.b). Consequently, this research shows that for swift trust in a disaster management context, it is important to consider the implications of how the temporary system is formed, due to the integration of surge staff. An EOC is unique to other temporary systems in that it is not completely ‘new’, but rather a combination of an established team and additional surge staff, woven together into the EOC, to create a new structure to coordinate the response (refer to Section 4.2.1).
152 Secondly, the data also suggested that those in leadership were more likely to have pre-existing relationships both with their colleagues in the EOC, and with colleagues in other government ministries supporting the response. This was shown particularly through the Current Operations team (the coordinating team of the EOC), who stated that having pre-existing relationships was helpful in coordinating the response and working together.
“I am from one of the central agencies in government. So for me, that is actually a bonus because in this exercise you are working with officers from all organisations in government and…when I want to speak someone, when I want something, I know exactly who to call and if it’s someone who has to provide information, there’s a high, very high possibility that I will already know that person…So [EOC 7 also] is just like me, [they] just about knows everybody in the government.” (EOC 11)
“…we knew each other beforehand, we’d worked together, in some cases we’ve trained together…So when you come in, you already know what turns you on, what turns you off.” (EOC 16)
“The only person that I met for the first time was [EOC 13]. Otherwise all the others I’d worked with them before…because I’ve been working with [my colleagues] for so many years.” (EOC 7)
“…if there was issues with operations, I’d talk to [the Director. Otherwise]…I just call up the minister in the front here. ‘This is the issue we have, we need a decision’. Dah dah dah dah, done!” (EOC 16)
The data showed that each member of Current Operations had a higher number of pre-existing relationships than new relationships. Consequently, each member of Current Operations could draw on their colleague’s network of relationships if they needed to be connected to another person in (or outside of) the EOC. Therefore, unlike swift trust theory, for those in positions of leadership, having a pre-established network of relationships prior to a national disaster operation was beneficial in delivering the response; as opposed to having to develop these under the time pressures of a national emergency.
4.5.2. Potential for working together again in the future Each respondent in the EOC was asked whether they expected to work with their colleagues again in the future. In response, 100% of the respondents said that they did expect to work with their colleagues again in some capacity. An analysis of the data showed that there were three ways in which the respondents expected to work together again in the future.
153 Through another national operation. Each of these respondents were surge staff and not permanent members of the EOC (apart from EOC 18).
“…yes [I expect to work with my colleagues again] during an operation such as this.” (EOC 1)
“…the next disaster, Zena, as it’s coming in now, probably do some work on that with them. And then next season as well if that comes off.” (EOC 2)
“I specialise in a lot of national operations. So in other national operations…” (EOC 7)
“…[A] response, such as this…” (EOC 8)
“…[I expect to work with my colleagues again] mostly around response yeh.” (EOC 14)
“…[I expect to work with my colleagues again] probably in the logistics.” (EOC 18)
The EOC and the work conducted by this office outside of a national operation. Apart from EOC 13, each of these respondents were permanent staff within the EOC.
“…we work together in this office, most of those guys.” (EOC 4)
“…because in my normal role, I work with the disaster service liaison officers, so I liaise with most ministries.” (EOC 6)
“…I will be working with these people…[in the future through] national duties…when I say national duties, it means elections, population census…we do a lot of public relations.” (EOC 9)
“I work with [my colleagues] inside these four walls. So disaster stuff for this time.” (EOC 12)
“…[I expect to work with my colleagues over] the long term…[considering] preparedness for the next one.” (EOC 13)
External training and disaster management correspondence with the government:
“…for example, international meetings and conferences…So we work in government and we meet all the time.” (EOC 7)
154 “Oh definitely…[in] training…” (EOC 8)
“…I conduct training at a regional and national level. A lot of the people will be there.” (EOC 11)
Considering these results, the data highlighted an important implication for swift trust in disaster management. The results showed that the members of an EOC have additional ways in which to work together in the future, apart from coming together for another national emergency (the sole reason the temporary system was created). Particularly, the permanent staff of the EOC responded that they expected to work with their colleagues as part of their on-going duties within the EOC. Similarly, EOC 7, EOC 8, and EOC 11, all disaster management experts or senior members in Current Operations, said they expected to work with their colleagues again through training, conferences, and other government duties. The continual requirements of effective disaster management, including mitigation, preparation, and ongoing training events, make future interaction for disaster management personnel likely.
These results are significant because swift trust theory argues that this heightened expectation of future interaction means the respondents are less likely to expect untrustworthy behaviour from their colleagues (or to act in an untrustworthy way themselves); increasing the likelihood of swift trust forming. Secondly, the results also show that the field of disaster management offers multiple ways in which disaster personnel might work together again; which is not limited to future interactions only through the same type of temporary system (an EOC formed for a national response).
Overall, the results showed that 64% of the relationships in the EOC were pre-existing, and 12 of the 18 respondents experienced new relationships. Most new relationships were experienced by the surge staff brought into the EOC as part of the response. However, those in positions of leadership and seniority (particularly in Current Operations) knew some of their colleagues prior to the response; which was an asset in coordinating the operation. Lastly, every respondent expected to work with their colleagues again in the future; given the multiple work areas within the field of disaster management.
Assembled by the contractor Swift thrust theory argues that in a temporary system, the respondents are, “assembled by a contractor” (Meyerson et al., 1996, p. 169). The theory defines the contractor as the person usually in charge of the temporary system, and therefore the one responsible for bringing the respondents together (Meyerson et al., 1996). Therefore, this section will first show that the
155 Director (EOC 17) was the head of the EOC and subsequently, that they were the contractor. It then will explore whether the Director assembled the respondents into the EOC.
4.6.1. The Director as the contractor An analysis of the data showed that there were five different ways in which the respondents identified the Director (EOC 17) as the contractor of the EOC: that is, the person who was the head of the operation:
Multiple respondents said that they had to get approval from the Director for decisions, documents and actions concerning the operation.
“…things we need to pass up to the Director. Get our minutes passes up to him, he gets it signed, and comes back out…” (EOC 4)
“…because of the government hierarchy, you have to follow the steps…[and you cannot] overstep [the Director].” (EOC 16)
“…a sitrep report…goes to the Director for vetting.” (EOC 5)
“If there were high level decisions that needed to be made, then it was through [the Director].” (EOC 14)
“…we don’t have that much authority to do things with [the Director’s] consultation eh? Because if we have to do anything from here, we have to consult him with approval, and then we can do it.” (EOC 15)
Respondents referred to the role of the Director as the coordinator of the EOC, as well as unique aspects of their role as head of the operation.
“…[the] Director…he coordinates the [EOC].” (EOC 18)
“…the Director, he has the coordination aspect, the responsibility, coordination with everybody.” (EOC 9)
“Because in this operation, in the Disaster Management Act and Disaster Management Plan, there is only one person authorised to talk to the media, not everybody, and that is the Director.” (EOC 7)
Respondents commented that there was a need to keep the Director informed of what was happening in the operation; given their role as the contractor of the EOC.
156 “With the Director, it would be on…information and feedback that he would give on messages that he would like present out in the public.” (EOC 1)
“Because the Director, he wants to know everything…[so I] go through him for everything.” (EOC 6)
“…I have to let [the Director] know what is happening in my project.” (EOC 8)
“…I’d always keep [the Director informed]. For major things, not for day to day. But, like, if things were going on, I’d just generally update him, say, ‘this is what is going on, what do you think?’” (EOC 10)
“…[my role was] was ensuring there was networking so that the Director was informed of everything that was happening.” (EOC 16)
“…all the information…[is] sent to the Director so he knows what is going on, [and then] he makes decisions…” (EOC 2)
Two respondents said that they would approach the Director for strategic guidance.
“…[I would get] strategic guidance from the Director…” (EOC 2)
“…if there was issues with operations, I’d talk to [the Director].” (EOC 16)
Respondents stated that they reported to the Director, even if they were from an organisation external to the EOC. This applied to both surge staff and permanent staff, as well as senior personnel from Current Operations (EOC 7, EOC 14).
“Well [EOC 9 and I] are the two principle officers here. We are the Deputy Directors…We work directly under the Director…” (EOC 6)
“…I report to [the Director] and not my line manager when it comes to response work.” (EOC 14)
“…my supervisor…during the operation, it’s the Director.” (EOC 3)
“Yeh, [Current Operations], we all work directly for the Director.” (EOC 7)
“I report to the Director.” (EOC 8)
Although the results showed that the Director was the head of the EOC and contractor, it remains to be explained whether they actively assembled the respondents into the EOC.
157 4.6.2. Whether the Director assembled the respondents in the EOC The National Disaster Management Plan states multiple times that the Director of the EOC is responsible for staffing a national response.
“The Head of [the EOC, the Director]…is responsible for the staffing and equipment of the centre during the emergency operation…” (NDMP, 1995, p. 27)
However, although there was a prerogative for the Director to assemble the staff within the EOC, the data showed that this only happened for some of the respondents.
Before discussing the results, it is important to note that the EOC comprised of both permanent and surge staff. Consequently, this section only considers the surge staff brought into the response; as it was not necessary for the Director to assemble the permanent staff in the EOC: they were already employed by the EOC to be included in an operation.
To learn whether the Director assembled the surge staff into the EOC, each respondent was asked to provide the person or organisation that asked them to come help in the response. This data is displayed in Table 4.6.a.
158 Table 4.6.a The person/organisation that asked each surge staff respondent into the EOC
Person/Organisation that Surge staff Respondent asked them into the Surge staff respondent’s role response Media – reporting and public EOC 1 EOC 3 correspondence Australian Civilian Corp - Logistics – International relations EOC 2 DFAT and planning EOC 3 EOC 17 – Director Head of Media
EOC 5 EOC 11 Support – Current Operations Current Operations – Situation EOC 7 Self reports EOC 8 EOC 12 Support – GIS and mapping Australian Civilian Corp - EOC 10 Head of Cluster Management DFAT Current Operations – Daily EOC 11 Self briefings EOC 13 EOC 17 - Director OCHA Liaison Officer Current Operations – Information EOC 14 EOC 17 - Director management EOC 16 EOC 17 - Director Current Operations – Logistics
The results show that for the surge staff, the Director only specifically asked four respondents (EOC 3, EOC 13, EOC 14, and EOC 16) to help in the response. For the remaining respondents:
Both EOC 2 and EOC 10 were sent to the operation by the Australian Government;
EOC 1, EOC 5 and EOC 8 were invited into the operation by senior colleagues, who they would support through their respective roles; and,
EOC 7 and EOC 14, as disaster management experts, reported to the EOC of their own accord.
“I pick up myself and I came, because I have a responsibility for my country. No- one told me to come here…I just reported here because I know my services are needed.” (EOC 7)
“We brought ourselves in.” (EOC 14)
159 Contrary to Meyerson et al’s (1996) criterion of a temporary system, the data showed that the Director did not assemble the respondents in the EOC. However, the results did reveal two important implications concerning swift trust and disaster management.
Firstly, as was the case in Fiji, many nations already have an EOC manned by permanent staff ready to engage in a disaster response should it arise. Therefore, in this setting, the role of the contractor (as the head of the EOC) in assembling the members for an operation is already limited. However, swift trust theory would argue that swift trust formation will be impacted by the head of the EOC’s active engagement in assembling the remaining surge staff to work in the EOC. This is because, in assembling these respondents, the contractor is selective about who to include in the operation. Consequently, the contractor’s selectivity helps the respondents to swiftly trust one another because they know that the contractor has vetted their colleagues.
“Inclusion [in a temporary system] is presumed to imply selectivity on the part of the contractor, and these judgments, in turn, are presumed to be predicated on sensible and more or less conscious criteria. Here, the credibility of the contractor – in terms of his or her reputation for creating and composing successful temporary systems – serves as a useful substitute for interpersonal history.” (Meyerson et al., 1996, p. 185)
Therefore, the results suggest that because the majority of respondents in the EOC were not selected for inclusion by the Director, the respondents were not able to use this ‘implied selectivity on the part of the contractor’ as a basis for forming swift trust with their colleagues.
Secondly, the data showed that some temporary systems, such as an EOC, do not have the opportunity to be created using respondents assembled only by the contractor. Rather, operational EOCs are usually a culmination of surge staff working alongside permanent staff. This is important to consider as, although swift trust theory emphasises the role of the contractor in selecting the members of a temporary system, it does not consider this unique case (which is common to the field of disaster management). Furthermore, the theory also does not consider the fact that any permanent staff in an EOC may not have undergone the process for inclusion by the contractor; unlike the surge staff. Consequently, this has an impact on trust formation as the permanent staff are able to rely on the contractor’s selectivity as a, “substitute for interpersonal history”, whilst the surge staff cannot.
Discussion of Research Proposition 1 Considering the previous results, the purpose of this section is twofold. Firstly, it will address research proposition one by showing that the EOC was a temporary system. Secondly, it will also
160 explain the implications that this has on the theory of swift trust in temporary systems as well as the wider disaster management literature. This is shown through a discussion of each of the five temporary system criteria previously analysed. That is, the temporary system must:
1. Be formed to address a task that is non-routine and requires interdependence.
2. Have a task that is defined by clear and specific goals and have a time limit on its duration.
3. Involve diversely skilled people from a small labour pool and overlapping networks.
4. Include members that have a limited working history and the potential for working together again in the future.
5. Be assembled by the contractor.
4.7.1. Formed to address a task that is non-routine and requires interdependence The case study revealed that the EOC was formed to coordinate the national response, and that the national operation was a non-routine task that required interdependence. These conclusions, and their implications for swift trust theory and the wider literature are discussed in the following three sub-sections.
Formed as part of the response An analysis of the results showed that the EOC was formed in order to address the national response and recovery to TC Winston. Although the EOC was manned by a skeleton staff prior to the operation, the post-disaster tasks and objectives of the operation meant that additional staff were required to help coordinate the response. The research found that Current Operations and Future Operations were two new teams that formed specifically to coordinate operational planning. Uniquely, their formation was a segmentation of the widespread responsibilities which fell to the EOC. This finding validates Goodman’s (1981) and Meyerson et al’s (1996) research on temporary systems, showing that the formation of the national EOC was a reaction to the importance assigned to the operation by the Fijian Government. That is, the task of responding to TC Winston was deemed significant enough that the EOC was specifically expanded; fulfilling one of the criteria for a temporary system. Secondly, the ‘first-time’ formation of Current Operations and Future Operations reinforced the findings of Kendra and Wachtendorf (2003), Waugh and Streib (2006) and Mendonca, Beroggi, and Wallace (2001) by showing that responding to the dynamic requirements of a natural disaster requires ingenuity
161 and flexibility. The previous structure adopted by the EOC was deemed unsuitable to meet the challenges of the response; therefore, the inclusion of Current Operations and Future Operations to coordinate the next 24 and 72 hours of planning respectively provided an example of where disaster management requires, ‘ingenuity’ (Kendra & Wachtendorf, 2003), ‘flexibility’ (Mendonca et al., 2001), and, ‘innovation and adaptation’ (Waugh & Streib, 2006).
Finally, the incorporation of Future Operations into the response also provided an extension to the literature concerning military involvement in a natural disaster response; particularly given the response operated, as Walker (1992) describes, as a formal partnership between the Fijian, Australian and New Zealand military forces. Currently, much of the literature concerning military involvement in a response is focussed on the United States. Therefore, this case is a unique contribution concerning the incorporation of military forces within a national operation in Fiji. Foremost, the case study showed that military involvement in a response can extend much beyond what Anderson (1970) would consider the usual period of involvement; that is, the first couple of days during the emergency period. Rather, Future Operations played an ongoing role, providing intelligence and logistical support (Noji, 2005). Most pertinently, as described by Anderson (1970), this case study provided another example of where the military was asked into the operation by the government, and subsequently responded to civilian authority (by Future Operations reporting ultimately to the Fijian Government). This reinforced the research of Walker (1992) by providing an instance in the Pacific where the military were held accountable for their involvement in a government led operation (with significant civilian leadership).
However, this case study is also significant and unique for two further reasons. Firstly, given the previous coup history within Fiji and the relationship between the National Government and the military, the involvement of the military in the TC Winston response and recovery could have turned to a leadership struggle rather than a cooperative effort. However, the research offers an example of military and government cooperation after a natural disaster against the backdrop of a contentious history. Secondly, the case study highlights the importance that Australia placed on their relationship with Fiji; particularly considering their denouncement of Bainimarama (the then Prime Minister of Fiji) and his military takeover of the government. Given Australia’s military forces were working alongside the Fijian military, Australia was actively taking steps to, as Herr and Bergin (2011) suggest, repair their regional relationship with Fiji and move past their recent historical tensions. As such, the case study contributes an example to the work of Birkmann et al. (2010) by showing that a request for international aid, and the subsequent multi-national operation, can be utilised as a medium to help repair international relationships.
162 Non-routine The case study also revealed that the response to TC Winston by the EOC was non-routine due to three factors. Firstly, the change in the structure of the EOC, due to the introduction of Current Operations and Future Operations, meant that new reporting lines had to be developed. Secondly, the response was found to be non-routine as the permanent staff of the EOC adopted new roles specifically for the response. Thirdly, the operation was shown to be non-routine due to the shift from the pre-disaster tasks of preparedness and mitigation to the post-disaster tasks of response and recovery. The data confirmed the operation conducted by the EOC was non-routine in accordance with the criteria set by Meyerson et al. (1996). That is, the adoption of new reporting lines, roles and tasks in the EOC showed that the response was unique, “relative to [the] routine procedures available in the organisation” (Meyerson et al., 1996, p. 168).
Additionally, the results also confirmed the concept of emergence (E. L. Quarantelli, 1999) as a critical part of disaster management. Emergence in a disaster management context is defined as the, “appearance of new groups and new ways of doing things…[in order to] deal effectively with disaster generated needs and difficulties” (E. L. Quarantelli, 1999, p. 8). The inclusion of Current Operations and Future Operations as new planning centres, the adoption of a new reporting structure, and the permanent EOC staff taking on new response-specific roles, showed that the concept of emergence was prevalent in the response to TC Winston. However, considering the post-disaster review (which is a responsibility of the EOC), it is recommended that further research be conducted to gauge the necessity and value of emergence in relation to the overall success of the operation; and therefore, if emergence is actually a necessary part of an EOC’s response and recovery.
Lastly, it is recommended that further research also be conducted to understand the impact that changing the structure of an EOC (including introducing new reporting lines) might have on a response. This is important to consider, as the primary role of an EOC is to coordinate the response (Militello et al., 2007; R. W. Perry, 1995; Wildman et al., 2012) and disseminate information. Any changes that may impact this already complicated process need to be understood.
Required interdependence The data found that the EOC exhibited two forms of interdependence: personal interdependence and team interdependence. Foremost, this finding further refines the criteria of interdependence in a temporary system, as outlined by Goodman and Goodman (1976) and Meyerson et al. (1996), by showing that interdependence occurred in two different forms.
163 Personal interdependence in the EOC was centred around three factors: making and approving decisions, sharing information, and producing outputs and deliverables for the response. These results validate the findings by Kendra and Wachtendorf (2003, p. 39), who argued that, “The EOC…allows for interpersonal communication, technically supported information exchange, and decision-making among the representatives of different agencies…”. Considering swift trust formation, the findings also challenge the theory proposed by Meyerson et al. (1996). That is, contrary to swift trust theory, the results suggest that for a disaster management context, swift trust may be impacted more by evidence on how a colleague makes a decision, shares information, and produces their work, than by any categorical assumptions.
Considering team interdependence, the case study found that although each team worked together in delivering the response, they all relied on Current Operations for strategic guidance and approval of decisions. The formation of Current Operations as the central planning unit for the operation was important to ensure that response data was directed to a central location and then disseminated throughout the wider EOC. This finding reinforces the work of Militello et al. (2007), who state that all information within the EOC should flow to a core team, to enable a more coordinated response.
Finally, this research provided a unique contribution to the disaster management literature and swift trust theory by showing that a respondent’s degree of interdependence was impacted by their level of seniority and expertise within the EOC. This finding furthered the work of Militello et al. (2007) who found that, in disaster training exercises, those with more experience were more engaged in the exercises and took on positions of leadership compared to their less experienced counterparts. Importantly, this research adds to Militello et al. (2007) by showing that for an actual disaster response, those with more experience not only took on positions of leadership, but that their degree of interdependence was higher when compared against their less experienced and junior counterparts. This finding also contributes to swift trust theory, as it shows that the degree of interdependence within a temporary system can be affected by the seniority and expertise of those within the network; which in turn, impacts the number of trusting relationships that need to be developed. Subsequently, further research is required to understand whether there is a correlation between the formation of swift trust and the quantity of trusting relationships a person must make.
164 4.7.2. A task that is defined by clear and specific goals and has a time limit on its duration…………………………………. The results showed that the response to TC Winston was a task that had three overarching goals, supported by clear and specific sub-goals and objectives. These findings reinforce Comfort and Kapucu (2006), who argue that shared goals, made clear to the respondents, helps them to adapt to the dynamic operating conditions of the response and recovery. Similarly, Kendra and Wachtendorf (2003) state that having response policies and procedures is important, and that these should be known by any respondents. Adding to Kendra and Wachtendorf (2003) though, the results showed that it is not enough for any response policies and procedures to be known; they must also be specific, which was the case in the TC Winston operation.
Secondly, Haas et al. (1977) state that disaster management can be subject to competing ideas regarding the desired goals of a recovery. However, the results of this research suggest that these concerns can be alleviated if there is a detailed, clear and specific set of goals from the outset of the operation. The data also suggests, as Jones and Faas (2016, p. 170) argue, that in a disaster response, “trust in relationships is enhanced by…identification with a common cause.” That is, having clear and specific goals and objectives for the operation may actually help trust form and enhance trusting relationships.
Finally, the case study showed that the EOC had a time limit on its operational existence: the declared period of a national emergency (74 days). Whilst it had a limited duration, it is important to recognise that disaster management is not a linear process of mitigation, preparation, response and then recovery. Rather, as Coppola (2006), E. L Quarantelli (1999) and Carter (2008) discuss, each of these four stages are linked and interdependent; so it is difficult to know when one stage starts and where one ends. Furthermore, given that the requirements of the response change as the operation progresses (as suggested by the respondents in this research) (R. W. Perry, 1995), the data showed that it if an EOC is to be considered a temporary system, it needs to be bound by a time limit. Otherwise, the duration of the EOC could organically extend to mirror the interdependent stages of disaster management and would not be, as Militello et al. (2007, p. 25) state, “ad hoc”; which Meyerson et al. (1996) argue is a critical requirement for swift trust.
4.7.3. Diversely skilled people, the labour pool and overlapping networks The case study showed that the respondents in the EOC had diverse roles and skills, and that they were drawn from a small labour pool consisting of overlapping networks. These criteria are addressed in the following sub-sections.
165 The diversity of roles and skills of the respondents in the EOC The results provided a unique contribution to the literature by providing a list of the roles for respondents involved in an EOC, as well as the unique expertise that was required to fulfil each role. The findings extended Meyerson et al’s (1996) definition of diversity, moving it beyond that of roles and skills, providing a unique contribution to the disaster management and swift trust in temporary systems literature. For working in an EOC, diversity also includes taking on additional responsibilities outside of one’s area of expertise and working longer hours due to the requirements of the national response. This finding also extends the applicability of Weick’s (1993) concept of ‘virtual role systems’, which describes how a person can take on additional roles when pressure is applied so that the organisation can continue to function, to the context of disaster management. Similarly, it provides a contemporary example of the ‘expansion of professional tasks’ as argued by Lalonde (2004), who states that during a disaster, personnel are required to take on tasks that are considered outside of their area of expertise.
Secondly, the results provide an extension to the disaster management literature by highlighting the skills and expertise required by those working in specific roles within an EOC. Multiple authors, including Carter (2008) and Anglin (2011), focus on the skills required to be a leader of the EOC. However, they do not consider the skills and expertise necessary to be a member in one of the other teams that operate in the EOC, such as logistics, procurement or media. This research provides one step toward filling this gap in the literature. Additionally, an analysis of these skills showed that the skill most prevalent within the EOC was information management, supporting the conclusions of R. W. Perry (1995) and Carter (2008). However, extending these authors’ work, the results also showed that national and international correspondence and leadership were skills required by multiple teams within the operation; and not just by those in positions of seniority as would be expected (such as Current Operations).
The labour pool and overlapping networks In accordance with swift trust theory, the results showed that the labour pool for the EOC was small; comprising of only seven organisations. Furthermore, the results showed that the respondents within the operation formed an overlapping network; with varying degrees of organisational interconnectivity for each person. As Meyerson et al. (1996) argue, a small number of overlapping organisations involved within the EOC increases the importance of swift trust; given the higher risk of reputational damage from untrustworthy behaviour being disseminated throughout the network. Regarding this overlapping network, it was concluded that the more respondents from an organisation that were included in the response, the more
166 likely that an interaction within the EOC would involve that organisation. Secondly, the degree of interconnectivity for an individual’s network was a function of their role in the EOC.
This case study also showed that the core team of an EOC can be made up of a select number of people from only a few organisations. This refines the findings of Waugh and Streib (2006, p. 2006), who argue that, “Disaster operations, particularly large operations, frequently involve a great many organisational and individual respondents”. And although there were multiple external agencies, government departments, and national service providers working with the EOC, the results showed that the operational and coordinating roles in the EOC could be undertaken by a smaller number of respondents, supported by the wider response network.
These findings also extend the work of Meyerson et al. (1996) and provide further clarity concerning swift trust theory and its conception of network interconnectivity. Swift trust theory focusses on how the interconnectivity of the entire network of respondents impacts trust formation. However, it does not focus on this at an individual level. This is important to consider, as the results showed that some respondents in the EOC had higher degrees of organisational interconnectivity than their colleagues. This meant that these personnel had more reach across different organisations to share untrustworthy behaviour if it was to occur. It is recommended that further research be conducted into the significance of personal interconnectivity in swift trust formation.
4.7.4. Members that have a limited working history but the potential to work together again………………………………. Contrary to what swift trust theory may predict, the case study found that 64% of relationships within the EOC were pre-existing (to some degree). The results showed that this was due to the composition of the EOC; with surge staff bolstering the capacity of the permanent EOC staff after the declared state of emergency. Consequently, it was found that an EOC is unique to other temporary systems in that it is not completely ‘new’, but rather a combination of an established team with additional staff incorporated into the EOC for the specific purpose of helping when a disaster occurs. These findings refine the research of Goodman (1981) and Meyerson et al. (1996) by showing that, in a disaster management context, the temporary system of an EOC will most likely be created around the pre-existing permanent EOC staff. Additionally, this work compliments the research of Kendra and Wachtendorf (2003), Waugh and Streib (2006), and Kapucu (2008), who all argue that the effectiveness of a national response, and its communication and collaboration, is better when those within the EOC have pre-existing relationships with one another.
167 Every respondent also said that they expected to work with their colleagues again in the future in one of three ways: through another operation; through the EOC and the work conducted outside of a national response; or through external training exercises and disaster management correspondence with government. These reasons also resonated in the literature. Chokshi, Seyed, Marinho Rodrigues, and Maurer (2014) stated that training exercises and simulated responses are a way in which disaster management personnel continue to relate to one another. Likewise, R. W. Perry (2004) argued that simulated responses to natural disasters are one way of intentionally building a working relationship between those who would typically be involved in the EOC.
Furthermore, in the Pacific, Fiji is the primary nation where many NGOs, INGOs, and other humanitarian service providers are located and operate from. As Roberts, Ongaro, Massey, Holzer, and Wayenberg (2011) claim, this approach of co-location enables these disaster management personnel to identify common needs, rehearse responses, and plan coordinated action. Most importantly, co-location enables relationships to be established prior to a natural disaster and helps, as Simo and Bies (2007) state, intersectoral collaboration to occur. Therefore, this research found that the field of disaster management actually relies upon on-going relationships. This seems to place the results at odds with swift trust theory’s criterion of a temporary system (that is, that personnel have a limited work history (Goodman & Goodman, 1976; Meyerson et al., 1996)). However, disasters will always be subject to personnel converging (Waugh & Streib, 2006) on the EOC to offer their expertise, and the incorporation of surge staff. Both of these factors result in the need for new relationships to form, and the need for swift trust to develop. Furthermore, as Meyerson et al. (1996) argue, the expectation of the respondents working together again is important in a temporary system; in a disaster management context, this means that those within an EOC are less likely to expect untrustworthy behaviour from their colleagues, which in turn, strengthens the need for swift trust.
4.7.5. Assembled by the contractor The case study revealed that the Director of the operation (EOC 17) was the contractor for the temporary system. There were five different ways by which the respondents identified the Director as the contractor. These included:
reporting to the Director, even if they were from an external organisation;
describing the role of the Director as the coordinator of the EOC;
168 keeping the Director informed of what was happening in the operation;
seeking strategic guidance from the Director; and,
getting decisions approved.
These results complimented the work of Waugh and Streib (2006) who stated that the head of the EOC (Director) was the coordinator and facilitator of the emergency operation, who would provide strategic guidance. However, applying Meyerson et al’s (1996) theory, the Director should not only lead the EOC, but also play a role in assembling its members. This is important because a member’s inclusion in the EOC by the Director would, “imply selectivity…predicated on sensible and more or less conscious criteria” (Meyerson et al., 1996, p. 185). It is this selectivity which acts as a substitute for interpersonal history between the respondents when forming swift trust. However, the results showed that the Director did not assemble the respondents for the EOC. Rather, the data showed that some temporary systems, such as an EOC, do not have the opportunity to be created using personnel assembled by the contractor. Although the Director may potentially mobilise and organise the EOC (Boin & 't Hart, 2010), there is little time for them to recruit a response team once a state of emergency is declared. This concept is supported by the wider literature, which highlights the many different stakeholders who are included in an operation but not recruited by the head of the EOC (Curnin & Owen, 2013; Kendra & Wachtendorf, 2003; R. W. Perry, 1995, 2003; E. L. Quarantelli, 1988). Consequently, although swift trust emphasises the role of the contractor in selecting the members of a temporary system, it does not consider the unique case of an EOC under the time sensitive operational pressures of a national emergency. As such, given the way an EOC is formed, there is less chance that the Director deliberately chooses who is to be included in the response. In this research, this criterion for a temporary system must ultimately be excluded as a contributing factor in the formation of swift trust.
The EOC as a temporary system The analysis and discussion of the results revealed that, in accordance with the criteria set by Meyerson et al. (1996), the EOC qualifies as a temporary system. Furthermore, the research applies and extends the literature on temporary systems to the field of disaster management, making several unique contributions:
Formed to address a task that is non-routine and requires interdependence – The EOC was specifically formed to address the non-routine task of responding to TC Winston. This task required both individual and team interdependence. Furthermore, this
169 research demonstrated that the seniority and expertise of a respondent within the EOC impacted their degree of interdependence; with a more senior respondent generally having to form more trusting relationships. Consequently, it is recommended that additional research is undertaken to understand this concept further, including whether there is a correlation between the formation of swift trust and the quantity of trusting relationships a person must make.
A task that is defined by clear and specific goals and has a time limit on its duration – The response was defined by clear and specific goals; which supported the requirements of a temporary system proposed by Meyerson et al. (1996) and Goodman and Goodman (1976). Also, as is essential to a temporary system, the EOC had a limited duration. This was important so the EOC could be separated out as a time-bound stage within the cyclic nature of disaster management; due to the interdependence of the four stages of mitigation, preparedness, response and recovery.
Diversely skilled people drawn from a labour pool with overlapping networks – The EOC was characterised by respondents who held diverse roles defined by their varied expertise. Additionally, the members of the EOC were drawn from a small labour pool which formed an overlapping organisational network. Particularly, this case study presented a unique contribution to the disaster management literature by providing a list of the roles for respondents involved in an EOC, as well by outlining the respective diverse skills and expertise that were required to fulfil these roles.
Members that have a limited working history but the potential to work together again – Contrary to other temporary systems, disaster management can have members who have worked together previously; as was the case for the EOC. This is due to there being multiple ways in which future interactions can occur for the respondents, other than just another operation (such as training exercises or pre-disaster work on preparation and mitigation). Consequently, it was found that an EOC is unique to other temporary systems in that it is not completely ‘new’; but rather a combination of an established team with additional surge staff incorporated for the specific purpose of helping in a response. However, there were multiple respondents who had not worked together prior to the TC Winston operation, and all the respondents also all expected to work together again in the future.
170 Assembled by the contractor – Although only some of the respondents within the EOC were specifically asked into the EOC by the Director, most were not. However, the Director did lead, coordinate and guide the response.
Although a temporary system, the results showed that an EOC can also be characterised by several unique attributes that reflect the nature of disaster management. These include:
Even though the task of the response may be defined by clear and specific goals at a macro-level, the day-to-day planning and logistics of implementing a national response can be very unclear and difficult to implement. This insight should be included within swift trust theory to provides further nuance to the characteristics of an EOC as a temporary system. The labour pool of an EOC will generally be comprised of government personnel from the affected nation. It is important to distinguish between the labour pool for the EOC, as a temporary system that is responsible for national coordination, and the preconceived notion that the labour pool for disaster management can be characterised by international respondents and NGOs. In disaster management, contrary to swift trust theory, it is ideal if respondents have a working history. Therefore, when considering the EOC as a temporary system, it is important to recognise that it is actually better if members have worked together before, and that there may only be a small number of personnel who are ‘new’ to the EOC. The role that the contractor plays in an EOC is less prescribed than that proposed by swift trust theory. The results showed that in a disaster management context, the head of the EOC, as the contractor, may not be involved in the EOC team’s formation. However, they do focus on the delivery of the temporary system’s goals; in this case, the national operation.
It is important to emphasise these attributes within this unique temporary system in order to show that the specific characteristics of any temporary system may mean that complete compliance with the theory may not be attainable. Simply, the example of the EOC also demonstrates that the characteristics of a temporary system may not be bound by only those included in the theory. It is recommended that further research be undertaken to determine the specific temporary system criteria for an EOC; to compliment those proposed by swift trust theory.
171
ANALYSIS OF RESULTS RESEARCH PROPOSITION 2
This chapter explores research proposition two, to determine whether the trust formed by the disaster management personnel within the EOC could be modelled by swift trust. Firstly, it provides an analysis of the results against each of the six swift trust criteria in accordance with Meyerson et al’s (1996) theory. It then discusses the implications of these finding in terms of swift trust theory, disaster management and the wider trust literature. Finally, it explains the applicability of swift trust to the EOC as a case study, by showing that trust in the EOC, although formed within the time constraints of the operation, was not theoretically swift trust in accordance with the theory.
Research Proposition 2: Swift trust in the EOC
Having determined that the EOC is a temporary system against Meyerson et al’s (1996) criteria, the second proposition of this research is to understand whether the trust exhibited by the disaster management personnel within the EOC, as a temporary system, could be modelled by the theory of swift trust. The following section evaluates each of the theory’s six trust criteria and their applicability to the EOC. For clarity, each trust criterion of swift trust is explained again.
1. Reputation – As members of the temporary system are drawn from a small labour pool and having overlapping networks, the chances are high that the members of this network will work together in the future. In the context of the EOC, this would suggest that a respondent will consider their reputation when deciding how to behave when in the EOC. Similarly, a respondent will also expect their colleagues to consider their own
172 reputations when deciding how to behave. As such, members of the EOC do not expect their colleagues to undermine them; which allows swift trust to develop.
2. Role – Swift trust develops faster when respondents interact around roles as opposed to personalities. However, if a colleague’s role is not clear, or they do not act in accordance with their role, a respondent may be hesitant to trust them; slowing the development of swift trust.
3. Categorical assumptions - As temporary systems are under time pressures, members rely more on categorical assumptions as opposed to evidence-based information when determining who to trust. This is because the time pressures imposed on the temporary system means that the respondents do not have adequate time to accurately determine whether their colleagues are trustworthy based on evidence. Examples of categorical assumptions include; their colleague’s role, culture, work industry norms, identity stereotypes such as appearance, etc. Examples of evidence-based information include competency, personality, work ethic, etc. As such, swift trust is characterised by a reliance on categorical assumptions which prioritises speed over accuracy when developing trust.
4. Confirmation of assumptions – As respondents rely on categorical assumptions when determining who to trust, they will look for their colleagues to behave in accordance with the assumptions that they have made. Additionally, the respondent may behave in a certain way so as to affirm the assumptions they have made about a colleague. However, if a colleague does not act in accordance with the assumptions that the respondent has made, then swift trust can be weakened.
5. Moderate interdependence – Swift trust is more likely to form when interdependence between members of the temporary system is moderate; as opposed to high. High interdependence means that a respondent is more vulnerable to untrustworthy behaviour by a colleague. Therefore, there is too much at risk for the respondent, and they subsequently require more than categorical assumptions to develop swift trust.
6. The contractor – Swift trust in temporary systems is also a function of the leader who brings the members together, called the contractor. The contractor’s reputation and the expectation of them not undermining the respondent may be all that is necessary to create an environment which allows swift trust to form, independent of the information about other members of the network.
173 Swift Trust Criterion #1 - Reputation Swift trust theory argues that the respondents within the EOC would not want to act in an untrustworthy manner. This is because they want to maintain their reputation, given the chance they will work with their colleagues again in the future. This enables swift trust to develop. As discussed in detail in Section 4.5.2, all the respondents within the EOC expected to work with their colleagues again in some capacity. As such, this section addresses swift trust criterion #1 by considering whether the respondents considered their reputations when working in the EOC. It then discusses breaches of trust within the EOC and the effects these had on some respondents’ reputations. Finally, it extends swift trust theory by providing reasons (other than reputation) as to why a respondent might not expect untrustworthy behaviour from a colleague.
5.1.1. Professional and personal reputations To understand the importance of reputation, each of the respondents was asked whether they, ‘considered their own reputation when deciding how to behave within the operation’. Seventeen of the 18 respondents said that they thought about their reputation when deciding how to behave whilst working in the EOC. However, contrary to swift trust theory, an analysis of the interview data showed that reputation did not only mean ‘professional reputation’. Rather, some respondents’ answers showed that they also considered their ‘personal reputation’.
Firstly, the following data considers the respondents who were concerned foremost about their professional reputation. A respondent’s answers were considered to be focussed on their professional reputation if they said they thought about their behaviour when working in the operation, with their colleagues, or in the future in another capacity.
“[I consider my reputation because] how can you trust a person when, you know, when working together with him or her in another operation when he cannot be trusted in one, so how can you trust him on another.” (EOC 7)
“…it’s because of the reputation that I have from the first 30 days of operation…Because I was able to be here on time, and was able to secure the directive given by the Director…” (EOC 3)
“Yes, yeh I would [consider my reputation]. For all of us in that room [Current Operations], we have reputations that we have to protect; especially for this work.” (EOC 11)
174 “[I consider my reputation] because the 30 days, you’re going to be working together…[and] I’ll have to work with [my colleagues] eventually and I would rely on them [in the future].” (EOC 12)
“My reputation, well certainly…for me the most important thing was that [my colleagues in the EOC] trusted me. And I had to earn that trust and I had to deliver on that trust…And it is the same in this case. I can’t be of use to them if they don’t trust me.” (EOC 13)
The data also showed that for EOC 2, as an international respondent, maintaining a professional reputation was significant because they were representing an international organisation. Uniquely, EOC 2 was the only international participant to state the importance of maintaining a professional reputation because of their representation of an international employer.
“Because I’m hired by DFAT, so you sit here, so in theory I represent the Australian government. You know, and I sit in the Fijian Ministry, so I have to [consider my reputation].” (EOC 2)
Uniquely, only one respondent considered the fact that working in the EOC meant their professional reputation was tied to the capabilities of their colleagues. As the authors of swift trust theory argue, “to understand swift trust in temporary systems is to appreciate the fact that relative strangers are uncertain caretakers of one’s goods” (Meyerson et al., 1996, p. 177). That is, ‘one’s goods’ can also extend to include one’s reputation.
“…yeh my reputation is at stake with your actions, so to speak…I mean, I’ll trust you because I know that if you screw up, my reputation is also screwing up.” (EOC 16)
Conversely, three respondents commented on maintaining their personal reputation (as well as their professional reputation) when working in the EOC. Answers were coded as related to personal reputation if the respondent commented on their personality or trying to uphold their relationships with their colleagues; as opposed to commenting about their work or role.
“…for me personally, [I consider my reputation because] I would have to work with these people outside the disaster, so I would try to keep that relationship good, I would make sure that, I don’t piss any of them off…” (EOC 12)
“Yeh sometimes you do [consider your reputation]. Yeh, I probably do. Yeh, especially if I get cranky.” (EOC 10)
175 “…I’m very careful of what I have to say to someone cause if I say a certain stuff about someone…probably might hit back at you. So you need to find the proper words to say to someone.” (EOC 18)
This categorisation of reputation into professional and/or personal is an important contribution and refinement of swift trust theory, which to date, has not made this distinction.
Secondly, although 17 of the 18 respondents said they considered their own reputation when deciding how to behave in the response, it was also important to ask whether they thought that their colleagues did the same. This was necessary, because if a respondent did not think their colleagues cared about their reputations, the perceived risk of them being affected by untrustworthy behaviour increased; negatively impacting the development of swift trust. Hence, each respondent was asked, ‘Do you believe people behave in a trusting manner in the operation because their reputation is at stake?’. Sixteen of the respondents believed that their colleagues also thought of their reputation when deciding how to behave during the operation. The two respondents who disagreed were EOC 10 (an international surge staff member in charge of managing the clusters) and EOC 17 (the Director of the EOC). Their answers were respectively:
EOC 10 – that the Fijian staff are less concerned with their professional reputation, given that some can be lazy with work hours or even turn up to work drunk, “and everyone knows that. Was their reputation damaged? No, everyone knows that’s what you do.” Comparing this response to the answers previously provided by the locals about their concern for their own reputation, this comment appears inaccurate.
EOC 17 – the Director of the EOC argued that they thought their colleagues did not consider their reputations when working, because they were more focussed on delivering the response than thinking about their personal/professional image.
Overall, swift trust theory would suggest that the respondents within the EOC would not want to act in an untrustworthy manner because they want to maintain their reputation, given the chance they will work with their colleagues again in the future. The theory implies that the respondents would be most concerned with their professional reputation. Although this was true for the majority of the members of the EOC, this was not the case for them all. Consequently, the results showed two things:
1. As swift trust theory predicts, the participants were concerned about their reputation when deciding how to behave during the operation. They also felt that their colleagues considered their own reputations when deciding how to behave in the operation. These
176 two factors meant that there was less chance of a respondent expecting untrustworthy behaviour from their colleagues; increasing the chance of swift trust forming.
2. Unlike swift trust, a respondent’s consideration of their reputation in the EOC did not only mean professional reputation. Rather, for some, the term reputation was expanded to include both professional and personal reputations.
5.1.2. Untrustworthy behaviour damaging reputations The results also revealed two cases where a person within the EOC had acted in an untrustworthy manner, leading to their reputation being damaged. It was important to explore these incidents, given:
the theoretical importance of maintaining a trustworthy reputation within the EOC as a temporary system,
the possibility of working with the members of the temporary system again in the future, and,
the risk of a member’s untrustworthy reputation disseminating throughout the wider network of disaster management organisations.
The first example was for EOC 18 (a local respondent working in the procurement team) not trusting their manager EOC 15 (another local respondent who led the procurement team). EOC 18 stated that their distrust in their manager was because their manager did not give them credit for the work that they were doing, and had released information that had reflected poorly on EOC 18’s reputation.
“…I’d prefer if, you know in the stuff of [the EOC], let it stay here, don’t leak information out if there is a problem in this ah…[and] I know more than what [my manager – EOC 15] knows. But [they are] getting paid for it. [They are] getting the credit...You have to prove yourself. If the report comes, you have to write your name. If not, people won't notice you…I don’t trust [them]." (EOC 18)
Furthermore, EOC 18 was required to continue working closely with their manager because they were the only two people within the procurement team in the EOC. Consequently, the ongoing working relationship between EOC 18 and EOC 15 meant that EOC 18 had to keep their distrust in EOC 15 confidential. This example showed that if a respondent experienced untrustworthy behaviour from their colleagues during the operation, that could damage the perpetrator’s reputation and lead to swift distrust in that person.
177 The second example was a more substantial case, where the EOC lost trust in the United Nations Office for the Coordination of Humanitarian Affairs (OCHA). To understand why this happened, it is necessary to first explain how the incident occurred. From the start of the response, OCHA were providing support to the EOC through information management and dissemination. However, the EOC had set a mandate that any information being released to the public was first to be approved by the Director or the Current Operations team.
A member of OCHA was working alongside the military forces within Future Operations. They were jointly producing data that was to be published as part of the response. However, the OCHA member took this information and published it on their United Nations website without gaining permission from the Director of the EOC or Current Operations. EOC 8 (who was working in Future Operations at the time of the incident) explained;
“…we found that…[the data] was published on the UN OCHA website, and some of the senior officers flipped because we had worked the whole night trying to get it done and all that kind of stuff. And it actually wasn’t referenced work, which we felt…[was] very unprofessional from OCHA to actually publish work that was done in a multidisciplinary team and solely, largely led by the Fijian military forces...So, I mean, that, when that happened…[the perpetrator was] told not to come back into the [EOC]...and for [the EOC] not to give out information to OCHA". (EOC 8)
The OCHA employee not only failed to follow the procedures for releasing information set by the EOC, but they also did not credit the joint effort made with the Fijian Government. Consequently, the EOC distrusted OCHA. EOC 16, a member of Current Operations, explained:
“[The EOC] felt it was information being leaked. And from the Director’s point of view, he felt he should have been consulted first before that information was released. Um, so that’s where…I think, yeh, like the trust then just went downhill.” (EOC 16)
This distrust was significant enough that the primary OCHA liaison to the EOC, post the incident, reflected that they could not overcome the problems that existed due to this infringement.
"[We as OCHA] were trying to share our information...but I would say, I don't think the [EOC] felt they needed much from us...in this situation, because of trust issues, there was a real decision, conscious or otherwise, it was conscious, but it wasn’t necessarily an official decision, but there was a pattern of behaviour that was to keep OCHA at bay.” (EOC 13)
178 The results also showed that this breach of trust by OCHA had ramifications on another UN agency, as well as the Australian Defence Force, and their involvement within the response. Firstly, the United Nations Disaster Assessment and Coordination (UNDAC) team were recommended by a senior member of Current Operations to become involved within the response. However, their affiliation to the United Nations (and, by extension, OCHA), meant that they were not brought into the operation.
“…the problem was in that first week, there was already issues between the [EOC] and OCHA around that information sensitivity issue. So when the UNDACs team arrived, they got embroiled in all of that, and it’s kind of sad they didn’t [get involved]…” (EOC 16)
Secondly, with hesitations being shown towards the international community concerning information security after the breach of trust by OCHA, the Australian Defence Force were also isolated by the EOC. Particularly, the military officers within Future Operations informed their local colleagues to start limiting the information released to the Australian Defence Force.
“Ok, (voice lowered for privacy), like I had an incident, um, when I was working [with Senior Australian Defence Force Member] for the first two weeks. So, um, I had been doing landing zones and also identifying zones where ADF [Australian Defence Force] relief was being sent out to. At some point in time, not some point in time, at a point in time that work with the ADF, there was um, ah, some, I wouldn’t say advice, but some orders not to give out too much information.” (EOC 8)
The reputational damage to OCHA by acting in an untrustworthy manner led the EOC to exhibit swift distrust toward this organisation and subsequently limit their involvement in the response, and that of another UN organisation (UNDAC) and the Australian Defence Force (ADF).
Conversely, although OCHA (as an organisation) was not trusted by the EOC, the data showed that there was still the potential for OCHA employees to regain the trust of respondents within the operation. EOC 13 was stationed to the EOC as the OCHA liaison officer after the breach of trust by OCHA had occurred. They found that, although institutionally OCHA was not trusted, they could personally build trust with their colleagues; even though they were affiliated with a distrusted organisation.
“I found that when I came in, I did develop good relationships and good trust with the staff in the [EOC] at a personal level…But for some reason, [at] the institutional level there’s this tension…So, even from my conversations with other people, national staff in the [EOC], often sort of said, ‘there’s issues with sharing with OCHA, but with you we
179 trust and we understand, like we don’t mind sharing information with you’. And so, I was very conscious from the word ‘go’ that I had to earn that trust and…If they trusted me with something, if they said, ‘we don’t want you to share this’, then I had to respect that.” (EOC 13)
This data shows that even though an incident can lead an organisation to be distrusted, trust can be built again by an individual of that organisation; provided they can develop a trustworthy reputation from their colleagues.
Finally, a comparative analysis of these two examples shows that the more people who are affected by an individual’s untrustworthy behaviour (only EOC 18 in the first case compared against (particularly) the senior leadership of the EOC in the second case), the more widespread the distrust in that individual. The distrust of EOC 18 in their manager was isolated to between those two people. However, the breach of trust by OCHA against the EOC impacted most respondents within the EOC. Regardless, both examples validate that maintaining a trustworthy reputation is important for swift trust to develop.
5.1.3. Additional reasons for not expecting untrustworthy behaviour Although swift trust emphasises reputation as a key reason why respondents would not expect untrusting behaviour from their colleagues, some respondents also offered additional reasons.
Multiple respondents stated that being a part of the Fijian civil service meant that they could trust their colleagues. A couple of respondents specifically mentioned that people within the civil service genuinely cared about Fiji and were passionate about helping in the response. Similarly, respondents suggested that being a part of the civil service meant that a person had to abide by the government code of conduct, and therefore were bound to act in a trustworthy manner.
“There is just this thing about the civil service where you just trust the other public servant. I don’t know how to explain it. There is this general trust in the civil service.” (EOC 1)
“…people who work in government, they are binded by what is called the ethics code of conduct and code of ethics; trust is one of them.” (EOC 7)
“I think there is a genuine, for most people, I mean public servants are here to serve the public and for the ones that work hard, they do it because, that is why they signed up.” (EOC 2)
180 “These guys out here [in the EOC], you know what, I think they do it because they are passionate about the job.” (EOC 8)
“…over here, there are people engaged in the civil service, there is a code of conduct. There is a code of conduct that everybody needs to abide [by]…and the values also. So you need to be seen to wear that every day, ah, when you are in the office and outside the office…If you don’t, ah, your job is at stake.” (EOC 9)
Uniquely, EOC 9 (as a permanent, senior staff member in the EOC) also stated that people were contractually obliged to behave in a trusting manner: they did not expect untrustworthy behaviour from colleagues as it would be a breach of their contract. The results showed that being a member of the civil service and expressing a real desire to help during the operation provided additional confidence to some respondents in being able to trust their colleagues. Importantly, this sentiment was not only expressed by the permanent staff in the operation, but also by some external surge staff too (EOC 2 and EOC 8).
In summary, as shown in Section 4.4.2, due to a small labour pool and overlapping networks, all the respondents within the EOC expected to work with their colleagues again in the future. In accordance with swift trust theory, the respondents considered their reputation (professional and/or personal) when deciding how to behave whilst in the EOC. Similarly, the respondents also believed that their colleagues considered their own reputations when deciding how to behave. The data also suggests that the members of the EOC did not expect untrustworthy behaviour from their colleagues given the damage this would have on the perpetrator’s reputation. All these factors (a concern for reputation, a small labour pool, and overlapping networks) helped create an environment where swift trust could develop. The results also showed that there were two recorded breaches of trust, with the reputational damaged suffered by the perpetrators being related to the number of people impacted by the untrustworthy behaviour. Lastly, although protecting one’s reputation was a significant contributor to the development of swift trust, some respondents also felt that personal motivations and being bound to the civil service provided further reasons as to not expect untrustworthy behaviour from their colleagues.
Swift Trust Criterion #2 – Interactions around role Swift trust theory would predict that swift trust will develop faster in the EOC if a respondent’s interactions are focussed on their colleague’s role in the operation as opposed to their personality. This is because, under the time constraints of the response, a role is easily defined
181 and less complex than a colleague’s personality; and thus provides some early evidence of whether a colleague can be trusted. Therefore, this section will first explore whether the interactions within the EOC were indeed based around the respondents’ roles. It will then describe the impact of role clarity and role behaviour on the formation of trust in the EOC. Finally, it will address the assumption within swift trust theory that role should be a swift trust criterion, and subsequently explain the importance of personality as a trust criterion for the EOC.
5.2.1. Interactions within the EOC based on a colleague’s role In order to learn what each of the respondents interacted with their colleagues about, it was first necessary to know which colleagues a respondent was interacting with as part of the operation. Using the Network Mapping Worksheet, it was possible to determine which colleagues a respondent was talking to, and how often these interactions occurred. The interview then asked about these connections and what the respondent specifically spoke to their colleagues about. Nearly every single respondent said they primarily spoke to their colleagues about the operation and their work. Explicitly, when asked ‘how many of their interactions with their colleagues were work related?’, EOC 2, EOC 8, EOC 10, and EOC 15 simply replied, “all”. Other respondents’ answers included:
“It was all about work.” (EOC 6)
“[All conversations were] work related. All work related.” (EOC 3)
“The majority of [my interactions were] work related.” (EOC 14)
Some respondents also specified that there was little time to speak about personal or social matters; that is, personality-based interactions.
“I have a lot to share with [my colleagues] on a personal level. But during phase one, I had hardly any time to do that…Even over meals and a cup of coffee, I’d be having it next to my laptop or something.” (EOC 11)
“Yeh, I mean, little side chats…[but] it’s still work. [Personal conversations] would be less than five percent of the conversations.” (EOC 13)
Specifically, for those within Current Operations, role-based interactions were not just limited to when they were working within the EOC. These respondents had an online chat group operating so that they could continue working on the response outside work hours. However, these interactions were primarily around work roles.
182 “…we formed a Viba chat group. We called it ‘Liaison Officer’ chat group. And that was even when we were at home and we needed information from each other about ‘where this vessel was going, and whether he would be here on time’.” (EOC 11)
The results showed that the majority of the interactions between the respondents within the EOC revolved around the operation and each other’s role in the response. This is significant because swift trust theory argues that, “relating in a temporary system is among roles as much as people” (Meyerson et al., 1996, p. 173). Therefore, according to swift trust theory, the respondents should have been able to obtain some initial evidence of whether a colleague could be trusted on the basis of these role based interactions. The following section details a recorded example of role-based interactions leading to trust formation between members within the Current Operations team.
Current Operations and interactions around role The case study showed that particularly for the members of Current Operations, role-based interactions led to trust. Although some of the team knew each other prior to the response, there was one member who was relatively new to this cell (EOC 14). Each member of the Current Operations team had a specific role, either: preparing the daily briefing for the Prime Minister, conducting GIS (mapping), liaising with the other parts of the EOC, or vetting the emergency response data. Furthermore, all four members reported straight to the Director, had access to sensitive information within the EOC, were positioned within the same office, and physically worked opposite each other. This combination of factors, in addition to their roles in leadership and subsequent responsibilities, meant that each member had to rely on the others to do their job, and that trust could form around their role-based interactions. EOC 11 provided an example of this trust forming quickly on the basis of role when working with their GIS specialist colleague (EOC 14), with whom they had not worked prior to the response.
"When I started doing, preparing the presentations to present before the council, there were lots of things that I felt was missing and which could have been done by [EOC 14] especially for mapping using GIS and all this. [They are ] the expert on that. So, I think I had, I had to trust a lot, ‘If I give you this information, do you think you could give me something’...So the first time I met [them], I don’t think I really actually talked to [EOC 14], it was a ‘hi bye’ sort of thing...The second time I met [EOC 14], [they were] in our cell. So I had to work with [them]. And then I asked [them] if [they] could help and [they] said ‘yeh, no worries’. And within a few minutes, [they] came back with whatever I had
183 requested for. And it actually helped me a lot in my presentation. So…we had to have that [trust]." (EOC 11)
Apart from showing a specific example of trust, EOC 11’s comment highlighted a general concern for Current Operations: the protection of operational information and data. The respondents in Current Operations were reluctant to share information without knowing who was going to use it and what it would be used for. This was particularly true for international partners after the breach of trust by OCHA. Example statements of this sentiment from members of Current Operations included:
“…it’s very important that you know whoever you’re talking to and sharing the information and that you don’t know his background or organisation, you need to be cautious of the information that you use.” (EOC 7)
“…I’d like to have his profile just to ensure to keep me…rather than just walk and say ‘hi’ and then we started to chat, ‘who are you, what are you, where are from’. Just give me your background so then I know you more before we continue.” (EOC 17)
However, in the example previously mentioned, even though EOC 11 did not know EOC 14, their colleague’s inclusion within Current Operations was enough to overcome their initial, systemic hesitancy to share information. This data leads to two conclusions. Firstly, as with the wider EOC, trust can form around role-based interactions. Secondly, this example highlights that ‘group inclusion’ can be a contributing factor in the formation of trust. As swift trust theory argues, “membership in a social identity group” (Meyerson et al., 1996, p. 174) can be sufficient for a respondent to develop swift trust without knowing their colleagues personally. The data suggests that ‘group inclusion’ was a factor for trust formation for some of the members of Current Operations, alongside role-based interactions.
5.2.2. Colleague’s having clear roles Swift trust theory also argues that role clarity is important to the development of swift trust. That is, if a colleague’s role is not clear to the respondent, they may be hesitant toward them; slowing the development of swift trust. This section will explore a lack of role clarity within the EOC, and the subsequent implications of this scenario for trust formation. Secondly, it will show that role clarity was considered important primarily by those in leadership and senior positions within the EOC.
184 A lack of role clarity in the EOC As part of the interview, each respondent was asked whether they knew what their colleagues did in the operation. The results suggested that there was a general lack of role clarity during the early stages of the response, particularly due to the incorporation of surge staff into the response. Furthermore, this lack of clarity was primarily experienced by the surge staff, who were new to the EOC and, as expected, were less familiar with any pre-existing roles within the office. However, some of the permanent EOC staff were also unsure of the roles of their colleagues in the operation.
“…you had people from all agencies, all different organisations. It was important for the team from [the EOC] to understand why we were all here. Because we were all here for just one purpose, which is to provide relief and response and all that. The team had to understand why we were here; which was something that really should have happened before the whole operation.” (EOC 11)
“Because the first one or two weeks of the operation, you are new to this office and you are new to the people you are working with, so you need to understand them, understand how they work.” (EOC 3)
“I came and I was told to go to the operation room [National Emergency Operations Centre]. Even in the operation room I saw, I see people sitting around. I didn’t know who, what was their role like in there.” (EOC 5)
“…[roles are] not clear and it needs to be made clear…” (EOC 2)
“…there are a lot of teams out there and I’m not sure what exactly they do.” (EOC 6)
“…there are a few who are just new to the system, to what is happening here…we need everybody here to understand what everybody’s doing.” (EOC 9)
“You came in here the early days, and from one day to the next, it was like little units sprouting up… Yeh, in the first week it was chockers. Cause no-one knew who was who. And everyone wanted to know what was going on.” (EOC 10)
The national disaster management documentation did not offer any further clarity around the roles of each EOC member during the response. Firstly, the National Disaster Management Plan, written to guide a national operation, only provided a high-level organisation chart outlining the structure of the national response, without detailing the specific roles required during an operation. Secondly, the EOC organisation chart (Appendix A) only showed the roles of
185 personnel for times outside of a national emergency. It did not provide descriptions of the reporting structure within the EOC, nor each respondent’s role, during a declared state of emergency. Lastly, the staff movement board, which was located in the EOC and designed to show who was working in the EOC each day (Appendix A):
was incomplete,
did not include all the respondents (including key personnel such as those in Current Operations), and,
seemed to be neglected (it was observed that the board was not updated to reflect real-time changes throughout the operation).
Furthermore, the staff movement board did not provide clear descriptions of any of the respondent’s roles. Rather, it only assigned titles to some respondents in accordance with the organisational areas of the EOC outside an emergency.
As stated, swift trust theory argues that role clarity is important for the development of swift trust. Within the EOC however, there was;
a lack of formal role clarity during the early stages of the response, and,
no documentation or up-to-date system describing the roles of the respondents within the EOC.
However, as shown in Section 5.2.1 the majority of respondents said that all their interactions within the EOC were based on the work they were doing with their colleagues. Consequently, although role clarity may have been lacking within the early stages of the operation, the data showed that the respondents would have had the opportunity to learn the roles of their colleagues, given they were working with them to deliver the requirements of the response. This is significant to recognise for two reasons. Firstly, although swift trust states that role clarity is important for the development of swift trust, it does not provide any clarification for this argument. The results of this research showed that the members of a temporary system may not need to know the roles of all the members of the system; rather, only those of the colleagues they are working alongside. Secondly, within a disaster management context, although formal role clarity may not exist, the members can still experientially learn their colleagues’ roles through working together.
186 The importance of role clarity for senior members of the EOC Finally, although swift trust theory values role clarity, the respondents expressed that they were generally willing to trust their colleagues from the outset; regardless of whether they knew their role or not. Rather than knowing a colleague’s role clearly, the respondents said they:
Were more concerned with their colleagues doing the tasks necessary for the response:
“I can be so worried about so many thing[s]…[but] I wouldn't really have to know [my colleagues]…I just had to trust my colleagues to do their part and I had to focus on mine…For the sake of the national exercise, you keep on doing what you have to do and trust people.” (EOC 1)
"…because of the nature of the work, we need to trust them that they can get the job done." (EOC 4)
“…[trust] just happens in the first hour we are working…when you work with them, you tell them, because you don’t have ah, the time to develop that, because we are on eight hourly shift in the [National Emergency Operations Centre], just need to get some idea of what they can do.” (EOC 6)
“…I don’t think we had enough time to get to know each other way more than we already, than the time we actually spent together…But I think for me, it was…because of our common goal of trying to…better respond…I think that’s why things actually worked out.” (EOC 8)
Started from a position of trust:
“I’m sure all the people sent to the operations here are all trustworthy people. So whenever the person comes in we just need to get along…” (EOC 3)
“I will generally walk up to someone and give them the benefit of the doubt from the word 'go'. And if they give me certain information or ask me for certain information, I will extend a certain level of trust right from the word 'go'.” (EOC 13)
“…my general standpoint is I start from a position of trust… I think generally you’d have to start, I mean a general trust.” (EOC 10)
“You know, I guess, you apply trust, and you task people differently.” (EOC 2)
187 “[I extend trust] pretty quickly...” (EOC 14)
Wanted to obtain some foundational information about their colleagues rather than only know their role:
“The only thing for me is to get to know the person’s name…where they’re from and why they’re there. And that’s it.” (EOC 11)
“I have to get the, background of the person from a previous colleague who they have worked together. So it can give me an indication that he can do the job...” (EOC 15)
Conversely, four of the respondents who were in positions of senior leadership were slower to extend trust to their new colleagues because of their general concern for protecting the EOC or, if they were working in Current Operations, due to information security and ensuring data was not misused. As such, these respondents did want to know a colleague’s role prior to extending trust. The results showed that a lack of role clarity would impact how trust formed for these respondents.
“I'd like to know the person in more detail, his or her background, and that's what I do for everyone who comes inside. I ask, I ask for like a profile, once they give that to me with their experience, then we sit down.” (EOC 17)
“If someone is new to the operation I have not met before, I will be very cautious about information that I share with him or her, in case, because I don't want that information to be misused or misinterpreted.” (EOC 7)
"I just don't trust anyone in the first time that I meet people because nowadays, there are a lot of people who can go around pretending to be someone." (EOC 9)
“During the disaster phase, it would probably be maybe a week or two weeks [before I trust them]. Just to see how they work, the conversations I have with them, see, you know, just to gauge...” (EOC 12)
(Field note) “During the extension to the emergency period (approximately two months after TC Winston struck Fiji), an international colleague (Australian, middle-aged male) was deployed to the EOC from the humanitarian organisation RedR. From speaking to him, he had been in many operations before and was experienced. He was tasked with writing a report for the response, and was required to liaise with Current Operations in
188 order to complete this task. However, through conversations and by observation, it became apparent that he was finding it difficult to secure a meeting with Current Operations for guidance on how the EOC wanted the report to be written. This may simply have been due to Current Operations and the Director being busy. However, given the response had slowed down by this stage (and that I was able to get meetings with the Director despite my peripheral role in the EOC), this was significant to observe as potentially another example of hesitancy in sharing information with an international partner due to an initial lack of trust (as with OCHA) in order to protect the EOC.”
Overall, the results suggested that a lack of role clarity did not affect the formation of swift trust within the EOC. The majority of respondents were willing to trust their colleagues from the outset of the operation. Only four respondents in positions of seniority were slower to trust their colleagues if they did not know their role in the operation. This was because they were concerned for the EOC and protecting how any information within the operation was used.
5.2.3. Role behaviour Swift trust theory argues that the development of swift trust for a respondent can also be negatively affected by a colleague not performing their role in the operation (role behaviour). Given that there was an initial lack of role clarity for some respondents, it would have been difficult for these respondents to assess if a colleague was in fact doing their job in the operation. However, as previously argued, over time, the respondents could have learnt their colleagues’ roles. This is significant because swift trust theory relies on both role clarity and role behaviour in the temporary system as an early indicator for deciding whether to trust a colleague. That is, both role and role behaviour are substitutes for other sources of evidence which are not initially available, but are gained over time. For many of the respondents within the EOC, because they were unsure of what their colleagues’ roles were in the operation, they could not rely on role clarity and role behaviour as a basis for trust formation.
Concerning role behaviour in the EOC, each respondent was asked how they would respond to a colleague who was working on a task that was not part of their assigned role. Eleven of the 18 respondents said that it would not affect how they interacted with that respondent in the future. These respondents were a mix of senior and junior members of the EOC, from multiple teams. They presented three main reasons why they were not concerned about a colleague behaving outside of their role:
Given the demanding requirements of a national response, members of the EOC were required to do extra tasks.
189 “I am of the understanding that everyone is working towards the three main focus areas. So if someone has to carry on another load for that, then so be it.” (EOC 1)
“…because this is an operation, we do not have all the resources that we want, so we let people assist in another [role].” (EOC 7)
“Everyone had like six hats, everyone was all over the place here.” (EOC 10)
Respondents assumed that their colleagues were multi-skilled and could take on additional roles.
“…we let people with multi-skill, they can specialise in one [role] but also assist in another.” (EOC 7)
“Maybe they have, ah, multi skilled.” (EOC 4)
Some respondents did not feel it was their prerogative to question what their colleagues were doing in the EOC and so kept to themselves.
“It is not, um, to me it is best for me to just keep to my work. If there are other peoples doing things outside their area of work, it’s the supervisor that should talk to them but not me.” (EOC 3)
“Oh no, I wouldn’t care much because I wouldn’t know if he’s given another [role]…there’s another motive for him to be doing this apart from what he’s
supposed to be doing.” (EOC 15)
“…it’s not my job to [judge my colleagues].” (EOC 10)
This data shows that, contrary to swift trust theory, role behaviour may not have played as significant a part in deciding whether a respondent would trust their colleague in the context of the EOC. The results suggest that for these respondents, the weight assigned to a colleague adhering to their role would seem to be less than what swift trust theory predicts.
Conversely, eight respondents said that a colleague doing a job outside their assigned role could negatively affect their perception of that colleague. Importantly, seven of these eight respondents were in positions of leadership or experts in the EOC, with four being from Current Operations. An analysis of these respondents’ answers revealed two main approaches they would take if a colleague was not adhering to their role in the operation:
190 To actively deal with it themselves:
“I’m one person that if I know you are supposed to be doing this and then you go ahead and also do this, I will ask you why you’re doing that.” (EOC 11)
“…they’re supposed to be doing this and they’re doing another thing…I would tell them straight.” (EOC 17)
“If they are qualified to do something else, why don’t you just push them to do that job rather than do something they are not qualified to do and do well?” (EOC 12)
“I’d like to think that if something like that did ever happen, that I’d try deal with it immediately rather than waiting, waiting for the response to be over…Just a conversation on the side.” (EOC 14)
“…if they screwed up their [additional role], it definitely would affect…I wouldn’t put them back into [their original role].” (EOC 16)
To not respond to their colleague’s behaviour but to take note of it:
“If they were supposed to be doing x but were doing y, you know, depending on y, whether they were not capable, couldn’t be bothered, or didn’t give a rats or just thought…that would be something that is locked away for next time. You just take a note…” (EOC 2)
“…there’s going to be some mistrust on that particular person because I know that they are here to do something and then they turn up to do other things…if you have people like that, ah, you know, sort of creates a lot of questions in the minds of the managers…” (EOC 9)
To fully understand the importance of role conformity to these respondents, the interview asked these participants how they would feel if their colleague performed well in their additional role; even though it was not their assigned job. Six of the eight respondents stated that if they saw the colleague perform well, this would counter their initial negative perception of them. However, only two of these six respondents said that it would actually positively affect their perception of their colleague.
Concerning the importance of role in swift trust theory, the data suggests that in an EOC, those in positions of leadership or experts are concerned with their colleagues adhering to their
191 respective roles within the operation. These results comply with what swift trust theory may predict in relation to role behaviour. Particularly, the majority of these senior EOC members felt comfortable to actively challenge their colleagues if they saw them performing outside of their role. However, as shown in Section 4.4.1, disaster management is characterised by personnel taking on additional roles and working longer hours in order to meet the requirements of the response. Therefore, the data suggests that the leaders of the EOC could actually inhibit the flexibility required to meet the demands of the operation, if they insist that the respondents focus only on their assigned role.
In summary, the results showed that for those in leadership and expert roles within the EOC, it was important that a colleague adhered to their respective role, as swift trust theory might predict. However, outside of these respondents, the remainder of the EOC were impartial to a colleague taking on an additional role; particularly if that colleague performed the additional role competently. This was because they believed either: the requirements of the response meant people had to take on more tasks; their colleagues were multi-skilled; or, it was not their place to challenge their colleague’s actions. The results show that for these respondents, the weight assigned to a colleague’s adherence to their role would seem to be less than what swift trust theory may predict.
5.2.4. The assumption of role as a swift trust criterion Swift trust theory argues that swift trust develops faster when the respondents within the EOC engage around each other’s roles as opposed to having personality-based interactions. However, there is an underlying assumption in this argument: that a respondent must be using their colleague’s role as a criterion in determining whether to trust them. However, the data showed that this was not always the case.
For the 18 respondents interviewed, each was asked what their trusting criteria were for both their local and their international colleagues. This was done to determine if there were any differences between how trust was conceptualised by, and extended toward, these two different groups of colleagues.
192 Table 5.2.a A summary of the respondents’ trust criteria applied to local and international colleagues
Trust Criteria Trust criteria applied to a Trust criteria applied to an Total Type local colleague international colleague Categorical 27 17 44
Appearance 2 3 5
Civil Servant 2 0 2
Fijian Culture 7 0 7 Parent 1 3 4 Organisation Role 11 8 19 Seniority & 4 3 7 Authority Evidence 34 28 62
Competency 12 9 21
Cooperation 4 2 6
Motivations 1 1 2
Personality 12 11 23
Work Ethic 5 5 10
Total 61 45 106
Table 5.2.a, as a high-level summary of the results, showed that:
Eleven of the 18 respondents said that they would use their local colleague’s role in the operation when deciding whether to trust them. Similarly, eight of the 18 respondents said they would use an international colleague’s role in the operation when deciding whether to trust them.
Twelve of the 18 respondents said they would use a local colleague’s personality when deciding whether to trust them. Eleven respondents said the same for an international colleague.
193 No respondent used role alone as a trust criterion. Conversely, three respondents reported they would rely on their colleague’s personality but not their role.
While swift trust theory assumes that role is a trust criterion, this data showed that a colleague’s role was not always a factor for trust formation within the EOC. Rather, there were more cases of the respondents relying on a colleague’s personality than their role when determining whether to trust them in the response. Specifically, when considering whether to trust either a local or international colleague, no respondents relied solely on role as a trust criterion. The data showed that swift trust does not always develop when a respondent’s interactions are focussed on their colleague’s role in the operation as opposed to their personality. Rather, although role may play an important part in the development of swift trust (because it acts as a substitute for more specific trusting evidence about a colleague), this was not universally the case for EOC. The data showed that there were additional factors that were also important in the development of swift trust. These factors are considered in Section 5.3, and primarily included the following:
personality,
competency,
Fijian culture,
levels of seniority and authority, and,
work ethic.
5.2.5. The importance of personality Swift trust theory argues that within the temporary system, role is a primary factor which helps trust form because there is not enough time for personalities to influence trust (given the time-sensitive task at hand). Simply, “swift trust is most likely when…interacting with roles rather than personalities…[as] swift trust is less about relating than doing” (Meyerson et al., 1996, p. 191).
Contrary to this idea, however, within the EOC, although respondents primarily had interactions based around role at the start of the operation, they naturally transitioned into more personal interactions as the temporary system progressed. Colleagues would share about their experiences of working in the EOC or the stresses induced by the environment.
194 “…because sometimes running the operation, you just get bored and frustrated sometimes, how the operations are running on the ground here, some things, you know, sometimes are not to your expectation, so sometimes we sit and discuss things that cool us down.” (EOC 9)
“…I would say ‘let’s go have lunch together’…like when we’re sitting in the office, let’s go have some cake. When it was ah...[EOC 7] had [their] birthday and they did cake in the office…” (EOC 13)
(Field note) “Within the EOC, sometimes there was afternoon tea and lunch brought out onto the central table of the office. Many respondents would gather around the table and have a conversation, stepping away from work to socialise. I was also included within these sessions, which helped build relationships with the staff.”
Similarly, the Current Operations team started doing ‘kava’ together after a couple of weeks into the operation. Kava is a traditional Fijian drink shared during a talanoa session.
“By the end of the second week, we started having kava together downstairs in the garage. So people started getting a bit more comfortable…mainly our Current Operations team. Then we slowly pulled the Future Operations people down and then the [National Emergency Operations Centre] guys come down. And then toward the third week, fourth week, then we opened the officer’s mess. So that’s were for us and Current Operations, not having to go up and down those stairs all the time, we just had a little kava session out the back here. And that’s where we brought the Future Operations people across. So we kind of call it the officer’s mess.” (EOC 16)
These examples show that over time, social interactions and a colleague’s personality became a source of evidence for a respondent in being able to develop trust with their colleagues. Furthermore, as previously described concerning Current Operations, these examples are instances of trust being influenced by ‘group inclusion’ (Meyerson et al., 1996; Popa, 2005). Having cake with the office or sharing kava were activities that helped unify those included, potentially strengthening a respondent’s social bonds and providing them with an additional source of information when determining whether to trust their colleagues. Similarly, being able to vent frustrations or simply have lunch with another colleague are further examples of ‘group inclusion’. These too provided opportunities for the respondents to interact with one another and learn of each other’s personality; which in turn provided more information when determining who to trust. Ultimately, these examples supported swift trust theory’s argument
195 that ‘group inclusion’ can be a potential contributor to the development of trust between members of the temporary system. However, most pertinently, these results are significant because they show that swift trust can be built off more than just an understanding of each other’s roles. Rather, as shown in Section 5.2.4, personality was considered important for trust by the respondents. Subsequently, this showed that the emphasis on role-based interactions adopted by swift trust theory may be inaccurate in the context of an EOC.
Swift Trust Criterion #3 – Categorical assumptions versus evidence-based information……………………………… When applied to a disaster management context, swift trust theory would suggest that the respondents within the EOC would rely upon categorical assumptions when deciding whether to trust their colleagues. This is because the time pressures imposed on the response means that the respondents do not have enough time to accurately determine whether their colleagues are trustworthy based on other, more nuanced evidence. Examples of categorical assumptions include: the institutional context of the temporary system, industry norms, roles, and even identity-based stereotypes, amongst others. Conversely, evidence-based information, gained over multiple interactions, is aligned with the formation of trust over time as the respondents are increasingly able to judge whether to trust a colleague due to their personality, work ethic, competency, motivations, etc., as the number of interactions increase.
This section will first discuss the trust criteria used by the respondents within the EOC. As this research also focuses on international personnel working in a response, it will then explain the trust criteria used by locals when deciding whether to trust their international colleagues. Finally, it will also discuss culture and the involvement of the military as contextual categorical assumptions that influenced trust formation in the EOC, particularly for the locals.
5.3.1. Trust criteria used by respondents within the EOC In order to address this swift trust criterion, each respondent was asked to describe what criteria they used when deciding to trust someone when working in the EOC during the operation. This question was asked with reference to local colleagues, and then international colleagues; in order to determine if there was a distinction in the criteria used by the respondents. The criteria that the respondents used were then grouped and classified as either a categorical assumption or evidence-based information in accordance with the distinctions made by swift trust theory. This data is presented in Table 5.3.a, which shows a summary of the trust criteria used by each respondent when deciding to trust a local or international colleague within the EOC. An analysis of the data showed that the members of the EOC relied on three primary trust criteria when
196 deciding whether to trust a colleague. These criteria were a colleague’s role, competency, and personality.
Table 5.3.a A summary of the respondents’ trust criteria applied to local and international colleagues
Trust criteria applied to a Trust criteria applied to an Trust Criteria Type Total local colleague international colleague Categorical 27 17 44
Appearance 2 3 5
Civil Servant 2 0 2
Fijian Culture 7 0 7
Parent Organisation 1 3 4
Role 11 8 19
Seniority & Authority 4 3 7
Evidence 34 28 62
Competency 12 9 21
Cooperation 4 2 6
Motivations 1 1 2
Personality 12 11 23
Work Ethic 5 5 10
Total 61 45 106
(Note: Before addressing the data, throughout their interviews, both EOC 5 and EOC 18 stated that they would take one and a half years and six months respectively before they trusted someone. In accordance with swift trust theory, these time frames are too long for swift trust to viable and as such, the trusting criteria of these two respondents are excluded from the following analysis. This has been done so as to not skew the data and provide a misrepresentation of what criteria were used by the remaining respondents when deciding who to trust.)
Role Overall, there were 19 counts on which another colleague’s role influenced the formation of trust for a respondent. An interview comment was coded against the ‘role’ trust criterion if the respondent said that trust was built through:
knowing a colleague’s job and purpose in the operation, or,
197 any interaction based around the work required for the response.
Some respondents valued understanding why a colleague was in the EOC and what their purpose was in the temporary system.
“…I have to get the background of the person from a previous colleague who they have worked together. So it can give me an indication that he can do the job...” (EOC 15)
“…we had a lot of interactions around what we were producing from [Current Operations]…[and] most of interactions with internationals are really all work related.” (EOC 14)
“…I would just need to know why they’re here, what they can and cannot do, their level of authority maybe...[and concerning internationals] I need to know [why] they’re here.” (EOC 11)
“…when you work with them…because you don’t have ah, the time to develop that, because we are on eight hourly shift…just need to get some idea of what they can do.” (EOC 6)
Although there were 19 counts of role being used as a trust criterion, an analysis of these results highlighted multiple significant factors. Foremost, swift trust theory predicts (as per swift trust criterion #2) that a colleague’s role is used when determining whether they are trustworthy, because there is limited time to obtain evidence through interactions. That is, the categorical assumption of a colleague’s role acts as a trust substitute for evidence available from an established relationship. Therefore, the theory would predict that the majority, if not all of the EOC respondents, would rely on the categorical assumption of a colleague’s role as a trust criterion. This was not the case, and this conclusion is supported by four pieces of evidence:
As explained in Section 5.2.2.1, the EOC was characterised by an initial lack of role clarity during the early stages of the response. Therefore, it would have been more difficult for the respondents that were using role as a trust criterion to determine whether their colleagues could be trusted; because they potentially did not know why their colleagues were included in the response.
As shown in Section 5.2.3, the members of the EOC took on additional roles in order to meet the requirements of the national operation; making it more difficult to determine what role a colleague originally had within the EOC, and what was an extra task.
198 The respondents relied on many trusting criteria other than role; the majority of which were not categorical assumptions but rather evidence based. The data in Table 5.3.a above shows that the respondents relied on evidence-based information almost twice as much as they did categorical assumptions. Therefore, with the respondents not being able to easily determine their colleagues’ roles in the operation, the results suggest that they may have turned to alternate trusting criteria when determining whether to trust their counterparts.
An analysis of each respondent’s individual trusting criteria showed that no respondent relied on role alone as a trusting criterion.
Therefore, the data suggests that, although role was an important trusting criterion within the EOC, it was not relied on to the extent that swift trust theory predicts. Rather, the results suggest that the respondents relied more on evidence-based trust criteria than categorical assumptions.
Competency Respondents not only wanted to know what their colleagues did within the EOC, but also that they were competent. A comment was coded against the trust criterion of competency if the respondent said that trust was built through a colleague being proficient in their role. Interestingly, while no respondents suggested role was the sole determinant or influencer of trust, of the 19 counts of role recorded, only three were not paired with competency. Overall, there were 21 counts of competency recorded as a factor influencing trust formation.
“…part of [trust] would have been the quality of the work they were producing…You could tell when someone’s made an effort. This is definitely someone you want to ensure stays on.” (EOC 14)
“…to me [trusting], as long as he’s been given the job, I expect him to deliver.” (EOC 17)
“…you just need to place our trust that they will get the work done. That is why they are here.” (EOC 4)
“…[trust is] also related to how they do their work…when they make the same mistake, you know something wrong.” (EOC 6)
“…there are qualified people, people with a lot of experiences, they have been part of the operations for some time, so you don’t really have to tell them how to think and how to look at things.” (EOC 9)
199 “…[for trust] I was looking at how they made their decisions [in their role]…” (EOC 12)
The results showed that role and competency were not easily separated as trust criteria. Furthermore, this data challenges swift trust criterion #2 by showing that for the respondents in the EOC, a colleague’s competency was as significant a trust criterion as knowing their role (competency being coded 21 times compared to role being coded 19 times).
Additionally, for the 12 respondents who relied on competency as a trusting criterion, nine of these were in positions of leadership or were experts brought in for the operation. Particularly, this included four of the five members of Current Operations, EOC 2 as an international logistics expert, as well as EOC 6, EOC 9, and EOC 12 as senior, permanent staff of the EOC. The data more readily suggests that those in leadership or those who had a specific area of expertise as part of the disaster trusted colleagues who displayed competency in their respective roles when working in the response. This is important to note because swift trust theory argues that members will value speed over accuracy when deciding who to trust. However, to understand the competency of a colleague takes time. Therefore, these results suggest that those in positions of leadership or who were experts in the operation required evidence to trust their colleagues, and did not favour speed and the use of categorical assumptions when forming swift trust. The data shows that they took a more balanced approach to the time-accuracy trade off than that suggested by swift trust theory. Significantly, this data shows that in the context of teh EOC, swift trust was not formed from categorical assumptions alone; but utilised evidence-based information.
Personality Lastly, the factor that respondents considered most important when deciding whether to trust a colleague in the EOC was their colleague’s personality. An interview comment was coded against ‘personality’ if the respondent referred to a personal trait of a colleague or anything that was specific to their character or behaviour (such as kindness, professionalism, generosity, honesty, not complaining, or simply greeting one another). In the context of trust for their colleagues, all except three of the respondents within the EOC used the personality of their local or international colleagues (or both) to some degree as the basis of trust formation. A total of 23 counts were recorded against the trust criterion of personality.
"The way they speak, the way they behave, the way they, you know, converse with you. From there I can know a person whom I can trust…I also count on their personalities." (EOC 15)
200 “…trust is something that comes from the inside…also, how, you know, they react in all the situations that they are faced with. The reactions that come out, it helps you judge what kind of [person they are]. You know, people who start complaining, complaining every day, you know that you are with some people who cannot, you know, be trusted to work…” (EOC 9)
“Something about, like, if I don’t know a person, I will not, I will not talk to him, I will be watching him unless, you know. If he is an open person and talks back to me and becomes friendly, then yes, I can share.” (EOC 7)
“…[trust is influenced by] the personality of the individual…it’s how they live up to their word…” (EOC 6).
“In local colleagues, what do I look for? Building some sort of friendly relationships with them. Being able to ask them questions, share information both ways...[and for internationals]…Honesty, their honesty.” (EOC 10)
"…it's how they talk, how their behaviour during that one week, how they express themselves…answering the telephones, call you to join in the tea room." (EOC 3)
“I just think they have to be honest with you…[and] caring, for yeh, for me and the people we are serving.” (EOC 8)
(Field note) “The day after their interview, I met EOC 10 in the stairwell on my way into the EOC and they had been thinking on their interview overnight. They wanted to emphasise that trust for them was also based on sharing of; information, experiences, stories, food, etc.”
Personality was the swift trust criterion mentioned most by EOC respondents (23 counts). Furthermore, those in positions of leadership as well as junior members of the EOC considered personality important when deciding who to trust. This is contrary to competency, which was valued primarily by those with seniority or with specific expertise. The data suggests that a colleague’s personality was a universal factor in determining trust; in that it was valued by the majority of respondents regardless of whether they were in positions of leadership or not. Furthermore, it also shows that again, speed was not solely favoured by the respondents within the EOC when deciding who to trust. Rather, the use of personality as evidence-based information highlighted the respondents’ desire for accuracy, rather than speed, when deciding whether to trust their colleagues. This is contradictory to what swift trust theory would predict.
201 Additionally, the high count of personality as a trust criterion was contrary to swift trust theory; which would argue the strongest criterion should have been a categorical assumption such as role. In reference to the theory, this means one of two things; either:
Personality, as a swift trust criterion, validated trust that first formed from categorical assumptions (in this case, primarily role); or,
Personality, as an evidence-based criterion, was just as pertinent as categorical assumptions in helping trust form (if not more so).
Given the evidence for role and competency as trust criteria, this research suggests that the latter proposition is true. The results suggest that an initial lack of clarity around roles during the early stages of the response meant that several of the respondents in the operation had to rely on other trusting criteria when deciding who to trust. As such, evidence-based criteria such as competency and personality became more pertinent.
It is not possible to determine from the data that, had roles been clearer, then competency and personality would have become less important. However, the data does show that of all the trust criteria utilised, in the EOC, respondents were predominantly determining whether to trust their colleagues by assessing a colleague’s personality, such as their: honesty, general behaviour, willingness to greet them, kindness, professionalism, generosity, and persistence. Significantly, and contrary to swift trust theory, this case study showed that personality, as an evidence-based criterion, was more pertinent in helping swift trust form than the other categorical assumptions adopted by the respondents.
Additional trust criteria used by respondents within the EOC Although role, competency and personality were the three most prominent trust criteria recorded in the data, additional trust criteria were mentioned by the respondents throughout the interviews. Specifically, after role, competence and personality, the next highest criterion was a colleague’s work ethic, mentioned 10 times. Work ethic was the code used when a respondent mentioned trusting their colleague if they were; diligent, punctual, delivering their work, etc.
“I would go for work output, I would go for commitment.” (EOC 17)
“…it’s also related to how they do their work…you give them something to do, they do it…” (EOC 6)
202 “…there’s also, you know, the willingness to help…The ‘can I help you with something?’ You know, ‘ok’. We’ve all got an extra workload but if someone’s got a bit of spare time and is willing to do the running around…” (EOC 14)
“…that they’re committed…if I need information or something from them and I’m here and they’re out somewhere, out of town, they would make an effort to come here. And I would do the same for them.” (EOC 11)
In light of the time constraints of the EOC, and the importance of competency as a trust criterion, it was consistent that work ethic also scored highly. Like competency, as an evidence-based criterion, a colleague’s work ethic was a trust criterion primarily for those in positions of leadership and seniority. For the six respondents that used work ethic as a criterion, five were senior staff or members of Current Operations. Furthermore, work ethic was not a trust criterion that could be determined from a first impression; rather it required a respondent to make a judgement over time. As such, work ethic appeared to be a criterion which further validated that trust in the EOC was built off evidence-based criteria more so than categorical assumptions. This again demonstrates that those in positions of leadership favoured accuracy over speed when deciding who to trust, contrary to swift trust theory.
The remaining trust criteria offered by the respondents had only a few counts in the data. They were concerning a colleague’s: seniority and authority in the EOC, degree of cooperation, appearance, parent organisation they came from, cultural background, motivations, and appointment as a civil servant. That said, although these factors did not occur as frequently in the data, some respondents weighted them just as important (if not more) as role, competency and personality when deciding who to trust.
Firstly, EOC 8, EOC 10, and EOC 11 all valued cooperation (through information sharing) as an important trust criterion. This is not surprising given their respective leadership roles and involvement within GIS, the clusters, and Current Operations. This criterion was also a by- product of the sensitivities placed around information security after the breach of trust by OCHA. As such, for these colleagues, information sharing held significant weight when deciding who to trust in the EOC.
“I think for trust as well it is the flow of information…I’m trusting that that person will also give me [emphasis stated by the respondent] information that is correct…” (EOC 8)
203 “I think if you are working with someone who’s reserved, who cannot really mingle with the others, cannot really share information with others, then I just feel that you’re not the right person to be in this operation.” (EOC 11)
Secondly, EOC 13 (as an international liaison officer from OCHA) was the only respondent who mentioned that they specifically considered their colleague’s motivations for being in the EOC. Furthermore, this was their only criterion for determining whether to trust another person.
"If I see somebody's values and being they generally want to do what's best for the people affected by the disaster, they will very quickly move up my scale of trust...but if I perceive someone is out for number one, then they might have a harder time moving up the scale of trust." (EOC 13)
Lastly, although only two local respondents considered being a member of the Fijian civil service as a criterion for trust (EOC 1, EOC 9), they placed a high value on its importance. Significantly, this trust criterion was only available to local colleagues.
“Any civil servant, they can be trusted.” (EOC 9)
“There is just this thing about the civil service where you just trust the other public servant. I don’t know how to explain it. There is this general trust in the civil service.” (EOC 1)
Despite swift trust theory’s emphasis on categorical assumptions, these additional trusting criteria raised by the data are primarily evidence-based (work ethic, cooperation, motivation). This is important to recognise because, when coupled with the other primary trusting criteria (role, competency, personality), the results again suggest that the members of the EOC utilised evidence-based trusting criteria more than categorical assumptions when determining whether to trust their colleagues. Although categorical assumptions did play a part (particularly role, as the theory predicts), they were not the overwhelming influence for respondents when deciding who to trust in the operation.
5.3.2. Trust criteria specifically applied by locals to international disaster management personnel As this research aims to provide insights into how internationals engage the trust of locals when responding to a disaster, it was necessary to also consider whether there were any unique trust criteria that the Fijian staff members applied only to their international colleagues. Consequently, in conjunction with the data in Section 5.3.1, every local participant was also
204 asked, ‘What three things could an international colleague do to gain your trust when working in the EOC during the response?’.
Foremost, the data showed that for locals to trust an international colleague, they needed to know the purpose of an international being in the EOC and how they were related to the TC Winston operation. Relating this to swift trust criterion #3, role was important; with some locals being very cautious of the intentions of any internationals coming into the operation and their access to response data and information.
“…[For internationals, I need to know] why are they here and what are they doing.” (EOC 1)
“…how they [internationals] are linked to the Winston operation.” (EOC 3)
“I need to know what they’re here…Why they’re here, what’ the purpose of them wanting all this information?” (EOC 11)
“…[Internationals should] state their purpose and how they are going to assist.” (EOC 4)
“You don’t just accept them into the system. You have to know how they operate and who’s…their source of funding and what are the ah, ah, what are the tasks that are given to them, and also to see a bit of their background. Where they have been operating before and stuff like that before you can trust them…Some of the groups coming in are well known, but the new ones…you need to do the background check.” (EOC 9)
“[Internationals should] have a discussion…[about] what [their] purpose is here. One thing is really, for locals, understanding what they’re here to do. What’s their role, what’s their responsibilities…” (EOC 14)
“[Internationals should] inform beforehand…Their intention. Their intention for coming over.” (EOC 17)
This data suggests that where an international colleague’s role was clear, the local would be more likely to trust them.
Some local respondents also said that the internationals could gain their trust through certain traits of their personality. The personality traits that the locals considered important were:
Humility in integrating into a locally led national response:
205 “Because they are coming into the situation, a lot would depend on how I see them adapting to the context…I expect them to put their modus operandi at the door and come in and work into the system…” (EOC 16)
“[International colleagues]…should be able to ah, listen. You know, and abide by, you know, certain, I wouldn’t say rules and regulations, but ah, maybe policies and directions. Respect, you know, they should respect the organisation and the director and what their wishes are. And just work closely with ah, with the team.” (EOC 7)
"…a number of international people I have worked with, they come off as 'I know better than you'. But you don't know the environment that I work in. I mean, you don't know what happens with the Indigenous people, I know it better.” (EOC 8)
“When [internationals] come in, they should come in with open hands to work with us.” (EOC 9)
“…they integrate with me, and it’s not going to be only me, it has to be my whole team...It’s the experience that [the international staff] actually have doing this work a lot of places and knowing that they are the visitor here and to be able to work with us. The international colleagues that we’ve worked with here, they have been doing that quite well.” (EOC 11)
“[Internationals should]…come in and try and understand how we work. You know, don’t try and force themselves in because they think they are international, they would know better. But because, every country is different and how we run things are different. So maybe come and understand how systems, how things work.” (EOC 12)
“[An international] must also [be able to] take on board criticism…” (EOC 17)
Having conversations and being kind (including personal greetings):
"I also count on their personalities. The way they speak to you and the way they respond.” (EOC 15)
“…say hello…And have a conversation. Like an introductory conversation…” (EOC 1)
206 “[Internationals should] introduce themselves…” (EOC 4)
"They regularly talk to you. At least they know you are there." (EOC 6)
“Don’t just talk about work, talk about other stuff, like rugby, you know, stupid things...If someone invites you to sit in the kitchen and have tea, join them. That helps build the trust and relation.” (EOC 12)
Being sensitive to the losses experienced by the local staff:
“Every single person was in here working knows someone or has been impacted or was impacted by Winston….[it is important to] spend a week just building that relationship and proving that you’re worthy enough to be where you are sitting.” (EOC 14)
Two local participants also specifically mentioned that the parent organisation of their international colleague would foster trust in that colleague.
“Their company and the experience of the company…[and] the reputation of the organisation.” (EOC 3)
“[I would be looking for] some sort of identity. Which organisations they are part of. Because we don’t want to open up our doors and then find out, that by opening…you know, we may have to do this ah [obligation], you know, in the near future.” (EOC 9)
Lastly, it is important to note that the international respondents had fewer criteria available to them on which to extend and receive trust, than locals; they were not able to build swift trust on the basis of being a civil servant or having a national cultural link.
Overall, these results have two major implications. Firstly, in terms of swift trust, the data showed that for locals, both categorical assumptions and evidence-based trust criteria played an important part in them trusting their international colleagues. Secondly, the data also raises important factors for international personnel to consider when deployed to another country’s response. The results suggest that for international personnel to be trusted by their local colleagues, they should:
Display humility by integrating into the response, remaining subject to the host government and not promoting their expertise over the local staff.
Be kind and introduce themselves.
207 Explain their role in the operation.
5.3.3. Culture as a categorical assumption used in the development of swift trust In addition to the aforementioned trust criteria, throughout the interview process, it became evident that both Fijian culture and the involvement of the military were potential influencers on how the respondents (primarily the locals) developed trust. As such, additional interview questions were added to understand both Fijian culture and the military, and the impacts that these factors had on trust formation within the EOC. This section focuses on culture as a categorical assumption used in the development of trust (with the next section addressing the military’s involvement as another categorical assumption).
Swift trust theory argues that categorical assumptions can extend to include suppositions based on culture. In this research, the influence of Fijian culture on trust within the EOC was individualistic; varying for each local respondent. (Refer to Appendix B for a description of the types of cultural relationships within Fiji). For the local respondents in the EOC, the data showed that:
seven felt that having a cultural link helped them form trust, and,
eight believed that having a cultural link meant that they were more likely to trust a local than an international.
Focussing on the former, although every local respondent within the EOC had a cultural link with their other local colleagues, approximately half felt that this was a factor that contributed to trust formation; in addition to the other categorical and evidence-based criteria.
“Ah yeh…myself and the person I am calling Kai…We are from the same province. So in that way, we have that natural bond, natural trust.” (EOC 4)
“Because, especially for Tau eh, for Tau, it’s like, like you’re Tauvu or cousins to joke with him and you know. But workwise, it’s easy to work with…” (EOC 15)
“Um, you know, having a familiarity with a person, ah, apart from those in the office, it makes a lot of difference…It also makes an instant connection. A literal…connection if you have that kind of relationship.” (EOC 17)
“In our traditional ah, I mean in our traditional cultural way of living, there, I don’t know how to explain it to you, but people from other clans, from other districts that are
208 traditionally linked to us, it makes trust much stronger. Because ah, in our traditional links, we respect, we respect one another… If you are trust me with some work, and you go away, leave me some work. I will not mess up with this work because you will not only look at me, you look at people behind me.” (EOC 9)
“It sort of does [help trust]…Ah, who, the person that I’d have a cultural link like a Mata is someone that I would, I would respect…Um, well I know with the Dreu, if I said something, or they asked me to do something, I would follow through. It’s out of respect that I would have to do it.” (EOC 12)
“[The cultural links] all have the elements of trust within that label. So if you’re a…Kai, the level of trust is a lot higher than if you’re a Tau. Um, or if you’re a Mataqali, there’s another level of trust there that’s specific to certain things, but not other things.” (EOC 16)
The data suggests that culture influenced trust formation for these local respondents; as it fostered a special kind of mutual respect. However, each Fijian had to personally establish if they would integrate their cultural relationships into the workplace. For the remaining local respondents, culture did not influence trust formation in the EOC.
"I think when it comes to work...it's just a step up from where our culture prohibits us to explore." (EOC 8)
“I’ve never really considered the traditional relationships as a decision maker for me.” (EOC 1)
“For me it’s not…for me, to trust a person I need to know about the person personally…Yeh, the [cultural] relationship doesn’t count.” (EOC 3)
"In Fiji, one thing they say about the culture…is that everywhere you go in Fiji, you would have a sort of cultural or traditional link…[but] when we bring that kind of relationship into our workplace, we need to know where to be able to draw the line…with me...there is a clear demarcation of how far I can go in that relationship...it really depends on the individual...there are people that take this relationship seriously." (EOC 11)
These results showed that not all the local staff in the EOC were willing to let culture influence how they developed trust. This may have been a result of the fact that the different types of cultural relationships warranted different levels of respect and ways of interacting; which could place the respondent in a difficult situation. (An example would be a junior local colleague
209 working under a senior local colleague in the EOC; however they are both Tauvus and therefore have a friendly rivalry which can be used to jokingly take advantage of one another. The behaviour between Tauvus would not be considered professional, especially between junior and senior members of staff; yet culturally, it would be accepted.)
As outlined above, half the locals (including both junior and senior members) chose to use cultural relationships in the EOC when deciding whether to trust another local colleague; affirming swift trust theory’s inclusion of contextual factors, such as culture, as a categorical assumption. The data also suggests that culture was more significant to some locals than others and subsequently, its influence on trust formation differed depending upon the importance ascribed by the respondent.
Lastly, the case study showed that eight of the local respondents within the EOC believed that having a cultural link meant that they were more likely to trust a local than an international. The results suggested that some locals found it easier to trust a local than an international because culture:
made it easier to communicate with a local than an international:
“[A local is] just a few steps ahead…Because if I am explaining something to you…I can only explain it to you in English. And if you really don’t understand, I have to try and break it down…But if there is a local…If I have that sort of relationship, then I can use that to…if it’s my Tau, I say, ‘see, that’s a problem of not doing studies when you are supposed to’ and that’s where the relation comes in.” (EOC 11)
led to them being more comfortable working with a local than an international:
“Because of the cultural links, yes…it’s just I’d be more, is comfortable the correct word, comfortable to work with, easier to work with a local than a [international colleague]…” (EOC 12)
“Yes, I suppose…Because especially for Tau eh, for Tau, it’s like…cousins to joke with him and you know. But workwise, it’s easy to work with him…” (EOC 15)
created links between the locals:
210 “Ah, in a sense…Yeh [it is easier to trust a local], because of the link and maybe the fear of not knowing them. [It is] dependent on the [international] person.” (EOC 4)
“Yeh, [I] should answer probably yes…Just because we have those links, those pre-existing links.” (EOC 16)
“Yes…Because you have a link with them.” (EOC 6)
“Yes [that cultural link]…could be a contributing factor to trust.” (EOC 8)
“…in our culture there is a lot of respect…because we have, ah, traditional links with one another.” (EOC 9)
The remaining local respondents said that they would not trust a local more than an international just because of culture.
“[Considering the influence of culture] not necessarily. I would go for work output, I would go for commitment. That would be my guidance when I deal with people. Whether it is my Mata, or for example, somebody from overseas, yeh.” (EOC 17)
“For me personally I have never really considered the traditional relationships as a decision maker for me.” (EOC 1)
"For me to trust a person, I need to know about the person personally…the [cultural] relationship doesn’t count." (EOC 3)
Considering swift trust criterion #3, the use of culture as a categorical assumption which promoted trust formation varied for each local. Approximately half of the respondents agreed that culture would help them when deciding whether to trust another local in the operation. Similarly, about half also said that they were more likely to trust a local colleague than an international colleague due to the way culture created a link between the locals and made communication and working together easier. These findings are significant because they show that, in a multi-national disaster operation, culture can be a categorical assumption used in trust formation. Furthermore, international personnel are automatically precluded from using culture as a trust criterion; which, for some of locals, is a contributing factor for trust. More generally for the field of disaster management, the results also showed that there are trusting criteria that an international colleague should be aware of when being deployed as part of an operation;
211 even though they themselves are not able to benefit from, or utilise, culture-based categorical assumptions when forming trust with their local counterparts.
5.3.4. The military’s involvement as a categorical assumption used in the development of swift trust The involvement of the Fijian military in the response to TC Winston affected how trust formed within the EOC. In order to understand this, it is necessary to understand how the military became incorporated into the operation.
The Fijian military’s involvement in the response was only meant to support the government led EOC, primarily through assisting the National Emergency Operations Centre (the response data centre for the operation). This was as per the National Disaster Management Plan.
“Staff from the Fiji Military Forces…will assist the [National Emergency Operations Centre] operating teams in communication, coordination and reporting work related to emergency operations, including evacuations and movement of relief supplies.” (NDMP, 1995, p. 27)
It was not expected that the military would inadvertently take over a key role of the National Emergency Operations Centre; which was meant to, “provide operational and logistic support to the emergency operation” (NDMP, 1995, p. 49). However, the military took on this responsibility through the establishment of Future Operations. Future Operations was not a part of the National Disaster Management Plan and therefore its establishment was unprecedented, with no clear process explaining how it was to be integrated back into the main EOC. Furthermore, Future Operations was manned by the Fijian military and coordinated by military commanders. The key consequence of the creation of Future Operations was that the response turned from a civilian response into a military-civilian operation.
Consequently, given Fiji’s military context and this change in the EOC’s original governance arrangements, each respondent was asked whether they though the involvement of the military in the response affected how trust formed. Of the respondents, five locals said that the involvement of the military affected how they formed trust in the operation, or (to use swift trust theory’s construct of categorical assumptions) that their perception of the military was a categorical assumption that impacted trust formation.
“[Due to the military, people] were a bit more hesitant to talk about personal stuff or share experiences that they had in the office or with decisions that’d been made…” (EOC 12)
212 “They [civilians] have that fear factor [toward the military]. They have to be careful with what they say.” (EOC 4)
“The one thing that was different was probably the large…the civil and military coordination and having that play for at least the first thirty days just to see how the two groups of people are interacting and getting over the whole hurdle of…the initial prejudice. And then going, ‘ok, so this is how we can actually work together’." (EOC 14)
The formation of trust was hindered for some respondents, not only because of their perception of the military, but also due to the military’s attitude toward civilians working in the EOC.
“For them [the military], trust played a big role. So the guys they let into the room [Future Operations] were the ones that they would trust. If they didn't trust you, they'd keep you at arms-length…[because of] the military thinking of 'everybody's out to get you'.” (EOC 16)
“Sometimes it looks like, sometimes, the trust is not there. They [the military] think that they are the only ones that are fit to be trusted to do job...” (EOC 9)
"…the military thinking civilians do not know what they are doing, which I think is very ignorant because I feel like people are being put in a certain situation of responding immediately to an event, you know, people actually perform better." (EOC 8)
(Field note) “Future Operations is physically isolated from the main office of the EOC; being in another room in the same building. This room is restricted and is manned by Fijian military personnel, supported by the Australian and New Zealand Defence Forces. The military have control over who they interact with and who they let into Future Operations.”
Although five local respondents stated that the military affected how they formed trust within the EOC, it appeared that the remaining respondents, who had either worked with the military or valued their input, were confident to interact with Future Operations and did not let the military’s attitude affect how trust formed for them.
“[The involvement of the military] created a different dynamic. Those people who were familiar with military ways got more out of it than those who were not.” (EOC 16)
213 “...they’ve really taught us a lot because they’re military person, they’re more disciplined and they work with time. They teaches us a lot of lesson…like it strengthens us to work more efficiently and improve on our work.” (EOC 15)
“…the way I see it in regards to this operation is actually ah, [Future Operations] makes the operation more efficient…I don’t know about the others, but for me, I am very much a, ah, what’s the word here, I am very much the same. The same person when they are around. It’s just the same.” (EOC 17)
Overall, the involvement of the military had a two-fold effect on the formation of trust within the EOC. Firstly, five respondents within the EOC were slower to trust members of the military due to their perception of the military. That is, in terms of swift trust criterion #3, these respondents had a categorical assumption about the military which influenced how they formed trust. Secondly, the involvement of the military changed the overall dynamic of the EOC. The shift from a civilian operation to a civilian-military operation meant that the process of forming trust was impacted by the inclusion, and subsequent protocols, of Future Operations. Those familiar with the workings of the military more readily adapted to this change; whilst others did so less easily. Although it is not possible to know what the EOC would have been like without Future Operations, the respondents’ comments showed that the military’s involvement distinctly altered the way that trust formed within the EOC.
Furthermore, the results provided valuable insights regarding military involvement in a disaster response. The case study particularly showed:
The military can become involved in the response and that this can lead to changes in the internal structure of the EOC.
The military can take on an expanded role of coordination, planning and logistics.
The involvement of the military can affect how trust forms in the EOC, as not all personnel feel confident working with the military. However, those who have worked with the military before (in another context) are more comfortable with the military’s integration into an operation.
5.3.5. Addressing Swift Trust Criterion #3 Reflecting on swift trust criterion #3, the results showed that role, competency and personality were the three primary trusting criteria used by the respondents within the EOC. Other less cited trusting criteria included: work ethic, being a civil servant, appearance, or a colleague’s parent
214 organisation. Additionally, the results also showed that two unique categorical assumptions, related to culture and the military, influenced trust formation for some members (particularly locals) in the EOC.
An analysis of the results showed that overall, the respondents did not rely more on categorical assumptions than evidence-based information when forming trust. Rather, the respondents favoured evidence-based information when deciding whether to trust another colleague within the EOC (the data contained 62 counts of evidence-based criteria to 40 counts of categorical assumptions). Although role, as a categorical assumption, was an initial trusting criterion used by many of the respondents, the data showed that competency and personality (as evidence-based criteria) were both more commonly used than role as trust criteria. As previously discussed, this appears to have been due to two primary reasons. Firstly, an initial lack of role clarity (see Section 5.2.2.1) meant that knowing a colleague’s role in the EOC was not simple. As such, the respondents may have relied on additional trusting criteria to compensate for not knowing a colleague’s role. Secondly, the EOC, and particularly those in positions of leadership, were cautious about sharing information and wanted to protect the EOC’s response data. Therefore, those in positions of leadership were more likely to consider the competency of their colleagues, their work ethic, and their intentions when deciding who to trust; to ensure that their own objectives (such as data protection) were not jeopardised due to a categorical assumption made. This was particularly true when deciding whether to trust an international colleague. Contrary to swift trust theory, the data showed that the respondents, and particularly those in positions of leadership and expertise, favoured accuracy over speed (that is, evidence over categorical assumptions), when determining who to trust in the operation.
Considering the integration of international colleagues into the EOC, swift trust theory does not describe in detail the affects that having a multi-national team could have on the formation of swift trust. The data showed that for some of the local staff to trust their international counterparts, their international colleagues needed to humbly integrate into the EOC and work alongside, and under, the national EOC leadership. That is, the local staff again required further evidence-based information when deciding whether to trust an international colleague.
Swift Trust Criterion #4 – Confirmation of assumptions Swift trust theory argues that a respondent will want to confirm the categorical assumptions they have made about their colleagues when deciding whether to trust them. This can be done via two methods:
215 A respondent might deliberately behave in a certain way so as to induce confirmation of the trusting criteria they have applied to their colleagues.
A respondent might look for their colleagues to behave in accordance with the categorical assumptions they have made about them.
Both these approaches are ways in which the respondents might obtain additional evidence to validate their categorical trusting criteria. If they observe a colleague not acting in accordance with the trusting criteria they had applied to them, then swift trust may be weakened. However, swift trust criterion #4 relies upon the supposition that categorical assumptions were primarily used in the formation of swift trust; which was not the case in the EOC. Therefore, this section first addresses whether the respondents sought to validate the trusting criteria they applied to their colleagues when using both categorical and evidence-based criteria. It then discusses the effects any unexpected behaviour by a colleague had on trust formation in the EOC.
5.4.1. Validation of the trusting criteria applied to a colleague Swift trust theory argues that a respondent may behave in a certain way in order to affirm their assumptions about their colleague, in an attempt to reduce any uncertainty concerning their trustworthiness. Within the EOC, three colleagues referred to deliberately behaving in a particular way in order to clarify their trusting criteria. These were EOC 2, EOC 9 and EOC 11; all of whom were disaster experts or senior members of the operation.
Firstly, EOC 2 and EOC 9 both emphasised competency as a trust criterion and thus sought to determine the proficiency of their colleagues. Both respondents stated that they did this by assigning their colleagues tasks and judging their competency.
“You task people differently. If you know someone is a flake, you’ll give them something simple to do and expect that it will probably not happen and so you plan for that not to happen and have a contingency plan in place so it doesn’t impact what you are doing. If you find someone you can rely on, you load them up.” (EOC 2)
“You don’t have to tell them but that is something that you can start asking questions, but not showing them that you don’t trust them, but with some questions….[then] you can slowly give responsibilities, not bigger ones, but test with some [smaller tasks].” (EOC 9)
Secondly, EOC 11 said that for their international colleagues, they would find out from an independent source why that colleague was coming into the EOC, and then cross-check this by
216 asking the international colleague in person. They did this to affirm their trusting criteria of honesty and cooperation.
“…before they come in, what I do, is to get an understanding of why they’re there, personally. And then I’ll ask them again as to why they’re there.” (EOC 11)
Although these three respondents deliberately behaved in a particular was to validate their trusting criteria, their motivations for doing so differed. The quotation provided by EOC 2 shows that they were concerned with the impact that trusting their colleagues would have on themselves and the work that they had to complete. However, both EOC 9 (a senior EOC staff member) and EOC 11 (Current Operations), as senior, local members of the operation, were more concerned with protecting the operation than their own interests. EOC 9 stated that part of their role was to determine who to trust in order to protect the EOC and the Director from being undermined.
“…I don’t like the people that you trust either. I cannot trust them just because of you [referring to the Director]…That way I am helping him because I just don’t think anyone who comes in will give him advice as a friend to the organisation.” (EOC 9)
Similarly, EOC 11 was conscious of the protections surrounding information dissemination and how this could be used against the EOC; particularly by international parties after the breach of trust by OCHA. As such, they would ask why someone wanted response data.
“Considering that we [Current Operations] were the hub of information, so I would really, I was just one person who would ask them ‘why? Why do you need this?’” (EOC 11)
Thus, the results showed that, although these respondents sought to affirm the trusting criteria they applied to their colleagues, their motivations for doing so were different; that being, personal protection (EOC 2) versus protecting the operation (EOC 9 and EOC 11).
Viewing these results in light of swift trust theory reveals two key implications. Firstly, it shows that all bar three respondents did not mention deliberately behaving in a particular way in order to validate their trusting assumptions made about their colleagues. Secondly, it shows that confirmation of assumptions may be less pertinent when the trust criteria used by the respondents relies primarily on evidence as opposed to categorical assumptions. This is because evidence-based trusting criteria require time to be validated. This presents a gap in swift trust theory. That is, it remains to be tested whether the respondents’ greater reliance on
217 evidence-based criteria meant that there was not the subsequent haste in trying to validate an initial categorisation of a colleague. The results suggest that the respondents were able to continually monitor their evidence-based trusting criteria (such as competency and personality) by observing their colleagues’ behaviour over time. Therefore, the results predict that in a disaster management context, if the members of a temporary system rely more on evidence-based trusting criteria, then swift trust criterion #4 becomes less relevant.
5.4.2. Expectations of behaviour Although only three respondents within the EOC sought to validate their trusting criteria by altering their behaviour, swift trust theory argues that swift trust can be damaged if a colleague does not behave as a respondent expects. Therefore, during the interview, each respondent was asked whether their colleagues behaved the way they expected them to during the operation, or whether they had to adapt to unexpected behaviour. This section will discuss the respondents’ answers concerning their colleagues’ behaviour and the implications this had on the formation of swift trust.
Within the EOC, 11 respondents said that their colleagues behaved the way they expected. However, ‘expected behaviour’ was classified by the respondents to mean different things:
Many of the respondents correlated ‘expected behaviour’ with their colleagues doing their job in the EOC and working long hours in order to meet the requirements of the response.
“…rather than adapt to unexpected behaviour, as in the negative, the adaptation was more to the positive, you know, like… People were willing to do the long hours, willing to eat tea and biscuit.” (EOC 16)
“I’ve been really impressed with the guys here…everyone stepped up.” (EOC 2)
“No I think given the, that for me was the really encouraging thing, was the long hours people put in to work. I didn’t expect that from so many people.” (EOC 14)
“I think people pretty much act the same, just faster.” (EOC 8)
Some considered ‘expected behaviour’ in a general sense and did not specify exactly what this behaviour looked like.
"I think they generally react normally." (EOC 6)
218 “…in general I’d say people behaved more or less the way I expected them to.” (EOC 13)
“I believe the environment here, as you can see, was ok under those circumstances.” (EOC 17)
EOC 9 stated that their colleagues behaved as expected because they were trained in disaster management.
“Yah [people behaved as expected]…mostly the people who come into our operations, that…[have] some kind of training…they sort of, have that idea ah, what is expected of them…” (EOC 9)
The remaining respondents, who said their colleagues did not behave how they thought they would, had different reasons for why this was the case. Both EOC 3 and EOC 4 stated that unexpected behaviour was a consequence of the stresses and scale of the operation.
"Sometimes we needed to adapt to special behaviours [by our colleagues] because there was a lot of pressure and tension.” (EOC 4)
“Because of the, maybe because of the extent of the event…sometimes people behave unexpectedly….Say for instance, Winston was a category 5 cyclone. If it was a category 3, 1, you know the extent of the work will not be as heavy as a category 5 cyclone.” (EOC 3)
However, when asked, both respondents said that the behaviour of their colleagues would not change how they would work in a future operation; the unexpected behaviour was perceived as a minor factor in retrospect. That is, their colleagues’ response to the stress of the operation did not advertently challenge any of EOC 3 or EOC 4’s trusting criteria; and hence seem to not have affected trust formation for these two respondents.
The remaining colleagues (EOC 7, EOC 10, EOC 11) used the trusting criteria of role, honesty, and seniority and authority (respectively) when determining swift trust. These respondents all experienced their colleague(s), at one point, not behaving in accordance with the criterion on which that trust had been based.
Role – “…a lot of people who are working in this operation have not been trained on this kind of operations. So they come in with different knowledge and background and attitude. So you got to adapt to them.” (EOC 7)
219 Honesty – “This is one in particular who I just don’t trust…because I feel that they’re not telling me all the information…It’s when you’re working with this person and they’re asking for information and you’re giving information all the time and then you say, ‘so what’s happening’ and they don’t tell. And then later on you go ‘oh’ (in realisation).” (EOC 10)
Seniority and authority – “…people who are in this office who should know what has to be done, instead of running…there was still things that could have been put in place already that were not in place.” (EOC 11)
Analysing these responses revealed multiple implications for swift trust theory. Firstly, for EOC 10, a breach of honesty (one of their trusting criteria) by a colleague led to swift distrust in that colleague. Conversely, for EOC 7 and EOC 11, despite not having their trusting criteria met by their colleagues’ behaviour, these respondents stated that this led them to distrust their colleagues. Therefore, the data showed that; if a colleague did not behave in accordance with a respondent’s trust criterion, then the impact this had on the formation of trust for a respondent was related to the value the respondent ascribed to that trusting criterion. In this example, it can be concluded that EOC 10 highly valued honesty, whilst the other respondents (EOC 7 and EOC 11) did not put as much weight on their respective trusting criteria; thus limiting the impacts of the colleague’s behaviour on trust formation.
These findings provide an original contribution to swift trust theory by showing that:
Not all trust criteria hold the same weighting for a person when determining trust.
If a colleague does not fulfill a respondent’s trust criteria, trust may still be possible, given a respondent generally relies on more than one criterion. That is, not every trust criterion is absolute in determining the existence of trust such that, when one criterion is not fulfilled, a colleague is automatically deemed untrustworthy.
Secondly, the results also showed that examples of swift distrust (as with EOC 10) provide a valuable insight into the prescribed value of each trusting criterion for an individual. That is, it may be easier to determine which trusting criteria are more important to a respondent by analysing examples of swift distrust rather than examples of swift trust.
In summary, the results showed that in the EOC, all except three respondents did not mention actively changing their behaviour to validate their trusting criteria. Secondly, 10 of the respondents said their colleagues did not behave unexpectedly. However, for the remaining
220 respondents that did experience unexpected behaviour by their colleagues, the impacts of this behaviour on trust formation differed. The results showed that if a colleague did not behave in accordance with a respondent’s trusting criteria, then the impact this had on the formation of trust was related to the value ascribed to that trusting criterion. The data showed more generally that a respondent does not weight all trust criteria equally.
Swift Trust Criterion #5 – Moderate interdependence Swift trust criterion #5 states that swift trust is more likely when interdependence between members of the temporary system is moderate. However, at the heart of swift trust criterion #5, swift trust theory, and indeed the wider trust literature, is the concept of risk. That is, the more a person relies on another person (interdependence), the more they have to lose if that person undermines them; thus, trust becomes more necessary. Therefore, to address swift trust criterion #5, this research sought to understand both the actual and perceived degree of interdependence between the respondents in the EOC.
This section first reiterates the network diagram for the members of the response, showing their actual degree of interdependence. It then provides a comparison of this data against a respondent’s perception of their reliance on their colleagues during the response. Finally, it shows that simply being a member of the EOC (that is, experiencing group inclusion) meant that a respondent was dependent upon their colleagues, regardless of whether they interacted with them directly or not.
5.5.1. The degree of interdependence between members in the EOC As explained in Section 4.2.3, the task of responding to TC Winston required interdependence. The actual degree of interdependence for each respondent can be gleaned from this data. However, in line with social constructionism, it was necessary for this research to consider each respondent’s perceived degree of interdependence within the EOC. This was necessary because, as an example, although a respondent may have a high degree of interdependence with their colleagues, if they believe that they do not rely on their colleagues, then in accordance with swift trust theory, they could feel less at risk because of their ‘perceived isolation’; thus minimising their need to develop trusting relationships. From the interview data, in order to determine each respondent’s perceived degree of interdependence, the respondents were asked how much they relied on their colleagues as part of the operation. The coded results (accompanied by sample answers) were:
Low reliance on colleagues – seven respondents, with responses including:
221 “Zero percent” (EOC 1)
“Not at all. If I’m not here, my job doesn’t get done.” (EOC 2)
“…most of the work was done by myself…” (EOC 4)
Moderate reliance on colleagues – four respondents, with responses including:
“[I rely on them] moderate.” (EOC 6)
“I have to do…ah 70% [of the work myself].” (EOC 7)
“The majority of what I did, I did on my own…[but] it has to be a partnership.” (EOC 17)
High reliance on colleagues – seven respondents, with responses including:
“Oh, [I relied on my colleagues] say like 90%.” (EOC 16)
“I rely on them for, I rely on them, in order to carry my work effectively.” (EOC 3)
“I relied on the relationships [with the EOC]…In terms of the report, I was heavily reliant.” (EOC 13)
“[I relied on my colleagues] a lot. They were the ones who actually gave me data to work with.” (EOC 14)
Swift trust theory argues that moderate interdependence is optimal. This is because with moderate interdependence, a respondent is not entrusting too much to their colleagues; thus minimising their risk of being undermined or affected by a problem. However, the data showed that only four respondents felt they had a moderate degree of interdependence with their colleagues. Conversely, seven respondents felt they relied heavily on their colleagues in the operation, and seven felt they worked with minimal dependence upon their colleagues. The analysis showed that the data seemed to correlate to each respondent’s role within the operation.
Respondents perceiving a low degree of interdependence:
o Support roles which provided assistance to those in positions of leadership (EOC 4, EOC 5, EOC 8, EOC 12).
222 o Respondents who worked on specific response tasks or in isolated teams (EOC 2 in logistics, EOC 10 in clusters, EOC 1 in media). Each of these respondents were either self-dependent or did not need many inputs from their colleagues.
Respondents perceiving a moderate degree of interdependence:
o Senior members of the EOC (EOC 17, EOC 7, EOC 9) including those in Current Operations and the head of logistics.
o EOC 6, as the respondent leading a shift team in the National Emergency Operations Centre.
Respondents perceiving a high degree of interdependence:
o Those working in Current Operations (EOC 11, EOC 16, EOC 14) who coordinated and planned the next stages of the response.
o The procurement team, who coordinated all the purchase orders and timesheets for the response (EOC 15, EOC 18).
o The OCHA liaison (EOC 13), whose role was to coordinate the international aid from OCHA with the EOC.
o EOC 3, as the head of the media team for the entire operation.
However, what remained to be understood, and what swift trust theory does not address, is why some members of the EOC in very similar roles (or the same team) had different degrees of perceived interdependence; particularly those in Current Operations. To address this, the interdependence data was tabulated against the Network Mapping Worksheet data to compare the number of connections a respondent actually had in the EOC against their perceived degree of interdependence (refer to Table 5.5.a).
223 Table 5.5.a The respondents’ recorded reliance on their colleagues compared against their perceived reliance on their colleagues
Percentage of Perceived reliance Number of Respondent Team recorded on colleagues connections connections EOC 17 Current Operations 22 14% Moderate EOC 7 Current Operations 16 10% Moderate EOC 9 Logistics 14 9% Moderate EOC 11 Current Operations 12 7% High EOC 16 Current Operations 12 7% High EOC 2 Logistics 10 6% Low EOC 18 Procurement 10 6% High EOC 14 Current Operations 10 6% High EOC 12 Support 9 6% Low EOC 13 OCHA 9 6% High EOC 5 Support 8 5% Low EOC 4 Support 7 4% Low EOC 15 Procurement 5 3% High EOC 10 Clusters 4 2% Low EOC 3 Media 4 2% High EOC 8 Support 4 2% Low EOC 1 Media 3 2% Low EOC 6 Support 3 2% Moderate 162 Total 100% (Average of 9)
The data provided one main reason why some respondents in the Current Operations team had different perceived degrees of interdependence. EOC 17 and EOC 7 respectively had the highest number of connections in the EOC, yet they both felt they relied only moderately on their colleagues. Conversely, the remaining members of Current Operations (EOC 11, EOC 14 and EOC 16), all felt they had a high degree of interdependence with their colleagues. The results suggest that for EOC 17 and EOC 7, their higher number of connections meant that it might not have been possible for them to rely heavily on many colleagues because there were simply too many relationships to be maintained. This conclusion is also supported by data from EOC 9, who, as a senior member of the operation, had a high number of connections, yet only perceived moderate reliance on their colleagues.
As well as considering discrepancies between the different members of Current Operations’ perceived degrees of interdependence, the data also revealed another unique finding. EOC 15, EOC 3 and EOC 6, despite having a low number of actual connections in the EOC, felt they relied heavily or moderately on their colleagues. This data is significant because it shows that although
224 a respondent may have only worked with a few people in the operation, they worked closely with those people and relied on them.
Overall, the data showed that there was an association between the respondents’ perceived level of interdependence and the actual number of connections they had in the EOC. However, there were some exceptional cases. These showed that a respondent may rely heavily on only a few connections in the EOC, or conversely, may rely only moderately on many colleagues. This data has two implications for swift trust theory:
The degree of interdependence within the temporary system is not only a function of the number of connections a respondent has with their colleagues. Rather, interdependence is also measured by how reliant a respondent feels they are on their colleagues. Consideration of both of these factors as determinants of interdependence refines the concept of this criterion, as it is presented originally in swift trust theory. A dual conception of interdependence (actual and perceived) extends the parameters of swift trust theory, to consider both the actual network of connections that may be the basis of trust formation, and the perceived degree of interdependence each member of a temporary system has on their colleagues (which may equally, or additionally, impact trust formation).
Secondly, swift trust theory argues that categorical assumptions provide insufficient evidence for a person to trust a colleague, if they have a high degree of interdependence with that colleague. The data also validated this conclusion; given that all those with a high perceived degree of reliance on their colleagues (EOCs 3, 11, 13, 14, 15, 16) relied on evidence-based information more than categorical assumptions when forming swift trust in the EOC.
5.5.2. Interdependence due to group inclusion Having analysed the individual degree of interdependence for each respondent within the EOC, this research sought to assess how the concept of group inclusion impacted interdependence. To do this, each respondent was asked about the likelihood that an operational problem in the EOC would affect just themselves, or whether it would impact the wider EOC. Examples could include the wrong supplies being ordered by the logistics team, or inaccurate information being provided by Current Operations or another member of the EOC. The purpose of this question was to understand the degree of risk a respondent believed they were subject to when accepting to be a part of the operation.
225 All respondents, except EOC 10, stated that an operational problem within the EOC would impact the wider EOC and not just themselves. This is significant when considering trust formed in the EOC because it indicates that respondents understood that simply being a part of the EOC meant taking on the risk of each member of the operation doing their job proficiently; regardless of whether they interacted with that colleague or not. Respondents cited two primary reasons for why a problem in the EOC would affect the wider office:
Tasks were interdependent:
“It affects the whole team…For me there is [a risk if I do not do my job]. People need to know what’s happening…” (EOC 1)
"Coordination work is very important. So if one cell is not doing their work very well or one member of the team doesn't contribute, it affects the whole team's operation." (EOC 7)
“I think it impacted the wider team…If you’re not going to give me very accurate information, I won’t be able to map that, and because I won’t be able to map that, we won’t be able to show it.” (EOC 8)
“It would affect the whole team…Because a lot of the things that we [do are interrelated].” (EOC 11)
“It impacts the whole team…Because like, all the works are related…So if [they do] not submit this, that will affect another [person’s] work, and that one will affect…” (EOC 15)
"I trust in you because at the end of the day, if you don't deliver, I get affected as well." (EOC 16)
“…[it affects] the wider team in the [EOC]…Because all things are related to each other.” (EOC 6)
EOC members took shared responsibility for the EOC:
“…everyone…[is] responsible. Everyone should be looking after each other…we have to share part of the blame.” (EOC 4)
“It affects wider…like I’ve said, the responsibility that is here on this post that I am putting…if one fails to perform, then it affects the whole [team].” (EOC 9)
226 “It impacts the whole team because of the reputation…cause what we’re meant to produce is not being done properly and it affects [the whole team].” (EOC 12)
“Even though we come here as individuals, you really develop a network with the people that you're involved with here. You almost become a little family. So you're always watching out for each other. So if something did happen...it would be difficult to say in some way that you wouldn’t be affected." (EOC 14)
“…it affects the whole team and it affects the project or the process.” (EOC 17)
The results show that regardless of a respondent’s perceived degree of interdependence, simply being a part of the EOC as a temporary system provided a baseline level of risk. This was due to the interdependence of tasks in the operation or the respondents taking a shared responsibility for the EOC delivering the requirements of the response. This meant that whether a respondent worked with another colleague directly or not, they still had a degree of interdependence due to their inclusion within the EOC. As such, considering swift trust, this research suggests that the degree of interdependence within the operation was at least moderate, if not high. This was due to the combination of: the perceived degree of interdependence for each respondent in their role, and the vulnerability of the whole EOC being affected by a problem. Again, this conclusion is supported by the greater reliance on evidence-based trusting criteria, which was an indicator that the degree of interdependence and subsequent level of risk was higher than what could be appeased by categorical assumptions alone.
Swift Trust Criterion #6 – The contractor The final criterion for swift trust relates to the leader of the temporary system who brings the team together (called the contractor). As shown in Section 4.6.1, for the EOC, the contractor and head of the operation was the Director (EOC 17). Swift trust theory argues that swift trust can develop between a respondent and their colleagues in the EOC based off the Director’s reputation. For swift trust, this is the only criterion that does not rely on a respondent’s colleagues but rather focusses on the role of the leader of the temporary system in allowing swift trust to form.
This section analyses how the reputation of the Director influenced trust formation in the EOC. Additionally, given the focus of this research on trust and international disaster management personnel, this section also explains how the reputation of the Director did not help trust form for the international respondents. Particular reference is made to the relationship between the EOC and OCHA.
227 5.6.1. How the Director’s reputation influenced trust formation To understand the influence of the Director’s reputation on trust formation in the EOC, the following question was asked of respondents that either reported to, or worked closely with, the Director (13 respondents in total): ‘Does the reputation of the Director help you to trust the other colleagues whom you have not met or worked with before?’. Of these respondents, eight said that the reputation of the Director helped them trust their colleagues, while five disagreed.
For those that said the reputation of the Director helped them trust their colleagues, there were several reasons why this was the case. Each respondent stated a particular attribute of the Director’s reputation that helped them to trust their colleagues. These were the Director’s:
Discernment:
“…because if the director trust them, so I have no reason not to trust them in work relations.” (EOC 7)
“I would see how director relies on [them] for certain information, refers people to [them]. Ok so, I will know ["that person's good"], yeh, I will need to get to know [them]." (EOC 11)
“Yeh, to some degree…[due to] the fact that he trust other people.” (EOC 16)
Judgement when selecting people to be included in the operation:
“Because the other people that are brought by the director, I’m sure they are the most qualified people to do the work the government is doing.” (EOC 3)
"I mean, look, for me, if you talk about the trust factor, just being able to, knowing that they're there with his [the Director’s] endorsement I think basically allowed that trust factor to be at the start of the conversation rather than being something that you worked towards." (EOC 14)
“Cause if he’s asking them to come and…if he trusts them, I can trust them.” (EOC 12)
Leadership and decision-making capabilities:
“Yes [their reputation helps]…Because I trust the Director’s decision.” (EOC 6)
228 “I think with the Director, one of [their] characteristics that I see um, evidently show, is how straight forward [they are] with people. If [they don’t] like what is happening, [they’ll] tell them.” (EOC 8)
EOC 16, a colleague who worked closely with the Director in Current Operations, provided a detailed example of how a local person external to the EOC, called Puji, was invited into Current Operations, and how trust was extended to them due to the Director’s endorsement.
"We all had no idea who he [Puji] was…a lot of us were hesitant in trusting him but the Director was ok with it and happy with what he was doing, so we all backed off and said, 'ok, we'll keep a cordial relationship'." (EOC 16)
As Current Operations were the information management hub of the EOC, and members of Current Operations were sensitive about sharing information, it was expected that these respondents would be wary of trusting a stranger. However, they trusted Puji due to the Director’s judgement; his endorsement was enough to create ‘cordial’ swift trust.
Of those that said the Director’s reputation did not help them form trust, three were the international participants in the EOC, and two local staff. (A discussion of the international staff and the Director’s reputation is provided in the next section). Considering the local staff, EOC 9 (a senior and permanent member of the EOC) said that they did not rely on the Director’s reputation (including his discernment, judgement and leadership capabilities) as a criterion for trust, because they wanted to judge people themselves. They felt that by applying their own trust criteria to a colleague, and not just extending trust because their boss did, they could protect the Director and EOC.
“…there needs to be an understanding between ah, I don’t like the people that you trust either. I cannot trust them just because of you [referring to the Director]…That way I am helping him because I just don’t think anyone who comes in will give him advice as a friend to the organisation.” (EOC 9)
The other local participant, EOC 15, simply stated that they wanted to know their colleague personally and did not rely on the reputation of the Director when forming trust.
Considering swift trust theory, the data showed that the reputation of the Director was only a factor for some of the respondents when forming trust with their colleagues. However, in conjunction with the other five aforementioned swift trust criteria, it can be concluded that the Director’s reputation alone was not enough for swift trust to form. As discussed, primarily in
229 Section 5.3, many respondents within the EOC required evidence in order to trust their colleagues. Accordingly, the data showed that the categorical assumption of the Director’s reputation was not a standalone factor for respondents within the EOC when forming trust. However, when considered in the context of disaster management, the results showed that the head of the EOC can encourage trust between their colleagues by; providing leadership and sound decisions, selecting appropriate personnel to be included in the response, and displaying discernment.
5.6.2. The Director’s reputation and its impact on international respondents extending trust………………………………. Of particular note, the results showed that all the international respondents within the operation (EOC 2, EOC 10, and EOC 13) said that the reputation of the Director did not help them trust their colleagues during the response. EOC 2 believed that the Director, although the head of the EOC, was somewhat ‘sidelined’ by Future Operations and the inclusion of the military into what should have been a civilian response. As such, they felt that the Director had a “function”, but the involvement of the military undermined the Director’s authority, and therefore, the Director’s reputation (including their endorsement) carried less weight.
“If I needed an answer on something, initially, [I would seek the] Director. After a while that wasn’t working…[the Director] got cut out of the loop…Because [of] the military takeover. And so the military run the operation…He [the Director] has a function, but [he] doesn’t make [the] decision.” (EOC 2)
EOC 10, another international respondent, similarly did not rely on the Director when forming their trusting relationships within the operation. Other results suggest that EOC 10 relied on honesty and sharing, particularly of information, when deciding who to trust (refer to Section 5.4.2). For them, the Director did not meet these criteria, and the Director’s reputation could not compensate for their failure to meet these more highly valued criteria.
“[The Director’s reputation]…makes no difference [when forming trust]…[the Director] is like a classic example of someone who does not give out information.” (EOC 10)
The data showed that this statement was not completely unfounded. The culture of information sensitivity and the breach of trust by OCHA early in the response seemed to have made the Director very cautious about whom they shared information with. The data revealed two examples of the Director’s hesitation to share information with internationals. Firstly, EOC 13, the third international respondent and a member of OCHA, was asked by the Director to write
230 a summary report about the response. In doing so, EOC 13 asked the Director to contact the respective government ministries for them in order to obtain the necessary inputs for the report. However, EOC 13 received resistance from the ministries and a lack of support from the EOC (even though EOC 13 was working for the Director to complete the report).
“I was there to write a report for [the Director]. So I asked [them] to have a look at the early draft and say, give me feedback. And I told [them] I needed [their] help, or the [EOC’s] help, in order to get information from the ministries. This didn’t happen. I was sort of stalled for a couple of weeks…[they] wrote an email saying someone would help me collecting information, but that person never followed up. So, from my point of view…it was an [EOC] report and they had to be engaged in it. It couldn’t just be written by me...[the Director] definitely was at some stage a little disappointed that…I didn’t get it done. And from my point, I explained to [them] there was a limit to what I could do without the [EOC] involvement…It’s a two way street.” (EOC 13)
The Director’s lack of urgency to aid EOC 13 (and ultimately, OCHA) in this process highlighted that at an organisational level, OCHA was not trusted. This was not an isolated incident: EOC 14, a local respondent and member of Current Operations, also stated that when they asked the Director to release data to the information managers in OCHA (whom EOC 14 had a working relationship with), the Director was reluctant. EOC 14 stated that the reluctance in sharing information was damaging; further validating that the Director was wary of trusting OCHA.
“OCHA has a large team behind them who could have provided some support. The perception of data not being used appropriately or being used outside of what it was intended for, I think, did more harm than good…So, having to negotiate with [the Director] to release a data set to an OCHA colleague for me was basically, really, it was a negotiation.” (EOC 14)
The Director’s decision to restrict OCHA’s involvement had consequences for the formation of trust in the broader EOC. As shown, many of the local respondents trusted the decisions and judgement of the Director. In essence, they trusted who he trusted. Therefore, the data would suggest that the conscious exclusion of OCHA (and also The United Nations Disaster Assessment and Coordination (UNDAC)) from the response by the Director may have been a catalyst for other members of the EOC to also keep OCHA at arms-length; as they trusted the Director’s decisions. As discussed in Section 5.3.2, the local staff wanted to know the role of their international colleagues in the operation, and to see them humbly integrate into the nationally led response. Furthermore, as explained in Section 3.6.2, this caution of the international
231 community was also a by-product of the recent response to Cyclone Pam in Vanuatu; where it was perceived by some of the senior local respondents in the EOC that the international community took over the response and excluded the local leadership. Therefore, although EOC 2, EOC 10 and EOC 13 as international respondents were included in the EOC, the reputation of the Director did not contribute to their formation of trust with others in the EOC. More broadly, the Director’s decision to resist the international community (particularly OCHA) from helping in the response, showed that erosion of group inclusion can also detrimentally impact trust formation.
In summary, the data suggests that the reputation of the Director was a contributing factor for some local respondents when forming trust within the EOC. However, even the local respondents required additional evidence when forming trust. Additionally, the Director’s hesitancy to share information with OCHA was one reason why his reputation was not a contributing factor for the international respondents when forming trust.
Discussion of Research Proposition 2 This section has two purposes, related to the results explaining research proposition two. Firstly, it will address research proposition two by explaining how trust formed within the EOC in accordance with swift trust theory. Secondly, it will explain the implications of these conclusions on the theory of swift trust in temporary systems, as well as the wider disaster management literature. This will be done through a discussion of each of the six swift trust criteria previously analysed:
1. Reputation,
2. Interactions around role,
3. Categorical assumptions versus evidence-based information,
4. Confirmation of assumptions,
5. Moderate interdependence, and,
6. The contractor.
Finally, it will conclude with an analysis of whether the trust formed in the EOC was indeed swift trust.
232 5.7.1. Reputation The results showed that the respondents considered their own reputation (professional and/or personal) when deciding how to behave whilst in the EOC. Similarly, the respondents believed that their colleagues considered their own reputations when deciding how to behave throughout the operation. This is significant as it provides evidence validating swift trust theory: that the members of the EOC (a temporary system) would not expect untrustworthy behaviour from their colleagues given the reputational damage that would be experienced by the perpetrator, should that untrustworthy behaviour be shared. Furthermore, as shown in Section 4.4.2, due to a small labour pool and overlapping networks, all the respondents within the EOC expected to work with their colleagues again in the future. As such, any reputational damage may not just be isolated to one operation; but extended to future interactions. These results are consistent with what swift trust theory would predict of a temporary system.
Secondly, the results also showed that there were two recorded breaches of trust, with the reputational damaged suffered by the perpetrators proportional to the number of people impacted by that untrustworthy behaviour. Lastly, although reputation protection was a significant contributor to the development of trust, some respondents also felt being a member of the civil service and having a desire to help the response provided further reasons as to not expect untrustworthy behaviour from their colleagues. These validate, to some extent, swift trust theory’s concept that categorical assumptions may serve as a basis for trust formation.
It is important to recognise that, in the context of the broader literature on trust, these results provide an alternate perspective on the importance of reputation in trust formation. Much of the literature, including Rousseau et al. (1998), Mishra (1996) and McKnight et al. (1998), argue that trust can be awarded based on someone’s reputation; regardless of whether the trustor has personally meet the trustee. Relating this argument to swift trust, it could be said that one might not expect untrustworthy behaviour from a person who has a trustworthy reputation. However, this case study provides a unique analysis of reputation from the perspective of the trustor. The results validated the work of Meyerson et al. (1996) by showing that the respondents considered their own reputation (either professionally and/or personally) whilst working in the EOC. Additionally, the respondents also thought that their colleagues were doing the same. This culminated in the expectation from the respondents that they would not be undermined by their colleagues due to the damage this would have on that colleague’s reputation; given they expected to work with some of the same colleagues again in a future capacity. Thus, this case study highlights an alternate perspective concerning the importance of the individual maintaining their personal reputation; and the subsequent impact this has on
233 trust formation. In summary, these results confirm Meyerson et al’s (1996) proposition that reputation is a key criterion in trust formation, but add a second dimension to this proposition: the data shows that trust can be impacted by how reputations are perceived from the perspectives of both the trustee and the trustor.
Additionally, the case study adds to the wider disaster management literature by confirming the importance of maintaining a favourable reputation in order to be trusted during a national operation. The breach of trust by OCHA provides a contemporary example of the work of Stephenson Jr (2005, p. 344), which states that, “while participants may trust a participant at the start of their relationship in a relief situation, a single miscue can foul this opportunity for the remainder of the emergency and certainly for organisational relations in the next.” As discussed, the results showed that The United Nations Disaster Assessment and Coordination team (UNDAC) were excluded due to their affiliation with OCHA and the United Nations, who had breached the EOC’s trust. Furthermore, the Australian Defence Force were also impacted by Future Operations limiting the distribution of information to them as an international organisation involved in the response. The case study therefore broadens the work of Stephenson Jr (2005) and Alvinius et al. (2010) by showing that reputational damage and swift distrust can extend to another organisation by association; reputational damage may not solely affect the perpetrator. This example also adds to swift trust theory (Meyerson et al., 1996) by highlighting the importance of international organisations adherence to the governance of the nation they are deployed to. This is important should they wish to remain a trusted partner; given that swift distrust and reputational damage can occur from only one breach of trust.
Significantly, this research also provides a unique contribution by providing evidence of ‘emotional contagion theory and the ripple effect’ (Barsade, 2002) impacting trust formation within a disaster management context. That is, the breach of trust by OCHA impacted the ‘mood’ of locals within the EOC toward trusting this organisation, which influenced, “the moods and then the judgements and behaviours of others”; particularly other local disaster management personnel. It is recommended that further research be undertaken to understand how the ripple effect and emotional contagion theory can impact trust formation in a disaster management context.
Finally, the results documenting the EOC’s distrust in OCHA provide an original contribution concerning the role of sanctions in a national response. Within the literature review, it was shown that the use of sanctions (such as contracts and terms of agreement) can help trust develop (Coleman, 1990). That is, trust can exist between personnel within an organisation
234 because there are consequences for breaking such trust; such as losing one’s job, etc. However, this research explores, ‘whether such sanctions exist during the recovery from a disaster, and if so, how these sanctions impact trust between local and international disaster personnel’. The case study linked the trust and disaster management literature by showing that such sanctions do exist in a national response. These sanctions were described by a representative from OCHA;
“…because of trust issues, there was a real decision, conscious or otherwise, it was conscious, but it wasn’t necessarily an official decision, but there was a pattern of behaviour that was to keep OCHA at bay.” (EOC 13)
The data showed that within a national operation, sanctions may not be formalised (taking the form of contracts, documents or agreements that would apply in an organisation), as suggested by McKnight et al. (1998). Rather, the ad hoc nature of an EOC as a temporary system seems to minimise the power of such formality; instead, behaviour was influenced through the creation of informal sanctions. The deliberate exclusion of OCHA from the operation showed that informal sanctions existed, such as unofficial decisions being made by those in power to exclude a member/organisation from the operation, or the risk of reputational damage within the network of disaster management organisations. The results showed that breaches of trust, and the subsequent imposition of informal sanctions, can lead to a member being excluded from the temporary system and gaining a reputation for being untrustworthy.
5.7.2. Interactions around role Within the EOC, the case study showed that a colleague’s role, and their respective role behaviour, were factors contributing to trust development. Respondents stated that there was a lack of clarity around their colleagues’ roles; however, contrary to swift trust theory, this did not affect the formation of trust within the EOC. Similarly, colleagues taking on additional tasks outside of their assigned role did not hinder trust formation. The data showed that although role may play an important part in the development of swift trust (because it acts as a substitute for more specific trusting evidence about a colleague), this was not universally the case for the EOC; as would be assumed by swift trust theory. Rather, the data showed that there were additional factors important in the development of trust; such as group inclusion and personality. Except for those in positions of leadership/Current Operations, the respective importance assigned to knowing a colleague’s role, and them adhering to that role, appears to be less than what Meyerson et al. (1996) and swift trust theory predict.
These findings have multiple implications for swift trust and the wider trust literature in relation to role clarity and role behaviour in a disaster management context. Firstly, the results
235 contradict the conclusion made by Curnin et al. (2015, p. 6) who espoused, “the importance of role clarity in the rapid formation of temporary organisations.” Although the authors’ research is built upon the theoretical framework of swift trust theory and role clarity applied to an EOC, their results do not provide a connection between role clarity and swift trust development. Rather, this connection appears to be assumed in their research; that is, if role clarity is important in an EOC, then it must also help swift trust form. However, this case study of Fiji has shown that in an EOC, although role clarity may be important for other reasons, the respondents were less concerned with using role clarity as a criterion for trust formation. Rather, the respondents stated that they either:
1. were more focussed on whether their colleagues met the requirements of the response, which could include taking on tasks outside of their assigned role; or,
2. started from a position of trust (that is, that trusting disposition was independent of role clarity).
This first finding extends swift trust theory and the argument of Meyerson et al. (1996, p. 191), who stated that swift trust, “is not so much an interpersonal form [of trust] as it is…[an] action form [of trust].” This ‘action form of trust’ can be defined as the respondents undertaking their assigned roles in the temporary system. However, the first finding demonstrates that the members of the EOC (excluding some of the senior leadership) were less concerned about their colleagues undertaking their assigned roles than effectively completing tasks as part of the response, even if those tasks were emergent. As Turoff, Chumer, Van de Walle, and Yao (2004, p. 15) state, “in a crisis it is never certain who will take on which role or which combination of roles. It is expected that people will be trained to be qualified in a number of different roles.” Similarly, these results support the Majchrzak et al. (2007) study of swift trust in emergent search and rescue groups, in which trust developed through action and getting things done. As stated however, the data did show that those in senior management uniquely considered role behaviour as a trust criterion, due to their concern for protecting the EOC. This is an important conclusion as it shows that there are different priorities assigned to role behaviour in an EOC based on the seniority of the respondent in the operation.
The second bullet point above, concerning a respondents’ trusting disposition, validates the work of Luhmann (1979) and Wiethoff and Lewicki (2005), who found that people more readily start a relationship by trusting, rather than distrusting, the other party. That is, trust is not just based upon the qualities of the trustee; rather, it also depends on the trustor (B. Barber, 1983; Gambetta, 2000; Hung et al., 2004; Lewicki & Tomlinson, 2003; McKnight et al., 1998; Tatham &
236 Kovács, 2010). As Mayer et al. (1995) highlight, a person’s disposition to trust is an important consideration concerning trust development.
The results also refined Meyerson et al’s (1996) use of Gambetta’s (2000) theoretical foundation for swift trust. Drawing on Gambetta’s (2000) ideology of trust and uncertainty, swift trust theory argues that, “in an effort to avoid uncertainty, the person is likely to be more trusting or more distrusting than the data warrant, simply in the interest of reducing uncertainty and getting on with the task” (Meyerson et al., 1996, p. 177). However, Meyerson et al. (1996) go on to argue that this desire to reduce uncertainty, and the consequential increased willingness to trust, is in response to a lack of evidence-based information. The theory argues that roles would not be overlooked when a respondent considers trust. However, only a small number of respondents in positions of leadership and in Current Operation used role behaviour and role clarity as a consideration when forming trust. Overall, the data showed that a lack of role clarity or role behaviour did not inhibit trust formation for most respondents. These findings provided an original contribution by refining the use of Gambetta’s (2000) theoretical model of uncertainty in swift trust formation in a disaster management context. That is, in reducing uncertainty and promoting trust within a disaster management temporary system, role was not as crucial as swift trust theory would argue.
Finally, the case study also provides a unique contribution to the literature by highlighting multiple examples of ‘the informal organisation’, as considered in new institutionalism. These include the EOC demonstrating a lack of formal role clarity, such as members taking on additional tasks outside of their expertise, and working outside formalised hours. These findings support the research of Lalonde (2004), who argues that resourcefulness and autonomy are critical commodities when operating in a disaster management context. Similarly, the existence of informal sanctions enforced against OCHA provides a new example of how information dissemination was an implicit EOC organisational value, that was not fully appreciated by OCHA as an external organisation (Helmke & Levitsky, 2004). As Zucker (1983, p. 5) writes, “…participants are not conscious of their central values and…common understandings are seldom explicitly articulated.” Therefore, it is recommended that further research be conducted to consider the applicability of new institutionalism to an EOC, and whether this theoretical approach can provide additional insights into trust formation within a disaster management context.
237 5.7.3. Categorical assumptions versus evidence-based information The TC Winston case study showed that role, competency and personality were the three primary trusting criteria used by the respondents within the EOC. Other less cited criteria included work ethic, being a civil servant, appearance, or a colleague’s parent organisation. The results also showed that two unique categorical assumptions, related to culture and the military, influenced trust formation for some members (particularly locals) in the EOC. Contrary to swift trust theory, the results showed that respondents, and particularly those in positions of leadership and expertise, favoured accuracy over speed (that is, evidence over categorical assumptions), when determining who to trust in the operation. Additionally, the data showed that local EOC personnel applied a unique set of trusting criteria to their international colleagues (i.e. their willingness to display humility when integrating into the EOC and work under the national EOC leadership). In reference to the wider literature on trust and disaster management, this section will first address the trust criteria adopted by the respondents. It will then discuss the unique categorical assumptions of the culture and military. Finally, it will consider the importance of maintaining a locally led national response, with the international community working under local leadership.
Most significantly, this case study provides a unique contribution to the disaster management and trust literature by detailing a comprehensive list of the factors that impact trust formation between disaster management personnel working together in an EOC. Secondly, in exploring the trust criteria used within the EOC, it became clear that contrary to Meyerson et al. (1996) and swift trust theory, the respondents did not rely more on categorical assumptions than evidence-based information when forming trust. Although Fahy (2012) found role was used by many respondents as a swift trust criterion in response to a crisis, the TC Winston case study showed that in the temporary system of the EOC, this criterion was not as significant as swift trust theory would predict. Rather, the respondents primarily relied on their colleagues’ competency and personality as trust criteria.
Breaking this down further, it was found that the results of this case study aligned with the research on trust in broader organisational theory, which argues that trust is influenced by three criteria: competency (ability), integrity, and benevolence (Ferrin & Gillespie, 2010; Lewicki & Tomlinson, 2003; Pirson & Malhotra, 2011; Rousseau et al., 1998; Saunders et al., 2014; Wiethoff & Lewicki, 2005). This is significant as the organisational literature does not specifically consider trust formation in temporary systems, such as that of an EOC. The similarity of trust criteria prominent in the EOC (a temporary system) with the trust criteria in non-temporary systems
238 (such as those in the organisational literature) suggests that swift trust theory may overstate the function of a temporary system in creating conditions conducive to trust formation.
Considering the three prominent organisational trust criteria, within the EOC, competency (ability) was coded 21 times; validating the work of Parsons (1963) and Coleman (1990) in which they argue trust can be dependent upon the capability of the other person. Furthermore, it was found that the data coded as personality (23 counts within the EOC) was able to be broken down into the organisational trust criteria of integrity and benevolence. Respondents stated that trust formed due to:
Integrity: A colleague’s openness and honesty (Mishra, 1996; Rousseau et al., 1998), personal motivations (Pirson & Malhotra, 2011), and their ability to live up to their word.
Benevolence: Whether a colleague was considerate, kind, sharing and generous, greeted the respondent, and did not gossip. It is important to recognise that these findings contribute to the trust literature by providing specific behavioural actions associated with the term benevolence, and help refine the high-level definitions of this term in the trust literature (such as that of Mayer et al. (1995)).
Furthermore, in finding that respondents relied more on personality than competency as a trust criterion, the data contradicts the findings of Mayer et al. (1995), Lewicki and Tomlinson (2003) and Pirson and Malhotra (2011) who argue that competency and integrity should be more important than benevolence as trust criteria early on in a new relationship. Similarly, the trust criteria used by the respondents also refines the work of McKnight et al. (1998) who state that, specifically in new organisational relationships, in the absence of knowing a colleague’s reputation, broad stereotyping can impact trust development. Although it was found that a colleague’s role was used as a categorical assumption, the respondents did not mention any other broad stereotypes as trusting criteria - except appearance; which had only five recorded counts out of a total of 106. Rather, the data suggests that the respondents formed trust using a combination of both categorical and evidence based information, and did not rely on stereotyping as much as may have been predicted by McKnight et al. (1998).
Apart from the main trusting criteria of role, competency and personality, the data also revealed that Fijian culture and the involvement of the military impacted trust formation for some of the respondents (primarily the locals). This confirms the research of Gambetta (2000), who argues that trust is contextual and situationally specific. That is, trust can be influenced by the wider
239 context (Lewicki et al., 1998) and social similarities (Kramer & Lewicki, 2010; MacDuffie, 2011). Discussing the influence of culture on trust formation, Zolin (2002, p. 5) claims that, “international temporary teams are not always so successful in developing swift trust, due to…differences in culture”. This case study found that, although culture did not help trust develop between the local and international personnel within the EOC, contrary to Zolin (2002), it did play an important part for some locals in developing trust with one another.
Finally, the case study also revealed one final consideration concerning the involvement of international respondents in a national emergency. Several of the local respondents stated that they were looking for their international colleagues to integrate into the EOC with humility; by working alongside and under the national EOC leadership. This provided a modern example for the argument of Hilhorst (2005, p. 359), which emphasises that, “…the need [for international personnel] to be accountable to local people…is perhaps equally important in the creation of trust. Working with local partners…having a participatory approach…and showing respect for local culture…can all contribute to the development of an environment of trust.” Furthermore, these findings support a wider part of the disaster response literature which argues that local capacity building is essential to disaster management. This capacity building can occur through the use of local expertise (M. Perry, 2007), local leadership (Lam, Lin, Tsai, & Chiu, 2010), and ensuring that the locals engaged in the response and not sidelined by international involvement (Hilhorst, 2005). The rapid, locally led response by Fiji after TC Winston counters the assumption that international response leadership is always necessary after a natural disaster in a developing nation; or even that it is the most helpful method of response. As argued by Berkes (2007) concerning the success in Samoa, it is imperative that developing nations have locally led response operations. Furthermore, this case study also showed that local personnel were actively seeking humble integration of their international colleagues into the operation. This reinforced the importance of international disaster response personnel being conscious of their involvement in another country’s operation; particularly given that there has sometimes been, “…little consideration displayed by many international actors for the specific context in which they [operate]” (Thévenaz & Resodihardjo, 2010, p. 16).
5.7.4. Confirmation of assumptions As with this swift trust criterion, the literature recognises the significance of behaviour as a method of validation for the trustor concerning their original trusting intention toward the trustee (Crisp & Jarvenpaa, 2013; Hung et al., 2004; Lewicki et al., 2006; Luhmann, 1979; Mayer et al., 1995). That is, the trustor can determine whether their initial trusting intention was appropriate based upon the behaviour of the trustee; and therefore decide whether to continue
240 trusting the trustee in the future (Mayer et al., 1995). However, what is unique to swift trust theory is that the trustor is assumed to have relied upon categorical assumptions when forming trust; given the ad hoc nature of the temporary system. Therefore, the theory argues that the trustor should have a heightened urgency in wanting to validate these assumptions, given the lack of evidence-based information available to them from which they may form swift trust. However, contrary to swift trust theory, within the EOC, the respondents relied upon evidence- based trusting criteria more than categorical assumptions. This subsequently reduced the relevance of a respondent needing to confirm their assumptions, as they were relying on trust criteria that were revealed over time (work ethic, personality, competency, etc.)
An analysis of the results provided some important clarifications for swift trust theory concerning how members of a temporary system confirm their trusting assumptions. Refining the work of Meyerson et al. (1996), the case study found four significant results for the EOC:
1. Only three respondents actively sought to validate their trusting criteria, and the motives for doing so varied between wanting personal protection and seeking to protect the EOC. This finding is unique, as it provides a range of motivations for someone wanting to validate their trusting criteria. This motivational distinction is lacking in swift trust theory.
2. A respondent does not weight all swift trust criteria equally.
3. Although a respondent may have experienced unexpected behaviour from a colleague in the operation, the bearing this had on the formation of trust was connected to the value ascribed to the trust criterion impacted by their colleague’s behaviour.
4. It may be easier to determine which trusting criteria are more important to a respondent by analysing examples of swift distrust rather than examples of swift trust.
Although each of these findings are unique contributions to swift trust theory, and apply the research of Meyerson et al. (1996) to a disaster management context, they also contribute more broadly to the literature of trust in disaster management. Multiple authors, such as Kumar and Paddison (2000), White and Eiser (2006), Chandrasekhar (2010, 2012), as well as The Emergency Capacity Building Project (2007) provide a comprehensive set of criteria for trust in disaster management. However, none of these authors provide weightings or an explanation behind the ascribed value of each trust criterion. This case study however has shown that, although intuitively assumed, not all trust criteria are weighted the same by a respondent. Therefore, it is recommended that further research be undertaken to understand, of the trust criteria
241 employed by an individual, which are the most important when forming trust in an EOC (subject to a national response).
5.7.5. Moderate interdependence Swift trust theory argues that swift trust is most likely when there is moderate interdependence between the members of the temporary system. This is because with moderate interdependence comes moderate risk of being undermined. Conversely, if the degree of interdependence is too high (and hence the degree of vulnerability for a respondent is also high), the use of categorical assumptions in forming swift trust does not provide the respondent with sufficient evidence to take such a risk; slowing the formation of swift trust. However, it must be stated that this proposition from swift trust is, as the authors state themselves, “our hunch…there are no certainties anywhere in these calculations, only implicit probabilities” (Meyerson et al., 1996, p. 175).
Therefore, considering the authors suppositions, the case study provided several unique clarifications to the concept of interdependence and the application of swift trust within an EOC:
The results found that the degree of interdependence within the operation was at least moderate, if not high. Swift trust theory argues that categorical assumptions provide insufficient evidence for a person to trust a colleague if they have a high degree of interdependence. The data validated this conclusion; given that all those with a high perceived degree of reliance on their colleagues used evidence-based information more than categorical assumptions when forming trust in the EOC.
It was found that the degree of interdependence of a respondent within the temporary system was a function of both the number of colleagues a respondent was connected to within the EOC, and how reliant a respondent was on these colleagues.
Regardless of a respondent’s perceived degree of interdependence, simply being a part of the EOC as a temporary system provided a baseline level of risk. This was due to one of two reasons: the interdependence of tasks in the operation, or a respondent taking a shared responsibility in the EOC delivering the requirements of the response.
The culmination of these findings is significant because they further clarify the work of Meyerson et al. (1996) and the concept of interdependence within a temporary system. That is, the interdependence of a member of a temporary system is a function of:
a member’s direct working relationships with their colleagues,
242 how reliant a member believes they are on their colleagues within the system, and,
the member’s inclusion in the system.
Addressing this last point specifically, the results also strengthened the argument that trust formation could be impacted by group inclusion; as predicted by Meyerson et al. (1996). The evidence suggests that the members of the EOC took responsibility for the operation, and were aware of how being a member of the EOC increased their degree of interdependence. These findings further validate the work of McKnight et al. (1998) and M Williams (2001), who argue that holding a common purpose and being identified with a group can lead to the members of that group perceiving each other as more trustworthy than people not in their group. Likewise, Kramer and Lewicki (2010) and MacDuffie (2011) argue that people are likely to extend trust to others who are in the same ‘in group’ or organisation as them. Put simply, Brewer (1981, p. 356) states there can be, “depersonalised trust based on category membership alone”. Similarly, focussing specifically on swift trust, the results validate the conclusions of Popa (2005, p. 79), who said that, “people trust others with whom they work (temporarily) because they perceive their group members as trustworthy”.
5.7.6. The contractor The case study showed that, contrary to Meyerson et al’s (1996) theory, the categorical assumptions regarding the reputation of the contractor (in this case, the Director of the EOC), were not substantial factors for the respondents when forming trust. Although no respondent said that they expected to be undermined by the Director, the Director’s reputation as the leader of the EOC was less influential than swift trust theory would suggest. This is because the respondents relied more on evidence-based information as opposed to categorical assumptions. Furthermore, the case study suggests that the theoretical model of an intermediary, as proposed by Coleman (1990), seems less applicable for disaster personnel within an EOC. This was particularly true for the international respondents; none of whom stated that the reputation of the Director helped them when forming trust in the operation.
In contribution to the disaster management literature more generally, the results did find that the head of the EOC can encourage trust between their colleagues by providing leadership and sound decisions, displaying discernment, and selecting appropriate personnel to be included in the response. These findings offer a contribution to the literature by providing evidence for specific leadership variables in a disaster management setting, which, as Schaubroeck, Lam, and Peng (2011) argue, can promote trust among colleagues and toward the leader. In particular, the last criterion (concerning the selection of personnel) supports the research of Burke, Sims,
243 Lazzara, and Salas (2007, p. 615), who argued that trust in leaders is impacted by, “the degree to which leaders are able to manage team composition to ensure the best combination of skills and knowledge exist within the team.” However, in the EOC, only some of the respondents were specifically selected for inclusion in the operation by the Director. One reason for this was due to the temporary system not being made from scratch; given the EOC had permanent staff prior to the operation. Therefore, further research is required to understand the degree to which the head of an EOC is able to actually manage team composition; and the subsequent impacts this has on trust formation.
Did swift trust exist in the EOC? Addressing research proposition two, the case study suggests that the trust exhibited by the respondents within the EOC was not accurately modelled by swift trust. Against the theoretical model proposed by Meyerson et al. (1996), the results showed that, regarding:
Reputation – The respondents did consider their reputations when deciding how to behave within the EOC.
Interactions around role – The respondents primarily interacted around roles, however role clarity and role behaviour were less important than other criteria when determining trust.
Categorical assumptions versus evidence-based information – Although categorical assumptions were used when forming trust (such as role, appearance, culture) to an extent, the respondents relied more on evidence-based information (such as competency, personality, and work ethic).
Confirmation of assumptions – Given the respondents’ reliance on evidence-based information when forming trust, it was less relevant for them to confirm their categorical assumptions about their colleagues.
Moderate interdependence – The EOC was characterised by a moderate/high degree of interdependence.
The contractor – The reputation of the Director, as the contractor, did not play a significant role in the formation of trust for the respondents.
The case study showed that the respondents’ trust criteria, and their approach to trust, was only partially consistent with swift trust theory. However, an overall analysis of the results did suggest
244 that the respondents trusted each other. Furthermore, given the time constraints of the national operation, the data would suggest that this trust was developed quickly. However, this conclusion raises a question concerning the use of the word ‘swift’ within swift trust theory. That is, this case study highlights an important factor which Meyerson et al. (1996) do not address as part of swift trust theory: ‘how long is it before the trust forming in a temporary system is no longer ‘swift’?’
Within their research, Meyerson et al. (1996) provide a list of different groups considered to be temporary systems, including (as examples): presidential commissions, senate selection committees, theatre groups, construction teams, auditing teams, juries, film crews, cockpit crews, paramedics, and humanitarian outreach workers. However, every one of these temporary systems has a lifespan. Thus, it is unclear as to what the timeframe is for swift trust to form, before it transitions into trust as it is more conventionally known. Furthermore, given that trust formation is dynamic and personally subjective, it appears that no universal timeframe could be reasonably allocated to when this transition should occur.
A second question that also weighs into this argument is, ‘how much evidence-based information is required before trust is no longer swift trust; because the respondent is not relying predominantly on categorical assumptions but is making an informed decision about trusting their colleague?’ This is not specified by Meyerson et al. (1996) and is yet to be addressed by the swift trust literature.
Therefore, considering this case study, the greater reliance on evidence-based information within the EOC, and the conformity of the respondents’ trusting criteria to that of the wider organisational trust literature, makes it difficult to separate swift trust out as a distinct feature of trust formation within the EOC. That is, within the EOC, it appears that trust formed quickly between the respondents, however it was not theoretically ‘swift trust’ as proposed by Meyerson et al. (1996), as it did not conform to each of Meyerson et al’s (1996) swift trust criteria.
Lastly, the results also showed that even though something may be deemed a temporary system, how trust forms in that system may not always be modelled by swift trust. This opens the possibility to additional models of trust being applicable to temporary systems (other than the model offered by Meyerson et al. (1996)); such as McKnight et al’s (1998) theoretical model of ‘Initial trust formation in new organisational relationships’, which is grounded in the organisational trust literature. Other trust models that could be applied to an EOC should
245 consider the importance of a colleague’s role, competency, integrity, and benevolence, as well as factors unique to a disaster management context on trust formation.
246
ANALYSIS OF RESULTS RESEARCH PROPOSITION 3
This chapter explores research proposition three; which considers whether the seven swift trust resilience factors impacted the resilience of trust in the EOC; even though this trust was theoretically not swift trust. First it provides an analysis of the results for each of the seven swift trust resilience factors. It then provides a discussion of these factors with reference to swift trust theory, as well as the wider trust and disaster management literature. Finally, it shows that trust in the EOC was impacted by some of the resilience factors applicable to a temporary system. However, through a paradigm of social constructionism, it argues that to investigate the resilience of trust appropriately means to also consider the subjective, personal and relational aspects of trust.
Research Proposition 3: Swift trust theory resilience factors
Although it was found that the EOC did not exhibit swift trust as per the theoretical model proposed by Meyerson et al. (1996), this research still sought to analyse the factors that the authors argued would influence the resilience of swift trust. This was important for two reasons. Firstly, the swift trust resilience factors are dependent upon the characteristics of a temporary system. Therefore, because the EOC qualified as a temporary system, it was necessary to consider whether these resilience factors were applicable. Secondly, the resilience factors of swift trust theory are not only beneficial in helping understand how swift trust forms and how trust is sustained. Rather, they also highlight additional aspects which are important for the wider trust and disaster management literature, including: the reputation of the leader of the
247 EOC, a person’s predisposition to trust in an EOC, the assumed qualifications of colleagues, and the importance of improvisation in disaster management.
Consequently, this chapter will consider the resilience of trust in the EOC in terms of the seven resilience factors, which Meyerson et al. (1996, p. 185) define as, “proxies for history – [that] might contribute to the development of trust in temporary systems...[and help it be] sustained throughout the life of the temporary system.” For clarity, each resilience factor is explained again:
1. The reputation of the contractor: The reputation of the contractor for bringing together successful temporary systems can substitute for a lack of interpersonal history between members. Furthermore, it is assumed that the contractor has used certain evaluative criteria in order to select who to include in the temporary system. Therefore, if the contractor has a reputation for bringing together a temporary system, and makes clear to the temporary system what criteria were used for someone’s inclusion in the system, swift trust can be more resilient.
2. Optimism: People within a temporary system expect, “better things to happen to them compared to the average group member” (Meyerson et al., 1996, p. 186). In other words, members of the EOC would expect an operational problem (delays in supplies, flow-on effects from inaccurate reporting of information in the situation reports, incorrect aid deliveries due to inaccurate surveys, etc.) to impact another colleague and not themselves. This attitude means that members are optimistic, which in turn makes swift trust more resilient.
3. A presumption of trust: Within a temporary system, even though trust is presumed, members can act as if trust were present. When a member observes a colleague within the temporary system acting as if swift trust were present, this affirms that member’s original assumption that swift trust is appropriate. This personal confirmation of swift trust being appropriate, initiated by their colleague’s behaviour, increases the resilience of swift trust.
4. Assumed qualifications: It is assumed that each member’s inclusion in the temporary system is based upon their expertise and experience; which has been gained elsewhere and does not require verification. If each member of the temporary system believes the other members to be qualified in their roles, then swift trust may be more resilient.
248 5. Hedges: The resiliency of swift trust can be altered due to members of the temporary system creating hedges. A hedge is defined as a ‘way out’ in order to, “reduce the perceived risks and vulnerabilities of trust by reducing interdependence and thus its perceived costs” (Meyerson et al., 1996, p. 187). Simply, a hedge in a temporary system reflects an attitude where one member trusts another, to an extent. Hedges can also lead to distrust if the other member(s) of the temporary system find out that those hedges exist. They can also result in overconfidence by the person who created the hedge, and to reduced group commitment.
6. Improvisation: Given temporary systems are formed to specifically address complex tasks in a set time, it is expected that things, “almost inevitably must lapse or break down on occasion” (Meyerson et al., 1996, p. 189). Consequently, swift trust may be more resilient in temporary systems that improvise solutions to problems because the members feel that the system is able to surmount inevitable setbacks.
7. Focusing on the task: As temporary systems operate with time constraints, they simply do not have time to develop harmful interpersonal and group dynamics that can erode trust. Rather, members are focussed on the task and ensuring its completion, which can make swift trust more resilient.
Each of these factors and their impact on the resilience of trust in the EOC are discussed.
The reputation of the contractor Swift trust theory argues that the contractor’s, “reputation for creating and composing successful temporary groups…serves as a useful substitute for interpersonal history” (Meyerson et al., 1996, p. 185). However, as previously explained under Section 4.6.2, the Director, as the contractor for the EOC, did not assemble the respondents for the operation. Rather, the Director specifically asked only four personnel to join the response (EOC 3, EOC 13, EOC 14, and EOC 16). The remaining respondents were:
asked into the EOC by other senior members of the response,
disaster management experts that reported to the EOC of their own accord, or,
were sent to the operation by the Australian Government.
249 Furthermore, as shown in Section 5.6.1, only eight local respondents considered the reputation of the Director when forming trust in the EOC. Particularly, only three commented on the Director’s judgement in selecting people to be included in the operation.
“Because the other people that are brought by the director, I’m sure they are the most qualified people to do the work the government is doing.” (EOC 3)
"I mean, look, for me, if you talk about the trust factor, just being able to, knowing that they're there with his [the Director’s] endorsement I think basically allowed that trust factor to be at the start of the conversation rather than being something that you worked towards." (EOC 14)
“Cause if he’s asking them to come and…if he trusts them, I can trust them.” (EOC 12)
The remaining five respondents who considered the Director’s reputation focussed on his discernment, leadership, and decision-making capabilities.
Additionally, unlike other temporary systems, the imposed time pressures of a national operation can hinder the opportunity for the head of an EOC to develop a reputation of creating a successful response team. This is because in Fiji, as with many other nations, the EOC is manned by a permanent, skeleton staff ready to engage in a national operation should it arise; thereby limiting the role of the Director in selecting personnel to be included in any forthcoming operation. As shown in Table 3.2.a, seven of the respondents within the EOC were permanent staff.
Even so, this research does recognise that the Director had a limited role in selecting the remaining 11 surge staff that were brought in to aid the response (personally selecting four – refer to Table 4.6.a). This was in accordance with National Disaster Management Plan, where it states that the Director of the EOC;
“Coordinates and manages the [EOC], including arrangements for staffing and equipment.” (NDMP, 1995)
Consequently, the results showed that the majority of respondents were not selected by the Director, and that the Director did not have a reputation for ‘composing successful temporary groups’. This ultimately meant that neither of these two factors were able to act as ‘substitutes for interpersonal history’ between the respondents and did not contribute to the resilience of trust in the EOC.
250 Secondly, swift trust theory also states that it is important that the contractor conveys the selection criteria used for a person to be included within the temporary system. However, within the EOC, no respondent said that they trusted their colleagues because they were actively aware of the criteria the Director used to include a colleague within the operation. And although the Director themselves said that they did use criteria, they did not say that they disseminated these to their colleagues in the EOC.
“…[before someone comes into the operation] I’d like to know the person more in detail, his or her background, and that’s what I do for everyone who comes inside. I ask, I ask for like a profile, once they give that to me with their experience, then we sit down.” (EOC 17)
Consequently, the criteria adopted by the Director for a colleague’s inclusion within the EOC did not help develop or sustain trust within the EOC; and therefore did not contribute to the resilience of trust.
These results are significant to recognise for three key reasons:
1. Although the EOC was a temporary system (in accordance with swift trust theory), it was not formed conventionally by the contractor (Director) asking people to join the group. Rather, in the field of disaster management, the data shows that EOCs (as temporary systems) are characterised by two unique factors:
o A core response team comprising permanent EOC staff.
o Additional surge staff, that are brought into the EOC to bolster the capacity of the EOC throughout the response. Furthermore, these surge staff can be asked into the operation by personnel other than the head of the EOC.
2. Swift trust theory does not specifically comment on the formation of a temporary system built around a pre-existing core group of people (such as the permanent staff of the EOC). This is an important consideration because only new personnel may undergo the selection process imposed by the contractor for inclusion into the system. That is, the inclusion criteria adopted by the contractor may only be applied to the new members coming into the temporary system. Consequently, adapting this to the EOC, this has an impact on trust formation and its resilience as the permanent staff included prior to the response are able to rely on the contractor’s selectivity as a ‘substitute for interpersonal history’ in relation to the surge staff brought into the operation. However,
251 any new surge staff cannot; given it is not clear what criteria were applied to the permanent staff when they were originally employed to work in EOC. Therefore, the results show how the EOC is formed (and temporary systems in general) can impact the influence of the contractor’s selectivity on trust formation and the resilience of such trust.
3. Although the Director of the EOC had selection criteria that they used to decide whether to include someone within the operation, these were not made clear to the wider EOC. Therefore, in accordance with swift trust theory, the data shows that only having inclusion criteria is insufficient in making trust more resilient. Further research is required to identify whether the selection criteria of the head of the EOC can make trust more resilient. Additionally, research is also required to explain how explicit the contractor’s selection criteria need to be if they are to be both: accessible to the members of the temporary system, and act as a substitute for history between the members of the EOC.
Optimism Swift trust theory argues that members of the EOC would expect an operational problem to affect their colleagues and not themselves. This personal optimism can make a respondent believe they are less vulnerable than their colleagues, meaning they feel less at risk and are therefore more willing to trust. As such, to gauge each respondent’s level of optimism, each was asked in terms of the operation, ‘Considering what could go wrong in the EOC, would you say you are less likely, as likely, or more likely to be detrimentally affected?’. The results are presented in Table 6.2.a.
252 Table 6.2.a Each respondent’s perceived ‘degree of impact’ for an operational problem in the EOC
Perceived degree of Respondent impact for an operational problem in the EOC EOC 1 As likely EOC 2 Less likely EOC 3 As likely EOC 4 As likely EOC 5 As likely EOC 6 Less likely EOC 7 More likely EOC 8 More likely EOC 9 More likely EOC 10 As likely EOC 11 More likely EOC 12 More likely EOC 13 Less likely EOC 14 As likely EOC 15 As likely EOC 16 As likely EOC 17 More likely EOC 18 More likely TOTAL Less likely 3 As likely 8 More likely 7
Unlike what swift trust theory predicts, only three of the 18 respondents said that they were less likely to be detrimentally affected by a problem in the EOC. Conversely, eight said they were as likely, and seven responded they were more likely to be detrimentally affected. An analysis of the results also showed specific reasons why the respondents placed themselves under their respective categories. These reasons are explained below in reference to the responses given:
Less likely to be impacted by a problem:
o Only one respondent, EOC 2 as an international colleague, displayed optimism as per swift trust theory. The respondent believed that, even if an operational problem impacted them whilst in the EOC, they could minimise the impact of it by quitting and leaving the operation altogether.
253 “Less likely. (“Why’s that?”) I can cut away and leave. You know, yeh, I mean, it’s a job, but you know, yeh…life isn’t impacted by it.” (EOC 2)
o The remaining two respondents who felt less likely to be affected by a problem in the EOC were in roles that were partially isolated from the rest of the EOC and unlikely to be impacted (EOC 6 was in charge of a small team inside the National Emergency Operations Centre, and EOC 13 was the OCHA liaison). However, neither of the answers provided by these two respondents suggested that they were optimistic about their outcomes in the operation; but rather realistic considering their roles.
“Less likely…Maybe because I am in charge of a small team, that team won’t be affected.” (EOC 6)
“I mean, generally, I don’t think I’m likely to be affected by it at all… I wasn’t really fulfilling a functional role within the [EOC]. And I don’t think, I mean the only way that that would happen is if they thought that I actually…sabotaged the operation. And then that would be detrimental.” (EOC 13)
As likely to be affected by a problem:
o Of these eight respondents, the majority simply answered ‘as likely’; and were not optimistic about their outcomes relative to their colleagues. Specifically, both EOC 4 and EOC 14, as local respondents, respectively mentioned that being a member of the EOC meant that one would actually share the responsibility for a problem and look out for their colleagues; again showing optimism was not a factor for them by willingly taking on their colleague’s problems.
“Because we have to share part of the blame.” (EOC 4)
“I think just as likely…you’re always watching out for each other. So if something did happen, even if it affected a small part of the group, I think invariably everyone else would…it would be difficult to say in some way that you wouldn’t be affected.” (EOC 14)
254 More likely to be affected by a problem:
o Each of these seven respondents were not optimistic about their outcomes within the EOC. Rather, the majority (being senior colleagues) stated that holding positions of leadership or being experienced meant that they were more likely to impacted by a problem.
“I’ll be more likely to be affected…Because I am in a position of management and supervisor in this office, so I take responsibility and account of it.” (EOC 7)
“Ah, yeh, more likely to be affected…leadership you know…” (EOC 9)
“…for me personally, it would be more [likely]… [because Current Operations] is one area of coordinating information, coordinating efforts, so, people who don’t know how things work here, this is the place where they will come to.” (EOC 11)
“I’d say more likely… most people assume because I have been here the longest, they think I’ll know [“know the answers”] yeh yeh yeh.” (EOC 12)
“…more likely to be affected…Because I’m the Director.” (EOC 17)
o EOC 8, EOC 9 and EOC 18 stated that the interdependence between members of the EOC meant that they were more likely to be affected by a problem.
“More likely…[the EOC] is the implementing office for my project and I need for nothing to go wrong in this office (laughs), so that I can set, so that this project can run smoothly.” (EOC 8)
“…Because you know, we have our corporate plan, our business plan over here, so ah, ah, if there, something happens, everybody down the line gets affected. It’s not only one person.” (EOC 9)
“More likely [because a problem will mean that my colleagues will work overtime]…and we get paid time and a half and double time. And they were relying on me to [process these hours].” (EOC 18)
255 In addition to this data, as explained in Section 5.5.2, the results also showed that 17 of the 18 respondents believed that an operational problem was likely to affect the wider EOC rather than just an individual. This was because the respondents believed that tasks were interdependent, and that they should share responsibility for the what happens in the EOC. Consequently, the data suggests that the respondents (excluding EOC 2) did not consider themselves less likely to be impacted by a problem compared to their colleagues and were not personally optimistic.
These results are significant for two reasons. Firstly, they show that, contrary to swift trust theory, all bar EOC 2 were not optimistic about their involvement within the operation, and that optimism did not impact the resilience of trust in the EOC. Furthermore, EOC 2’s optimism was founded on them leaving the operation all together; and not on their belief that they would not be impacted by a problem in the EOC. This is important to recognise because it shows that ‘optimism’ (as considered by swift trust theory) does not only include being less likely to be impacted by a problem whilst in the temporary system, but also the degree to which the member can leave the temporary system altogether (thus limiting their degree of exposure and personal vulnerability).
Finally, the data also provided a correlation between the respondents’ perceived exposure to the consequences of an operational problem in the EOC and their respective roles. Those who felt they were less likely to be impacted by a problem (EOC 6 and EOC 13) had roles that were generally isolated from the rest of the office. Conversely, those who felt more likely to be impacted were generally in positions of leadership (EOC 7, EOC 9, EOC 11, EOC 17), experienced disaster management personnel (EOC 12) or subject matter experts (EOC 8).
A presumption of trust Swift trust theory argues that within a temporary system, even though trust is presumed, members can act as if trust were present. When a member observes a colleague within the temporary system acting as if swift trust were present, this affirms that member’s original assumption that swift trust is appropriate. This personal confirmation of swift trust being appropriate, initiated by their colleague’s behaviour, increases the resilience of swift trust. To address this resilience factor, this section first considers the assumption of swift trust theory that ‘trust was presumed’ within the EOC. It then discusses whether the respondents thought their colleagues behaved as expected and whether this affirmed an assumption of trust. Finally, it considers signage within the EOC and the impacts that this had on trust being presumed and the resilience of trust.
256 6.3.1. The assumption that trust is presumed To adequately address this resilience factor, it was necessary to analyse an underlying assumption; that is, swift trust is more resilient if the trustor chooses to start from a position of trust whilst working in the EOC. Therefore, in order to understand whether this assumption adopted by swift trust theory was valid, each respondent within the EOC was asked, ‘How well would you have to know a colleague you had never met, before trusting them during the operation?’. An analysis of the data (as shown in Section 5.2.2.2) revealed that 11 of the respondents adopted a trusting disposition as described by swift trust:
“How well? I guess I wouldn't really have to know [my colleagues]…for the sake of the national exercise, you keep on doing what you have to do and trust people.” (EOC 1)
“You know, I guess, you apply trust, and you task people differently.” (EOC 2)
“…I’m sure all the people sent to the operations here are all trustworthy people. So whenever the person comes in, we just need to get along…” (EOC 3)
“No [I do not need to know my colleague well]. You just need to place our trust that they will get the work done. That is why they are here.” (EOC 4)
“Um, I feel that, [trust] just happens in the first hour we are working with them.” (EOC 6)
“I think maybe, how well? Well, not that well. Well, I don’t think we had enough time to get to know each other way more than we already…” (EOC 8)
“Quickly…I generally start trusting.” (EOC 10)
“No [I do not need to know them well]…where they’re from and why they’re here. And that’s it…it’s a personal [thing], to work with everyone…” (EOC 11)
“During the disaster phase, it would probably be maybe a week or two weeks [before I trust them]. Just to see how they work, the conversations I have with them, see, you know, just to gauge...” (EOC 12)
“I will generally walk up to someone and give them the benefit of the doubt from the word 'go'.” (EOC 13)
“[I extend trust] pretty quickly yes.” (EOC 14)
Six respondents were slower to trust, due to the following reasons:
257 Wanting to protect the EOC; through information protection (EOC 7), ensuring their colleagues were competent (EOC 15, EOC 17), or knowing their colleagues were honest (EOC 9):
“So if someone is new to the operation I have not met before, I will be very cautious about information that I share with him or her, in case, because I don’t want that information to be misused or misinterpreted. So, it’s very important that you know whoever you’re talking to and sharing the information and that you don’t know his background or organisation, you need to be cautious of the information that you use. Especially when you are working with government, that’s very important.” (EOC 7)
“Yeh, it takes a little bit of time to get to know him better…Ah I have to get the, background of the person from a previous colleague who they have worked together. So it can give me an indication that he can do the job or is good team player.” (EOC 15)
“I would ah, have a brief session…I’d like to know the person more in detail, his or her background…their experience…and that’s what I do for everyone who comes inside.” (EOC 17)
“I just don’t trust anyone in the first time that I meet people because nowadays, there are a lot of people who can go around pretending to be someone.” (EOC 9)
Needing to feel personally secure because they had previously been undermined:
“Mmm, it will take me long…a year and a half…I have always been on the losing side eh…[so] when I look at people and I trust them, I need to be secured.” (EOC 5)
Wanting to become more familiar with their colleagues by working with them:
“Mmm, six months…Ah, it depends on how I work with [my colleagues]…it usually takes a while for me to trust someone.” (EOC 18)
Uniquely, EOC 16 (as a member of Current Operations) was the only respondent who stated that their disposition to trust was dependent upon the role and nationality of the other person. They
258 argued that it would take: one day to trust a member of the Fijian military; a couple of days for a local colleague; and a week for an international colleague.
Overall, the results showed that whilst 11 of the respondents had a predisposition to trust in line with swift trust theory, seven did not. Therefore, contrary to the theory, it is not fair to assume that within an EOC as a temporary system, that the members will start from a position of trust. Furthermore, the data showed that for those who were slower to trust, there were several unique reasons for this occurring. For both EOC 5 and EOC 18, these respondents were concerned with trust in a personal sense; wanting to feel more familiar with their colleagues or personally secure before extending trust. However, the data also revealed that for the respondents in positions of leadership (EOC 17 and EOC 7 in Current Operations, EOC 9 as the head of logistics, and EOC 15 as the head of procurement), their primary reason for not having a quick trusting disposition was to protect the EOC. These four respondents highlight an important implication for swift trust theory, and trust development in general, which is that some members of a temporary system consider more than just their personal vulnerability when deciding whether to trust a colleague. In forming their trusting disposition, these four respondents also considered the risk of the EOC being undermined. This is important to recognise as most of the trust literature only focusses on the development of trust based on the risk to the trustor. However, this case study has shown that some personnel also place a high (if not higher) priority on protecting their organisation (EOC) when forming trust; particularly those in positions of leadership.
6.3.2. Whether behaviour affirmed a presumption of trust Swift trust theory also argues that if a member observes a colleague within the temporary system acting as if swift trust were present, this affirms that member’s original assumption that swift trust is appropriate; creating a self-fulfilling cycle. Therefore, the interview asked the respondents whether they thought their colleagues behaved the way they expected them to whilst in the response. As discussed in Section 5.4.2, 12 respondents stated that their colleague(s) behaved as they expected in the operation. Conversely, six respondents stated that, at least at one point during the operation, their colleague(s) did not behave as expected.
Secondly, swift trust theory argues that trust becomes more resilient when a member has a predisposition to trust and that this is affirmed by a colleague’s trusting behaviour. Therefore, each respondent’s trusting disposition was compared against whether they felt their colleagues behaved as expected whilst in the EOC. The results showed that only seven of the 18 respondents had both a predisposition to trust and felt their colleagues behaved as expected
259 throughout the operation. The case study suggests that for the EOC, the resilience of trust was less affected by the respondents adopting a presumption of trust. Rather, contrary to swift trust theory, the results highlighted that within an EOC, as a temporary system, it is not guaranteed that there will be a, “collective perception that swift trust is reasonable” (Meyerson et al., 1996, p. 186). That is, an individual’s predisposition to trust may not be overruled simply by their inclusion into the temporary system; even though it has been shown that ‘group inclusion’ can impact trust formation. The data showed that self-protection or a desire to protect the EOC may be greater prerogatives for an individual than presuming trust.
6.3.3. The impact of signage on the presumption of trust Finally, on the walls within the operation were signs that listed the values of the EOC; one being trust (refer to Appendix A). To appreciate the influence of these signs on creating an environment where trust could be presumed, each respondent was asked, whether they were aware that trust was a value of the EOC, and if so, did it impact their behaviour throughout the operation? Although 17 of the respondents were aware of the signs, only three respondents said that it affected their behaviour. Most respondents said that it did not change how they behaved. These respondents felt that:
The signs were redundant or not important to the EOC leadership.
“…I don’t think it has been championed by our management.” (EOC 1)
“No, we just don’t bother…[the signs are] just for record and people don’t bother reading it.” (EOC 6)
“No, cause I mean, no-one reads them [laughs]” (EOC 12)
“I’ve seen them but I did not take time to [read them].” (EOC 15)
The signs were not personally applicable.
“Oh look I’ve read it, you know, but that is for the Fiji public service and I’m not a member of the Fiji public service.” (EOC 2)
Signage cannot make you trust someone.
“How can you put something up on a poster that says ‘trust’! I mean, that’s, that’s one of the things that I’d get rid of that. You know, you could write, ‘build trust with other people’ but, you know. That’s something you can do, or anyhow, or try to.” (EOC 10)
260 “Because I trust people…[the signs do not influence me].” (EOC 4)
“I know that is the intention to aim towards, but I don’t know that you would trust some just cause the code of conduct says ‘trust’.” (EOC 14)
Particularly, when asked whether the signage influenced trust formation, EOC 12, EOC 16 and EOC 18 all laughed at the concept. The results showed that although the code of conduct was displayed in the EOC and encouraged trust, it seemed to have done little to actually create an environment where trust was presumed. This data also compliments the trusting criteria of the respondents in that they required evidence-based information when forming trust in the operation; and relied less on categorical assumptions such as the influence of signage.
Lastly, the case study highlighted that the resilience of trust (its development and whether it is sustained) is unlikely to be influenced by a displayed code of conduct. Rather, the respondents’ answers showed that trust is not a trait that can be presumed because of organisational expectations, but rather is a personal decision. And although signage may play a part in making trust known as an expectation, the authority of such signage is negligible. Consequently, the results showed that trust is more likely to form if the expectation of trust is: championed by the leadership in the EOC, made applicable to all staff (local and international), and actively encouraged between the members of the EOC rather than just visually displayed.
Assumed qualifications Swift trust theory argues that swift trust can be more resilient if the respondents believe their colleagues are included within the operation because they are qualified in their role; with their expertise and experience being gained elsewhere and not requiring verification. Therefore, this section will first address whether the respondents knew their colleagues’ qualifications. It will then discuss whether the respondents thought their colleagues were qualified for their respective roles in the operation and the impact this had on the resilience of trust in the EOC.
6.4.1. Whether the respondents knew their colleagues’ qualifications To understand whether the respondents assumed that their colleagues were qualified for their roles in the operation, each respondent was asked, ‘Of the local/international colleagues you had not met or worked with before, do you know their professional qualifications?’. This was asked for both local and international colleagues to determine whether there was a difference in the expectations for being qualified.
Simply, within the EOC, every respondent (bar the Director) answered that they did not know the professional qualifications of either their local or international colleagues. Specifically, the 261 Director only stated that they knew some of the qualifications of the international colleagues; as he had potentially received their resume prior to these personnel coming into the operation.
6.4.2. Whether the respondents thought their colleagues were qualified Swift trust relies upon the assumption that a respondent’s expertise has been gained prior to the temporary system forming. However, given none of the respondents knew their colleagues’ qualifications before the operation, this research sought to assess whether the respondents thought their colleagues were actually qualified; demonstrated, as swift trust argues, by competent role performance. Subsequently, to understand whether a colleague’s assumed qualifications impacted the resilience of trust in the EOC, each respondent was asked, having observed their colleagues working throughout the operation, ‘Do you think your local/international colleagues are qualified to do their job?’. Again, this was asked separately for local and international colleagues. Focussing first on the respondents’ answers concerning their local colleagues:
Ten stated their local colleagues were qualified.
Six said that their local colleagues were not always qualified in their roles.
Two stated that their local colleagues’ qualifications did not matter when working in the EOC.
For the 10 that said their local colleagues were qualified, they had the following reasons:
Due to a colleague coming from a reputable parent organisation and that this organisation would send across the ‘best’ personnel to help in the operation. This comment was only made by local colleagues:
“To me, it is easy to, because when I know they are sent to work in the [EOC], they are one of the best in the ministries or departments.” (EOC 3)
“Yes, I think [they are qualified]…because they were being sent from others ministries, I assumed they are senior officers…” (EOC 18)
“Ah, I think [they are qualified] yah, because ah, ah, we don’t just get people anyhow to work here. They get released from their different ministries. So they are administrators, managers from that agency releases to us the people whom they trust. People that they think can take up emergency operations.” (EOC 9)
Due to their colleagues delivering on their work: 262 “[For the colleagues] that I actually got to work directly with…oh yeh…when I see them do their work, yeh…” (EOC 11)
“Yeh yeh [they are qualified], [the EOC] isn’t rocket science. It is pretty straight forward, pragmatic kind of work. You just need to be practical…” (EOC 2)
“For the level of work they have been doing, I definitely think [they are qualified].” (EOC 8)
“Um yes [they are qualified]… because the output they always take, they always put, whatever the input and the output is perfect.” (EOC 5)
“From what they’ve done during the operations, [they are] definitely [qualified].” (EOC 14)
“…yes to some extent. I can judge the type of qualification they have. The way they deliver out their work every day.” (EOC 15)
The six respondents who stated that they thought their local colleagues were not always qualified argued that these colleagues:
May have lacked training in disaster management (even though they may have come from reputable parent organisations):
“Mmm, can I say no?...I mean, some of them were not qualified…cause [the staff in the National Emergency Operations Center] were staff that were pulled in from other organisations that had very little experience or training in systems.” (EOC 12)
“Well the guys from the public service commission selected them to do their jobs so obviously they would know what they are doing because they are senior officers…[but] judging from the reports that we see from the first few weeks…they did not really know what sort of information we needed to extract from the data we were receiving…So we had to keep on grooming them during that period.” (EOC 4)
“Ahh, I would suppose so by them being sent over to work…[but] it can be yes and it can also be no. You see here, Fiji is a developing country. A lot of, I would say, many people who are actually working ah, may not have that qualification
263 for that particular position. But some of those people go through that experience, throughout the years.” (EOC 17)
“Yeh, some [locals] were the wrong person [for the role], but anyway.” (EOC 16)
Did not always produce the work required of them whilst in the EOC:
“Sometimes you have reservations…Just about the quality of work, their meant to do something, you look at the quality of work and you begin to doubt. That’s when I want to check on their background.” (EOC 6)
Finally, two respondents stated that they were not concerned about the qualifications of their local colleagues, but rather whether they could deliver the work required for the operation.
“If they are able, if they’re able to do the work, then I don’t mind if they’re qualified to do, as long as they can do the work.” (EOC 7)
“As you work with them, you start to figure out can they [do their job]…like qualifications…actually I’m still unqualified in lots of the things I do, it doesn’t matter. It’s about can you do it or not. Have you got, you know, have you got an ability to do it.” (EOC 10)
Conversely, for the international personnel within the response, 16 of the respondents stated that their international colleagues were qualified. (Note: EOC 3 did not feel they could accurately answer this question). These respondents:
Assumed their international colleagues were qualified given the colleague’s parent organisation:
“…I guess [they are qualified]; if the Australian Government sent them.” (EOC 1)
“Ah, well, for the known organisations, like the trust is there between the government and that organisation. So the people that they send almost the same.” (EOC 9)
Simply assumed that their international colleagues were qualified; sometimes without providing additional reasoning:
“Yeh, probably, yeh sure.” (EOC 2)
“[They] should be…[I] assume.” (EOC 4)
264 “I would suppose so… Well, probably I’d say, looking at an international colleague you would expect that them being international, they would have some sort of qualification.” (EOC 17)
“Yes, I assumed they were qualified...because they were brought in…” (EOC 12)
“Yes…no doubt.” (EOC 15)
“They must be qualified because if they do not qualify, they wouldn’t be here. (laughs)” (EOC 7)
“Yeh, definitely. Definitely.” (EOC 8)
Felt their international colleagues were qualified given their engagement with the work in the EOC, delivering on this work, and/or having previous experience with disaster management:
“Yes [they are qualified] (said quickly with assurance). We speak the same language [concerning the work in the EOC].” (EOC 10)
“…[they are qualified because] well the ones that I’m working in are actually producing the stuff that they are required to do it.” (EOC 12)
“Yeh [they are qualified]. It’s just how they came across, asking questions.” (EOC 11)
“Yes, cause their very, their technology is more advanced than us…They are good at what they do, the information, and the output is very perfect.” (EOC 5)
Only one respondent (EOC 6) felt that their international colleagues were not always qualified given that they sometimes produced substandard work whilst in the EOC.
“Because, from my understanding, when they came into work, they must have gone through some screening processes and they check their qualifications…But um, when we actually work with them, and you’re not getting the desired output, the result that you want, then you begin to doubt.” (EOC 6)
265 Table 6.4.a A summary of whether the respondents thought their local and international colleagues were qualified for their roles in the EOC
Not Colleague type Qualified N/A Qualified Local 10 6 2
International 16 1 1
Again, swift trust theory argues that the members of the temporary system assume their colleagues are qualified and proficient in their roles. However, “deviations from…competent role behaviour call in question the “professionalism” of the transgressor” (Meyerson et al., 1996, p. 190) which can impact the resilience of trust. Considering this, although the EOC did not exhibit swift trust, these results were still applicable as they highlighted multiple significant factors concerning trust and its resilience within the EOC.
Foremost, the case study shows that the parent organisations for both local and international colleagues can act as a substitute for the qualifications of that colleague. Multiple local respondents stated that they felt their local colleagues were qualified given they came from the civil service (refer also to Section 5.3.1.4 for the civil service being a trust criterion). Similarly, some respondents assumed their international colleagues were qualified and competent given their affiliation with an international organisation; such as the Australian Defence Force.
Multiple respondents also stated the importance of both their local and international colleagues delivering on their work. Furthermore, if a colleague failed to produce their work, this led to a respondent questioning the capability and qualifications of that colleague. Additionally, given the use of competency as a trust criterion by 13 of the 18 respondents, the data suggests that inadequate role performance impacts the resilience of trust within the EOC. That is, because that respondent cannot rely on their colleague to perform within the operation, trust in that colleague may deteriorate or not develop at all. As such, the data shows that, although the EOC did not exhibit swift trust, the development of trust and it being sustained (resilience) throughout an operation can be impacted by a colleague’s perceived qualification; which is verified through competent role performance.
Secondly, the data presented a significant contribution to the field of disaster management response; which is that international disaster personnel can simply be assumed to be qualified with little proof required by their local colleagues. Apart from the use of parent organisations by two local personnel, seven other respondents (six of which were locals) stated that they
266 assumed their international colleagues were qualified without providing further justification. On the other hand, the data showed that some local respondents recognised that the local staff may not have been fully qualified given they were not appropriately trained in disaster management. This difference in assumed qualifications and competency is important to recognise because it places a prerogative on the international organisations sending personnel to aid a response to make sure that they send the appropriate person(s). That is, given there may be a subliminal assumption by the host nation that those coming from overseas are qualified, if that international colleague is not qualified or does not perform competently (as shown with OCHA in Section 5.1.2), this can decrease the resilience of trust or even lead to distrust in that organisation entirely. Therefore, the data showed that it is in the international organisation’s best interest to ensure that those sent to aid a response are qualified and competent in their respective roles; even if these qualifications may simply be assumed by their local counterparts.
Furthermore, this expected competency must not be exploited by the international personnel aiding the response. Although the international respondents may be assumed to be qualified, as discussed in Section 5.3.2, the local leadership also expected this assumed competency not to be exploited. Rather, they expected that their international colleagues would work in partnership with the local leadership; humbly integrating into the response without exerting their expertise as a prerogative for running the operation.
Overall, the data showed that competent role performance was used to verify the assumed qualifications of a colleague. However, if a colleague did not adequately perform their role, then this led a respondent to believe their colleague was not qualified. Provided the reliance on competency as a trust criterion by the respondents within the EOC, the data suggests that the assumed qualifications of a respondent can impact the resilience of trust. Put simply, if a colleague is found to be proficient, this can help trust develop and be sustained throughout the life of the temporary system.
Hedges The fifth resilience factor for swift trust considers hedges. A hedge is defined as a ‘way out’ in order to, “reduce the perceived risks and vulnerabilities of trust by reducing interdependence and thus its perceived costs” (Meyerson et al., 1996, p. 187). Furthermore, hedges can lead to the member who created it:
to become overconfident,
267 to have reduced commitment to the temporary system, and,
being perceived as untrustworthy if the other member(s) of the temporary system find out that the hedge exists.
Within the operation, the data showed that there were two respondents who had hedges; EOC 2 as an international colleague working in logistics, and EOC 8 as a local surge staffer working on mapping (GIS). This section will address each of these hedges, the respondent’s approach to their hedge, and the respective impact of the hedge on the resilience of trust.
Firstly, throughout EOC 2’s interview, it became apparent that the respondent was not attached to the EOC and was willing to leave the operation if they felt they were becoming too vulnerable to the risks of being involved in the response. This was most evident when they stated;
“I can cut away and leave. You know, yeh, I mean, it’s a job, but you know…life isn’t impacted by it.” (EOC 2)
Furthermore, as argued by swift trust theory concerning hedges, EOC 2 also made comments throughout their interview which emphasised their (over)confidence in their ability and role in the operation. They also made statements that showed that they felt able to isolate themselves from the EOC and have a reduced level of commitment to the temporary system.
(Over)Confidence in ability, role and access to senior staff:
“…like, if you work in development for a long time, you understand that, um, you know…one of my superpowers is my ability to judge character and you get good at it. After a while, you can tell who is going to be reasonable and who isn’t. I guess that is one of the benefits of age and experience I suppose.” (EOC 2)
“If I’m not here, my job doesn’t get done…[my colleagues] don’t do what I do.” (EOC 2)
“…if I needed a decision, I’d talk to the Minister…usually he’d find me.” (EOC 2)
Reduced commitment to the temporary system:
“…[the office code of conduct] is for the Fiji Public service and I’m not a member of the Fiji public service…I’m held to the Australia [Government].” (EOC 2)
“I can cut away and leave.” (EOC 2) 268 Positively, swift trust theory also argues that hedges can help trust form because the member with the hedge is more willing to trust; knowing that they have a process in place to limit their vulnerability. However, although EOC 2 had a personal hedge in place, as shown in Section 6.3.1, they also had a predisposition to trust their colleagues. As such, it is unclear whether their predisposition to trust was a product of their hedge, and what impact this had on the resilience of EOC 2’s trust. Furthermore, the data showed no evidence of another colleague being aware of EOC 2’s hedge. Thus, although the hedge did exist, because it remained confidential, it had no discernible impact on the resilience of trust for the wider EOC.
The second recorded hedge belonged to EOC 8, whom had applied for a scholarship to study abroad prior to TC Winston. When specifically asked about their hedge, the respondent said that they believed it would not lead to distrust because seeking further training was encouraged in the EOC.
“I know for a fact, this office is always looking to build capacity and furthering their staff’s education…You know, I believe in people actually bettering themselves whether in education or financial, you know, leaving to go to a better opportunity. I definitely think it’s personal growth. Not only personal but financially, you know. At the end of the day, it’s not only about money but it’s experience you are going to get out of the new job yeh.” (EOC 8)
Furthermore, they also said that they would return to the EOC, and would support anyone who was trying to ‘better themselves’; showing that their hedge had not impacted their level of commitment to the temporary system.
“…if someone was leaving who had the same expertise as me to do a better job, then, you know, then definitely, good for them…[and I would come back], definitely, definitely.” (EOC 8)
This group commitment was also highlighted when they said that they intended to hand-over their work appropriately and train another colleague to be able to fill their role once they left. Similarly, this also showed that EOC 8’s perception of their hedge was not one which promoted overconfidence and a belief that they were the only one capable to do their role.
“I mean, if you knew that you were leaving, I mean if you were planning to leave, like in my case, I’ve already implemented, I’m actually drawing up a work plan where I’m gonna try and train everybody in this office in GIS. [“Handover properly?”] Yeh, basic GIS, and then, you know, that would be the wiser thing to do.” (EOC 8)
269 As EOC 8’s hedge did not lead them to be overconfident or have reduced group commitment, the results suggest that the hedge had a limited impact (if any) on how they formed trust with their colleagues and its resilience. Rather, as with EOC 2, the respondent already had a predisposition to trust their colleagues within the EOC. Similarly, there were also no recorded instances of EOC 8’s hedge being known by their colleagues; which could have led to distrust and impacted the resilience of trust for the wider EOC.
Finally, considering hedges more generally, the EOC was a temporary system that had at least two senior respondents (EOC 7 and EOC 11 from Current Operations) displaying characteristics in direct contrast to a hedge. Both these respondents, as trained disaster managers, were enthusiastic to be involved in the response; as opposed to protecting themselves from the risks of being involved in the operation and distancing themselves from the group.
“I pick up myself and I came, because I have a responsibility for my country. No-one told me to come here…I just reported here because I know my services are needed.” (EOC 7)
“I want to share is how I actually got to come in. The thing is, I am part of a, there is a group of the officers who are in the civil service who are actually trainers who develop people to work in emergency operations centres to do disaster management work. So it was in my role as a trainer and as someone who has worked in a number of operations, um, that brought me to [the EOC].” (EOC 11)
Overall, although two hedges were recorded within the EOC, neither of these were perceived by the wider EOC and there was no observable impact (positively or negatively) on the resilience of trust within the operation. Conversely, the data did show that, in accordance with swift trust theory, that hedges may result in the holder having (over)confidence in their abilities and a reduced level of group commitment. However, the effects of a hedge on the member appear to be related to their perception of the hedge’s purpose. That is, EOC 2 was willing to leave the EOC for personal protection. On the other hand, EOC 8, although still willing to leave the operation, was doing so for professional development and further training; and intended to return to the EOC in the future. Although there were only two hedges present in the EOC, this data does suggest that hedges in a temporary system:
can exist yet still have no significant impact on the resilience of trust, particularly if they are not known by the wider temporary system;
may not lead to overconfidence and lowered commitment to the temporary system, contrary to swift trust theory; and,
270 can be perceived and justified differently by their respective holders.
Improvisation Swift trust theory argues that trust may be more resilient in temporary systems that are able to improvise solutions to problems because the members feel that the group is able to overcome inevitable setbacks. However, as shown with multiple other factors within swift trust theory, there remains an underlying assumption. In this case, the theory assumes that improvisation is a part of the temporary system. Therefore, to address whether this was a fair assumption for the EOC, each respondent was asked, ‘When problems arise that affect the EOC, are solutions allowed to be improvised or is there a strict procedure that is to be followed?’.
Within the EOC, 13 respondents stated that solutions were simply allowed to be improvised (Note: both EOC 1 and EOC 13 did not answer the question). Nine of these respondents answered briefly, affirming that improvisation was commonly accepted within the operation:
“No, it’s improvised.” (EOC 2)
“Ah, it can be improvised.” (EOC 7)
“(Responded quickly) Oh just improvised.” (EOC 9)
“(Responded quickly) No no, improvised.” (EOC 10)
“Oh we would improvise.” (EOC 11)
“Ah solutions were allowed to be improvised.” (EOC 12)
“Ah, improvised.” (EOC 15)
“No no, you can improvise.” (EOC 17)
“It can just be solved.” (EOC 18)
The other four respondents believed that improvisation was a part of the response due to the inadequacy of the Standard Operating Procedures (SOPs). These respondents stated that improvisation was necessary because the SOPs were: out of date, not familiar, insufficient in detail, and/or deemed irrelevant.
“[You can] make up a solution…our SOPs (Standard Operating Procedures) not really, it’s outdated…It’s not up to the standard practice that is currently being used.” (EOC 4)
271 “…there is an SOP, Standard Operating Procedure that we use. They’ve got cyclone plans, what have you. But what it doesn’t tell you is, ‘how do you take, how do you pick up the gaps in the information using something else…[so, yes, there is improvisation, which is] the whole discussion around ‘what do we do now?’” (EOC 14)
“I guess there is a procedure to follow, but…I’m not familiar with it…[the solutions are] just what we thought.” (EOC 5)
“Well there is written procedures but [they] are drafts and no one reads them. So basically, what our understanding is of it, um, and if that still doesn’t fit into that box, we have no choice but to come up with something new.” (EOC 16)
Contrary to these opinions, three respondents argued that solutions were not allowed to be improvised; rather a respondent should inform their superior of an operational problem in the EOC. That is, there was a hierarchical reporting process in place should a problem arise, and the respondent was not encouraged to improvise a solution themselves.
“…there is a procedure that needs to be followed…they will need to talk through their supervisor, their supervisor then to the director, the director then to the PS (Permanent Secretary). So there is a channel that needs to be followed…There are regulations that cover the [EOC] operations…” (EOC 3)
“That is a difficult [question], because…you may want to respond immediately but [that is not encouraged]…because the Director, he wants to know everything…But sometimes it’s difficult and you need to help to respond to something immediately…[there are] channels to follow. Sometimes I feel it limits our efficiency…I want to develop some SOPs and some plans for future.” (EOC 6)
“Protocol, protocol…Going to the more senior person to tell them about the problem…I can’t do it alone.” (EOC 8)
Swift trust theory argues that improvisation makes trust more resilient because it, “inspires confidence that unexpected but unavoidable setbacks…are surmountable and survivable” (Meyerson et al., 1996, p. 189). However, although the case study showed that most respondents within the EOC adopted improvisation as a technique to solve unforeseen problems, the results did not necessarily show that improvisation made the respondents more confident in the ability of the temporary system to overcome problems. However, the respondents’ answers did show that improvisation:
272 was a necessary part of natural disaster management and working with their colleagues in the EOC; and,
was required, because the Standard Operating Procedures were out of date and did not provide adequate solutions to the specific problems faced during a response.
Furthermore, by considering this data in context of the wider results, this research believes that if improvisation did impact the resilience of trust in the EOC, it was on the condition that a colleague was proficient when improvising a solution. Firstly, as explained, improvisation was a necessary part of the operation, with the EOC relying on it throughout the response. Examples included:
The creation and incorporation of Future Operations into the EOC for the first time.
The creation of new reporting lines between each different operational group (Future Operations, Current Operations, the National Emergency Operations Centre, and the National Disaster Management Office).
Implementing and adapting the requirements within the Standard Operating Procedures.
The incorporation of surge staff into the response.
Secondly, as shown in Section 5.2.3, this response was characterised by personnel taking on additional roles and working longer hours so as to meet the requirements of the operation; which was itself another act of improvisation. Although those in positions of leadership were cautious of personnel taking on additional roles, the majority said that trust was not negatively affected if their colleague performed their additional role proficiently. Finally, the majority of respondents within the EOC used their colleague’s competency as a criterion for developing trust. Therefore, the culmination of this data suggests that if improvisation was to make trust more resilient, it was dependent on whether a colleague was deemed proficient when improvising.
The results also highlighted multiple significant contributions to disaster management and trust formation. Firstly, the fact that the Standard Operating Procedures were out of date and insufficient increased the need for improvisation in the EOC. The data also showed that Standard Operating Procedures need to be relevant, adaptable and applicable to the requirements of a
273 national emergency. They also need to recognise and incorporate improvisation as a necessary part of disaster response.
The data also showed that internal reporting hierarchies can hinder improvisation; which may impact trust development. The results showed that some colleagues felt they were unable to improvise because they needed to report any problems upwards to their manager. Such constraints can lead to bottlenecking and slow the response, as members are waiting on approvals from those in senior positions. Conversely, those in positions of leadership (EOC 7, EOC 11, EOC 14, EOC 16, EOC 17 from Current Operations) or disaster management experts (EOC 2, EOC 9, EOC 10, EOC 12) where much more comfortable with improvising. Their answers to the interview showed that improvisation was a necessary part of the response and that they were able to do it without seeking approval. The data suggests that there is a correlation between seniority/expertise and feeling authorised to improvise a solution.
Although the results showed that hierarchy can constrain the response, the data did also show that hierarchy is important so that: a unified message is presented to the public from the EOC about the progress of the response, and the EOC is protected in terms of dissemination of information and information security. Considering this practically, given that disaster management requires continual adaptation to the dynamics of the response, if members of the EOC feel constrained by their reporting requirements, this may impact the speed at which the response and recovery can be managed. However, the case study shows that there needs to be a balance between members of the EOC feeling free to improvise, whilst not damaging the integrity of the EOC as the office charged with managing and leading the operation.
Focussing on the task Swift trust theory argues that due to the time constraints imposed upon temporary systems, such systems simply do not have time to develop harmful interpersonal and group dynamics that can erode trust. Rather, members are focussed on the task and ensuring its completion, which can make swift trust more resilient.
As shown in Section 5.2.1, the respondents within the EOC stated that the majority, if not all, of their interactions with their colleagues throughout the operation were related to the response; with minimal time for social interactions. Similarly, as discussed in Section 5.4.2, the respondents also stated that their colleagues behaved as expected, focussing on the task of the operation; with personnel taking on extra workload and working longer hours to meet the requirements of the operation.
274 However, the results also showed that, contrary to swift trust theory, that focussing on the task at hand can result in hasty actions that subsequently lead to distrust. As shown in Section 5.1.2, OCHA published response data without the approval of the Director and the EOC, which caused them to be distrusted by the EOC and the Fijian Government. This incident highlights the consequences when haste in responding is used as justification (knowingly or not) to bypass the protocols of the temporary system; which can erode trust as opposed to making it more resilient.
It is important to recognise that untrustworthy behaviour is not nullified because the temporary system is under time pressures. Swift trust theory argues that in a temporary system, “there is simply not enough time for things to go wrong” (Meyerson et al., 1996, p. 190). However, what the theory fails to recognise is that the time constraints imposed on the temporary system may actually induce behaviour which is damaging to the resilience of trust. Therefore, the data shows that the EOC was generally focussed on the task of responding and had little ‘extra’ time to develop harmful interpersonal dynamics given the requirements of the operation. However, the case study also shows that the time constraints imposed on the EOC, and focussing solely on the task of responding, can lead to hasty behaviour which erodes trust, rather than making it more resilient.
Discussion of Research Proposition 3 In consideration of the aforementioned results, this section will address research proposition three by discussing the resilience of the trust formed in the EOC with relation to the seven swift trust resilience factors. It will also explain the implications of these conclusions on the theory of swift trust in temporary systems as well as the wider disaster management literature. This will be done through a discussion of each of the seven swift trust resilience factors previously analysed:
1. The reputation of the contractor,
2. Optimism,
3. A presumption of trust,
4. Assumed qualifications,
5. Hedges,
6. Improvisation, and,
275 7. Focussing on the task.
6.8.1. The reputation of the contractor The case study found that the Director did not have a reputation for ‘composing successful temporary groups’, nor did they make explicitly known the selection criteria they applied to any member included within the operation. Subsequently, this meant that neither of these two factors were able to act as ‘substitutes for interpersonal history’ between the respondents and did not contribute to the resilience of trust in the EOC.
These findings add to the work of Goodman and Goodman (1976) and Meyerson et al. (1996) by showing that temporary systems can be created without the personnel being selected by the contractor. Furthermore, swift trust theory would argue that this would mean that trust in the EOC may be less resilient because the respondents are not able to rely on the contractor’s selectivity as evidence when trusting their colleagues. However, the results showed that the formation of trust in the EOC was only partly impacted by the Director’s ability to form the temporary system. Rather, as shown in Section 5.6.1, the respondent’s used the Director’s discernment, leadership and decision-making capabilities, and personal endorsement of colleagues as substitutes for evidence when deciding who to trust in the EOC. These findings extend the work of Hyllengren et al. (2011), who found that, in a military context, the reputation of the leader impacted the development of swift trust. Particularly, the authors found that swift trust in the leader was a product of the individual and relational characteristics of the leader (including their management and leadership skills). This research has shown that, within an EOC, these characteristics of the leader also help the members of the EOC to trust their colleagues; even if the trust formed was not theoretically swift trust.
Secondly, the data adds to the work of Goodman and Goodman (1976), Packendorff (1995) and Meyerson et al. (1996) by showing that temporary systems can still exist, even if they are not formed primarily by the contractor. Rather, for disaster management, the case study shows that EOCs (as temporary systems) can be formed through two types of personnel coming together:
A core response team comprising permanent EOC staff.
Additional surge staff, that are brought into the EOC to bolster the capacity of the EOC throughout the response. Furthermore, these surge staff can be asked into the operation by personnel other than the leader of the EOC.
This finding has flow-on effects for swift trust theory as Meyerson et al. (1996) do not consider the formation of a temporary system built around a pre-existing core group of people (such as
276 the permanent staff of the EOC). This is an important consideration that requires further research, given:
The assumption in swift trust theory that personnel are included into the temporary system based upon selection criteria applied by the contractor.
Only new personnel will undergo the selection process imposed by the contractor for inclusion into the system. This creates an imbalance as any new members to the EOC are not able to apply the contractor’s inclusion criteria to their colleagues who were a part of the EOC prior to a national operation. This is because they were permanent staff and may have undergone different selection criteria to that applied in a response situation.
Consequently, the results show that how the EOC is formed (and temporary systems in general) can impact the influence of the contractor’s selectivity on trust formation and the resilience of such trust.
Finally, although the Director of the EOC had selection criteria that they used to decide whether to include someone within the operation, the data showed that these were not made clear to the wider EOC. As explained in Section 5.6.1, endorsement by the Director helped trust form for a few colleagues in the operation, however, none of these respondents specifically knew any of the criteria used by the Director. As such, in accordance with swift trust theory and the research of Meyerson et al. (1996), having inclusion criteria is insufficient in making trust more resilient. Rather, criteria must be used explicitly if they are to be both: accessible to the members of the temporary system, and act as a substitute for history between the respondents. However, given that the respondents relied more on evidence-based information (such as personality and competency) than categorical assumptions, the results suggest that for the EOC, any inclusion criteria used by the Director, even if known, may not have increased the resilience of trust for the respondents towards their colleagues. As highlighted in Section 4.8, it is recommended that further research be undertaken to consider the influence of the head of the EOC, as the contractor, on trust formation; recognising that trust is less impacted by how they assemble their team and endorsement of others (as per swift trust theory), and more their personal attributes.
6.8.2. Optimism An analysis of the results showed that the respondents (except EOC 2) were not optimistic about their degree of vulnerability in being a part of the operation. Therefore, contrary to Meyerson
277 et al. (1996), the data found that optimism did not impact the resilience of trust in the EOC. It is important to note that swift trust theory includes the concept of optimism from the research of Weinstein (1980); who argues that people are unrealistically optimistic about what will happen to them compared to others. However, Weinstein (1980) argues that optimism is a partial product of a person’s perceived control over a situation. Significantly, and in opposition to this claim, the results showed that:
Those in positions of leadership or experienced disaster management personnel, who had a degree of control over their situation, were not optimistic about their outcomes. These respondents expected to be impacted by a problem in the EOC specifically because of their position. That is, the results suggest that those in leadership were aware of the dynamic nature of a national response and that problems were a part of this; which subsequently influenced their degree of optimism.
Some respondents felt that they should share responsibility for a problem in the EOC, regardless of how it eventuated. Therefore, even though these respondents had control over whether to become involved in a problem within the operation, they would still choose to for the sake of the response.
Conversely, the data also showed that optimism was related to control in two cases:
It was found that EOC 2 was optimistic about their outcomes in the response; given that they had control over whether to leave the operation whenever they wanted.
The majority of the respondents stated that they would be impacted by any problem in the EOC given the interdependent nature of the response. Thus, the respondents recognised that they did not have control over this factor, making optimism unlikely.
The case study provides a valuable contribution to swift trust theory, and ultimately the research of Weinstein (1980), by showing that optimism can be related to perceived control; however this is not always the case. Within the EOC, those in positions of leadership, although having control, approached problems realistically; knowing their involvement was necessary. Additionally, personal optimism, through perceived control, fails when a person chooses to use that control to become involved with the problem.
Finally, the case study also extended the concept of optimism as used by Meyerson et al. (1996) in swift trust theory. The authors primarily refer to optimism in relation to a person’s belief that they are less likely to be impacted by a problem within the temporary system. However, as
278 demonstrated by EOC 2, the results extended this definition by showing that optimism does not only include a person’s perception that they are less likely to be impacted by a problem whilst in the temporary system, but also the degree to which the member can leave the temporary system altogether (thus limiting their degree of exposure and vulnerability).
6.8.3. A presumption of trust Although many authors agree that a person’s trusting disposition is important for trust formation (B. Barber, 1983; Gambetta, 2000; Hung et al., 2004; Lewicki & Tomlinson, 2003; McKnight et al., 1998; Tatham & Kovács, 2010), Meyerson et al. (1996) hypothesise that in a temporary system, the members will start from a position of trust. However, the results showed that whilst 11 of the respondents had a predisposition to trust in line with swift trust theory, seven did not. Consequently, the data shows that it is not fair to assume that within an EOC as a temporary system, the members will all start from a position of trust and that the resilience of such trust will be able to rely on a member’s presumed predisposition to trust.
Additionally, the results also show two primary reasons why a respondent did not have a predisposition to trust. These were, a member: considering their personal vulnerability (EOC 5, EOC 18), or considering the EOC and wanting to protect the operation (EOC 7, EOC 9, EOC 15, EOC 17). Focussing first on trusting disposition and personal vulnerability, much of the literature considers trust in terms of the personal risks that the trustor takes on when deciding to trust a colleague. A person’s willingness to be vulnerable and initiate trust (or distrust), as McKnight et al. (1998) would argue, is a product of their faith in humanity. Similarly, Lewicki and Tomlinson (2003, p. 2) argue that someone’s disposition to trust is a, “function of the degree to which trust has been honoured in that individual’s history of prior interactions, and may have its most profound effect in novel or ambiguous situations.” The example of EOC 5, who stated that they had experienced incidents where they had been “on the losing side” (EOC 5), confirmed this concept. Particularly, this respondent’s answer provided a contemporary example of how, as Delhey and Newton (2003) state, distrust can occur due to a previous lack of success or well-being.
Secondly, this case study not only provides a valuable contribution to swift trust theory, but also the wider trust literature, by showing that not having a predisposition to trust can be based upon a person wanting to limit vulnerability to something other than themselves; in this case, the EOC. It is also important to note that this was only stated by those in positions of leadership. This is significant to recognise because, whilst some trust models consider the role of contracts and organisational safeguards (McKnight et al., 1998; Rousseau et al., 1998) in helping trust
279 form, they do not consider that the trustor may consider the organisation itself something worth protecting and using this as an antecedent for trust.
The case study also showed that trust cannot be presumed because of implied organisational expectations such as signage; but rather is a personal decision. Particularly, the results showed that signage had a negligible impact on creating a trusting environment. Rather, the results suggest that trust is more likely if it is:
Championed by the leadership in the EOC and actively encouraged between all the members of the EOC. This supports the work of Meyerson et al. (1996, p. 191) who state, concerning the contractor as the head of the temporary system; “the contractor may play an important role…in being not only the architect and facilitator of swift trust but also its centurion.”
Made applicable to both local and international staff. This is of pertinence and requires further research; given disaster management often requires international cooperation.
6.8.4. Assumed qualifications Swift trust theory argues that if each member of the temporary system believes the other members to be qualified in their roles, then swift trust can be more resilient. The case study found that the respondents did not know their colleagues’ qualifications prior to the operation. Furthermore, the results showed that the respondents:
Believed their international colleagues were more likely to be qualified than their local colleagues.
Used a colleague’s parent organisation as a substitute for judging a colleague’s qualifications. This is an example of reputation categorisation, which McKnight et al. (1998, p. 480) argue is a process that can occur when forming trust in new organisational relationships. “A person may be perceived as a competent individual because she or he is a member of a competent group”. Also, contrary to Zolin’s (2002) study on swift trust, this research found that reputation categorisation (a form of stereotyping) actually helped trust form rather than hindered it. Similarly, this complimented the work of Alvinius et al. (2010), who argued that for on-the-ground disaster response, swift trust was dependent upon the organisation that a member of the temporary system came from.
280 Used competent role behaviour to determine whether a colleague was actually qualified for their role in the EOC; in accordance with swift trust theory (Meyerson et al., 1996). If a colleague delivered on their work, this would affirm assumptions about them being qualified. However, if a colleague did not produce their work, they were judged to be not qualified in their role.
Relied upon competency as a trust criterion. This meant that competent role performance could impact the resilience of trust by confirming or invalidating a respondent’s initial assumption about a colleague’s qualifications. This confirms the work of Meyerson et al. (1996) and McKnight et al. (1998, p. 484), with the latter stating, “unless [a colleague] does something to seriously violate the positive expectations a [person] develops from the initial…interaction, the [person’s] trusting intention is likely to endure”.
Focussing on the field of disaster management, the results also provided a valuable contribution by showing that international disaster management personnel can simply be assumed to be qualified by their local counterparts. This finding has two implications. Firstly, given the use of competent role behaviour in confirming a colleague’s qualifications, it is in an international organisation’s best interest to ensure that they send qualified personnel to a response; given that incompetence can result in distrust in the international organisation (as shown with OCHA).
Secondly, international organisations must not exploit the fact that their staff may be assumed to be qualified. This specifies the sociological trust theory of B. Barber (1983) to a disaster management context, where they write that ‘individuals within a society must trust that those with knowledge and expertise are not going to abuse this power’. Specifically, it also points to a much larger problem currently within the disaster management sphere, which Telford and Cosgrave (2007) state as a lack of appropriately qualified international disaster personnel. Considering the response to the 2004 Indian Ocean tsunami, Telford and Cosgrave (2007, p. 12) stated that there were, “major weaknesses in international staff profiles, staff quality, and continuity. These weaknesses are highlighted year after year.” Similarly, Scheper, Parakrama, and Patel (2006) argue that poor strategic planning by international disaster organisations led to the recruitment of inexperienced staff in an effort to achieve quick results after the tsunami. In light of these authors’ research, the case study has shown that it is imperative that an international organisation ensures their staff are qualified before being sent to a response. This is of particular importance because they are likely to be assumed to be qualified and, “…poor results from an incompetent international can impact the wider perception of the international
281 community” (Telford & Cosgrave, 2007, p. 12). This also points to the larger argument that the case study highlighted in Section 5.3.2, which is that the international community should also work in partnership with local actors (Scheper et al., 2006), both prior to the disaster to reduce the amount of surge staff needed, as well as during the response and recovery stages. Furthermore, they should not use their assumed expertise as a means to run the operation; as R. Barber (2015) argues happened in the response to Cyclone Pam in Vanuatu.
6.8.5. Hedges Within the EOC, there were two hedges recorded, however neither of these were perceived by the wider EOC and there were no observable impacts on the resilience of trust within the response. The results refined the concept of hedges as described by Meyerson et al. (1996) by showing that hedges do not always lead to increased personal confidence as well as a reduced level of group commitment. Additionally, the data also showed that hedges can be perceived and justified differently by their respective holders; and contrary to swift trust theory, they are not always used to reduce personal vulnerability when in a temporary system.
Secondly, it is necessary to consider the purpose of hedges as described by Meyerson et al. (1996) in relation to the results and wider literature. As swift trust theory defines, a hedge is a ‘way out’ in order to, “reduce the perceived risks and vulnerabilities of trust” (Meyerson et al., 1996, p. 187). However, as identified in Chapter 2, what the disaster management literature is lacking is an understanding of the risks associated with trust when working on the national response and recovery. That is, what did the respondents consider to be the risks specific to the operation that required trust. Addressing this question, the interview data showed that the respondents considered the following factors to be risks for the operation:
The security and dissemination of information.
The reputation of the Government.
A perceived lack of productivity in delivering the response.
A perceived lack of control over running the operation locally (which was reinforced after the breach of trust by OCHA when they published data without the EOC’s permission).
Albeit not an extensive list, these factors provide an original contribution to the literature by highlighting specific risks that require trust when working on a multi-national response. However, it is important to recognise that a hedge is unable to mitigate against any of these
282 risks at an operational level. Rather, in accordance with the definition of Meyerson et al. (1996), a hedge is a form of personal insurance for a respondent if they were to cause one of these risks to eventuate. Thus, the case study suggests that the role of hedges (in being able to mitigate against risks) is limited in disaster management given that there are risks associated with running a response that do not make any one person vulnerable, but rather the reputation of the operation as an entity.
Finally, the data also revealed an additional refinement to the use of hedges in swift trust theory. Although hedges may provide a ‘way out’ for an individual, they are unable to mitigate against reputational damage; should the reason a person needs a way out be because they were untrustworthy and damaged the operation in some capacity. This is of particular pertinence to the field of disaster management, given that the network of respondents is small and future work is likely; through training exercises (Chokshi et al., 2014; R. W. Perry, 2004) and on-going collaboration and planning (Roberts et al., 2011; Simo & Bies, 2007). These factors combine to inhibit the use of hedges, as they are unable to mitigate the risk of damage to a person’s reputation. Also, this shows that there may be a correlation between the size of the network the temporary system is formed from (labour pool) and the use of hedges; which requires further research.
6.8.6. Improvisation The results showed that improvisation was an accepted part of the response and recovery. This was due to the nature of running a national operation, as well as the Standard Operating Procedures for the EOC being out of date and lacking detail. Only three respondents stated that they would not improvise during the response, as they felt that they had to report any operational problems to their superiors. Additionally, the results also showed that if improvisation was to affect the resilience of trust in the EOC, it would be dependent on a colleague improvising proficiently. This extends the work of Meyerson et al. (1996) by showing that improvisation itself does not necessarily increase the resilience of trust; but rather, in accordance with the wider argument in swift trust theory of competent role behaviour, proficiency when improvising is important.
Secondly, considering disaster management, the results showed that improvisation must be balanced against the internal hierarchy and reporting process of the EOC, and the applicability of the Standard Operating Procedures. That is, using Harrald’s (2006) typology, disaster management should approach a response in a ‘balanced and adaptive’ manner; that balances
283 discipline (Standard Operating Procedures and the internal reporting hierarchy) with agility (the ability to improvise and change).
The results also showed that the EOC was subject to a hierarchical reporting structure; where some members felt that to improvise would be to bypass this structure. The internal reporting structure of the EOC helped protect the response from operational risks such as: information security and dissemination, ensuring a unified message was presented to the public, and that information was passed through the appropriate channels (as explained in Section 6.8.5). However, improvisation is necessary in disaster management given the time constraints, and should not be limited only to those in positions of leadership or seniority; as the case study shows. As G. Webb (2004, p. 52) argues, “individual role incumbents sometimes short-circuit bureaucratic decision-making processes…when [they] perceive the demands of the situation to be such that following routine procedures would cost valuable time”. Conversely, the ‘short- circuiting of bureaucratic decision-making processes’ (G. Webb, 2004) can cause distrust; as was demonstrated with the breach of trust by OCHA when they bypassed the leadership of the EOC and published response data without approval. As such, the case study shows that hierarchy and reporting lines are important in natural disaster management to preserve the integrity of the EOC as the organisation charged with running the operation. However, they need to be balanced against improvisation, so that the response is not unnecessarily slow.
Additionally, the results showed that the Standard Operating Procedures for a national operation must not only be up to date and deemed relevant by their users, but they also must incorporate improvisation as an accepted part of a response. This affirms Webb’s (2004, p. 60) argument, where they write that, “effective responses to disasters involve flexibility…[and] those responsible for preparing communities and organisations for disasters should plan for improvisation to occur…[and build] flexibility into their organisational and community disaster plans.” Put simply, the Standard Operating Procedures for a national response should provide sufficient guidance whilst also recognising that, “emergency management officials should plan to improvise” (G. R. Webb & Chevreau, 2006, p. 70).
6.8.7. Focussing on the task Considering this final resilience factor, the results showed that the respondents were focussed on the task of responding and had minimal, if any, additional time to develop harmful interpersonal dynamics given the requirements of the operation. Furthermore, apart from the two breaches of trust recorded in Section 5.1.2, there were no other recorded instances of harmful interpersonal dynamics between the respondents. Therefore, as the research of
284 Meyerson et al. (1996) and Jarvenpaa and Leidner (1998) claim, trust can be resilient in the temporary system when the members are focussed on achieving the task of responding. This concept also validates the research of Majchrzak et al. (2007), who found that swift trust is developed through action.
However, provided the time constraints on the EOC, the case study also showed that there are risks if the respondents prioritise speed over due process when delivering the response. Such behaviour can actually lead to distrust; as was demonstrated by the breach of trust by OCHA. This points to a larger concept in the disaster management literature addressed by Scheper et al. (2006), who argues that focussing on the task at hand too intensely can result in the wider strategic approach being forfeited. They commented particularly on the poor strategy of including inexperienced international personnel into the 2004 Indian Ocean tsunami response, “Pressure for quick results not only led to recruitment of inexperienced staff but also such pressure on the most experienced so that they too were drawn into the result-oriented mode. All this reflected underlying problems of strategy” (Scheper et al., 2006, p. 36). However, their argument points more generally to a specific consideration for the application of swift trust theory to temporary systems, such as those within disaster management. That is, although Meyerson et al. (1996) argue that focussing on the task of responding may help make trust more resilient, such an approach must not substitute the need for strategic management; especially when considering a national operation. Rather, as shown by the EOC and OCHA, focussing on the task at hand without considering the wider strategy can lead to the erosion of trust.
The applicability of the swift trust resilience factors for trust in the EOC The purpose of this chapter was to determine if the resilience of trust in the could be determined by the swift trust theory resilience factors. Foremost, it is important to recognise that the trust displayed by the respondents within the EOC was not modelled by the theory of swift trust. As such, it would seem logical to assume that the theoretical resilience factors, which are customised for swift trust in temporary systems, would not apply. However, this was not uniformly the case. The results showed that the development, and sustaining of trust (resilience), was impacted by:
the respondents assuming their colleagues were qualified, and this being demonstrated through competent role behaviour;
improvisation being a necessary part of the operation; and,
285 the respondents focussing on the task of undertaking and coordinating the response.
Yet, using the paradigm of social constructionism and interpretivism, the results show that swift trust theory does not adequately address how the resilience of trust can be impacted by the relational side of trust throughout the duration of the temporary system. That is, if resilience is defined as an attribute of trust which enables it to, “survive those moments and incidents that occur during the life of a temporary group… [that] call into question or threaten to disrupt trust” (Meyerson et al., 1996, p. 184), then it must also focus on the relational (dyadic) qualities of trust. It is these qualities, independent of the temporary system, that are just as likely to ‘threaten or disrupt trust’. And perhaps this was outside the scope of swift trust theory, but it remains an element of trust that cannot be excluded when considering resilience as a unified concept. As such, it is recommended that further research be conducted in order to understand how the subjective and relational aspects of trust within an EOC impact its resilience. This should be done with particular consideration of the characteristics of the EOC as a temporary system and the wider disaster management context.
286
CONCLUSION
The aim of this research was to understand what factors impacted trust formation for disaster management personnel within an EOC during a natural disaster response. This was investigated through the theory of swift trust in temporary systems. This chapter provides an overview of the literature and then summarises each of the three research propositions. It then explains the research contributions, which addresses the research aim and question. Finally, it considers the limitations of the case study and further research that could be undertaken to better understand trust formation in an EOC.
Overview of the literature The literature in Chapter 1 showed that developing nations are disproportionally impacted by natural disasters, experiencing higher mortality rates and economic losses as a percentage of national GDP than the developed world. These losses are only expected to worsen due to urban growth; with the developing world expected to absorb 95% of the global projected urban growth over the coming decades. In order to recover from a natural disaster, many developing nations seek international assistance. However, there is little research considering how trust forms in multi-national teams (with local and international disaster management personnel) working on such a response. Therefore, the aim of this research was to explore what factors impact trust formation for local and international disaster management personnel who are working together in the response and recovery after a natural disaster, in a developing nation.
More specifically, the research explored:
1. What are the characteristics of the temporary system of local and international disaster management personnel in the EOC, charged with coordinating the response?
2. What factors influence trust formation and are these factors different for local and international disaster management personnel?
287 3. What factors may impact the resilience of trust formed for local and international disaster management personnel?
However, addressing the research aim was complicated by the fact that trust has many definitions across various disciplines. Much of the disaster management literature has only considered that trust could be affected by a natural disaster, without considering how it may form in that context. Conversely, any research that has considered how trust forms, and the antecedents of trust, has not been applied to a natural disaster context. Consequently, adopting a social constructionism and interpretivism paradigm, this research sought to test whether the theory of ‘swift trust in temporary systems’ (Meyerson et al., 1996) would be an applicable construct through which to understand how trust forms within an EOC during a natural disaster operation.
On the basis of swift trust theory and the sub-questions above, three research propositions were put forward. Data from these propositions, when combined, provides a detailed understanding of how trust formed within the EOC. In addressing these propositions, the case study also provided important insights into the antecedents of trust for local and international disaster management personnel, as well as contributing to the wider trust and disaster management literature. Therefore, the next section discusses both the key findings for each research proposition and their contribution to knowledge.
Research proposition 1 During the response and recovery to a disaster in a developing nation, an EOC exhibits the properties of a temporary system.
The purpose of research proposition 1 was to determine whether the EOC was a temporary system in accordance with the criteria proposed by Meyerson et al. (1996). This was necessary because the trust criteria asserted by swift trust theory rely upon the conditions of a temporary system for their relevance. Overall, the results found that during the response and recovery to TC Winston, in accordance with the research of Goodman (1981) and Meyerson et al. (1996), the EOC qualified as a temporary system. Therefore, in addressing the research proposition and considering each temporary system criterion, the case study makes the following contributions to knowledge:
The response and recovery to TC Winston was a non-routine task that required the formation of the EOC. In particular, the results found that interdependence was one characteristic of this temporary system (which Meyerson et al. (1996) considers
288 important for trust formation). However, this research elaborates on the research of Meyerson et al. (1996), demonstrating that two forms of interdependence existed within the EOC: personal interdependence and team interdependence. This research also provides a unique contribution concerning interdependence in the EOC: the results found that the degree of seniority and expertise of the respondents within the EOC impacted their degree of interdependence; furthering the research of Militello et al. (2007) by examining interdependence in a real-time disaster operation.
It was found that the EOC had clear and specific goals for the operation. This contributes to the research of Kendra and Wachtendorf (2003) by showing that the goals of the operation must not only be known by the disaster management personnel within the EOC (as these theorists would suggest), but that they should also be specific. These results also supported the assertions of Jones and Faas (2016, p. 170), who argue that in a disaster response, “trust in relationships is enhanced by…identification with a common cause.” Furthermore, the research responds to Haas et al. (1977), by suggesting that having clear and specific goals can help prevent the formation of competing ideas about the purpose of the recovery.
This case study extends the work of Meyerson et al. (1996) by showing that diversity in an EOC, as a temporary system, includes taking on additional responsibilities outside of one’s area of expertise and working longer hours due to the operational requirements of the national response. This finding unites the work of Weick (1993) on ‘virtual role systems’, with that of Lalonde (2004), by showing that, in the context of an EOC as a temporary system, personnel take on additional roles outside of their area of expertise. The results also present a unique contribution to the temporary system and disaster management literature by providing a list of the roles of respondents involved in an EOC, as well as each role’s unique expertise. Particularly, in support of R. W. Perry (1995) and Carter (2008), it was found that information management was the most utilised skill within the EOC. However, this case study furthers these authors’ work by outlining that national and international correspondence and leadership, as particular skills, were required by multiple teams within the temporary system of an EOC.
In support of Meyerson et al. (1996), it was found that an EOC, as a temporary system, consists of personnel drawn from a small and overlapping network of disaster organisations. This conclusion refines the findings of Waugh and Streib (2006) by showing that operational and coordinating roles in the EOC can be undertaken by a small
289 number of personnel. Furthermore, the results further the work of Meyerson et al. (1996) by emphasising the importance of an individual’s degree of interconnectivity in a temporary system; as opposed to only network interconnectivity (as per swift trust theory). The case study shows that some respondents had higher degrees of personal interconnectivity, which meant they had more reach across different organisations, enabling them to easily share untrustworthy behaviour were it to occur.
Contrary to Goodman (1981) and Meyerson et al. (1996), it was found that an EOC is unique to other temporary systems as it is not completely ‘new’, but rather is the combination of permanent disaster management personnel, with surge staff brought in for the specific purpose of helping during a national operation. This conclusion also compliments the work of Kendra and Wachtendorf (2003), Waugh and Streib (2006), and Kapucu (2008) in the disaster management context, who argue that pre-existing relationships are beneficial for the coordination of a national response. Similarly, many of the respondents expected to work together again in another operation, through the EOC outside of a disaster, or through training exercises (Chokshi et al., 2014; R. W. Perry, 2004), fulfilling this criterion of a temporary system (Meyerson et al., 1996).
Contrary to Meyerson et al. (1996), it was found that the head of the EOC, as the contractor, may not assemble the respondents to be a part of the temporary system. Rather, as supported by the wider literature considering the types of disaster management stakeholders (Curnin & Owen, 2013; Kendra & Wachtendorf, 2003; R. W. Perry, 2003; E. L. Quarantelli, 1988), it was found that some temporary systems, such as an EOC, do not have the opportunity to be created using personnel assembled by the contractor once a state of emergency is declared. However, although the contractor did not contribute to the formation of the EOC, they were charged with coordinating the operation (Waugh & Streib, 2006). This is a useful insight and case study contribution to disaster management literature, which may be less concerned with temporary system formation and more concerned with the details of authority and management structures within an EOC.
Apart from addressing the specific criteria of a temporary system, the analysis of the EOC as a temporary system also contributes to the wider disaster management literature through the following findings:
In support of E. L. Quarantelli (1999), the EOC confirmed the concept of emergence as an important aspect of disaster management. Emergence was shown through: the 290 creation of Current Operations and Future Operations, the adoption of a new reporting structure, and the permanent EOC staff taking on new response-specific roles.
The inclusion of Current Operations and Future Operations in the response and recovery provides a contemporary example of responding to the dynamic requirements of a natural disaster with ingenuity (Kendra & Wachtendorf, 2003), flexibility (Mendonca et al., 2001), and innovation and adaptation (Waugh & Streib, 2006).
This research’s attention to the expanded role of Future Operations within the operation extends the literature concerning military involvement in a natural disaster response and recovery. It also provides a unique contribution to the disaster management literature concerning governance of the response and recovery, outlining an example in which the military were held accountable to government authority within a national operation (Walker, 1992). These results also contradict the findings of Anderson (1970), showing that the military can take on an operational role which extends beyond the first few days of the disaster and emergency period.
Considering military involvement, the case study supports the findings of Birkmann et al. (2010), showing that a request for international aid and the resulting multi-national operation, can be used to help repair international diplomatic relationships.
Research proposition 2 The trust that develops between disaster management personnel within an EOC is in accordance with the trust criteria proposed by swift trust theory.
Research proposition 2 tested whether the trust that developed between the respondents within the EOC was in accordance with the trust criteria proposed by swift trust theory. The results found that although trust formed quickly between the respondents, the theoretical antecedents of trust described by swift trust theory generally did not align with how the respondents formed trust. However, the case study showed that a respondent’s competency, personal reputation, role-based interactions, as well as their degree of interdependence are all characteristics of swift trust theory that should be considered when setting up and operating an EOC. In addition to swift trust theory, the results also highlighted the importance of additional trusting criteria, such as culture, as important factors that can influence trust formation and that should be considered when forming and running an EOC. Nonetheless, even though swift trust theory did not exactly map trust formation within the EOC, the results still provided a valuable
291 contribution to the application of swift trust theory to an EOC, as well as providing significant insights into how trust forms in this environment.
In agreement with Meyerson et al. (1996), the results showed that maintaining a trustworthy reputation (professional and/or personal) was important to the respondents within the EOC; as a trustor and trustee’s concern for their reputation discouraged untrustworthy behaviour. Diverging from the work of Rousseau et al. (1998), Mishra (1996) and McKnight et al. (1998), which focusses on how a professional colleague’s reputation affects trust formation, the case study highlights the importance of a member of the EOC maintaining their personal reputation, and the subsequent impact that maintenance of this reputation has on trust formation in this context.
Similarly, the case study’s documentation of the breach of trust by OCHA, and the subsequent damage to their reputation as an organisation, extended swift trust theory (Meyerson et al., 1996) by emphasising the importance of international organisations adhering to the governance of the nation they work alongside. This is because, as Stephenson Jr (2005) argues, on-going reputational damage, and ultimately distrust, can be caused by a single incident.
The case study showed that, except for those in a position of leadership, the weight assigned to knowing a colleague’s role and subsequently them adhering to that role was less than what swift trust theory would predict (Meyerson et al., 1996). This conclusion also disagrees with the findings of Curnin et al. (2015), who argued that role clarity was important in an EOC, and therefore must help swift trust form. Rather than emphasising role clarity, the results of this case study support the work of Turoff et al. (2004) in arguing that the respondents: were more concerned with their colleagues meeting the requirements of the operation, assumed their colleagues were multi-skilled, or did not feel it was their prerogative to challenge their colleagues taking on multiple roles. The case study results also provide an original contribution to the trust literature by refining the use of Gambetta’s (2000) theoretical model of uncertainty in swift trust formation in a disaster management context. The results show that knowing a colleague’s role was less important in reducing uncertainty and forming trust for a respondent than initially suggested.
Many respondents started from a position of trust (Luhmann, 1979; Wiethoff & Lewicki, 2005), emphasising the importance of trusting disposition as an important factor in trust
292 development (Mayer et al., 1995). However, the results also added another perspective to the broader trust literature by showing that those in leadership in the EOC may not be disposed to trust; being more hesitant in order to protect the wider operation.
Rather than aligning with swift trust theory (Meyerson et al., 1996), the results showed that trust formation in an EOC is more aligned with the organisational trust literature; where the three primary antecedents of trust are competency (ability), integrity, and benevolence (Ferrin & Gillespie, 2010; Lewicki & Tomlinson, 2003; Pirson & Malhotra, 2011; Rousseau et al., 1998; Saunders et al., 2014; Wiethoff & Lewicki, 2005). Additional, less cited trust antecedents included a colleague’s: appearance, parent organisation, seniority and authority, degree of cooperation, motivations, and work ethic. Considering these antecedents of trust, the results diverge from the research of McKnight et al. (1998), who argued that stereotyping is an important aspect in forming trust early on. Furthermore, divergent from swift trust theory (Meyerson et al., 1996), the case study showed that although an EOC operates under time pressures, disaster management personnel may still favour accuracy over speed when deciding who to trust.
In support of Meyerson et al. (1996), it was found that categorical assumptions such as local culture and the military can also impact how a respondent (particularly locals in the EOC) form trust. These findings validate the work of Kramer and Lewicki (2010), MacDuffie (2011) and Lewicki et al. (1998), who stated that trust can be influenced by context and social similarities.
As the basis of their trust toward international colleagues, locals look for these colleagues to integrate into the EOC with humility, by working alongside and under the national EOC leadership. This reinforces Hilhorst’s (2005) argument, which states the importance of international disaster management personnel being accountable to local leadership when forming trust.
The results show that although a member of an EOC may have experienced unexpected behaviour from a colleague in the operation, the bearing this had on the formation of trust was connected to the value ascribed to the trust criterion impacted by their colleague’s behaviour. This is an original contribution to swift trust theory (Meyerson et al., 1996). In the same vein, it was found that not all trust criteria are weighted equally, and that it may be easier to determine the importance of a trust criterion to a respondent by observing examples of distrust.
293 The case study clarifies the concept of interdependence explored by swift trust theory (Meyerson et al., 1996), by showing that interdependence in the EOC (as a temporary system), was a function of: a member’s direct working relationships with their colleagues, how reliant a member believes they are on their colleagues in the EOC, and the member’s inclusion in the EOC. Particularly, the case study affirms the impact of group inclusion on trust formation, as the research of Kramer and Lewicki (2010), MacDuffie (2011), McKnight et al. (1998), M Williams (2001), Brewer (1981) and Popa (2005) would suggest. The results confirm that in the EOC context, holding a common purpose and being identified with a group can lead to members of that group perceiving each other as more trustworthy.
Contrary to Meyerson et al. (1996), the reputation of the Director (the contractor) was not a substantial contributor to trust development. This showed that the theoretical concept of an intermediary, and their impact on trust formation, is less significant in an EOC than some theorists would suggest (Coleman, 1990). However, this case study does affirm that the head of the EOC can impact trust formation by: displaying leadership and sound decisions, showing discernment, selecting appropriate personnel to be included in the response, and through endorsing a colleague. These factors offer specific traits of a disaster management leader that, as Schaubroeck et al. (2011) argue, can contribute to that leader being trusted by their peers.
As with Research Proposition 1, the analysis of swift trust in the EOC also contributed to the wider disaster management literature through the following findings:
Informal sanction can exist, be enforced, and influence behaviour within an EOC. Informal sanctions can include measures such as an unofficial decision to exclude a member or organisation from the response, or the risk of reputational damage, which limits an EOC member’s agency in that context.
In support of Thévenaz and Resodihardjo (2010), it was found that international disaster management personnel must be conscious of their involvement in a national operation in a developing nation. In order for trust to be most effectively established in an EOC with a multi-national team, the response and recovery should be locally led (Lam et al., 2010), using local expertise (M. Perry, 2007), with locals not being excluded from the operation by their international colleagues (Hilhorst, 2005).
294 Some of the risks that influence trust when working on a multi-national response include: the security and dissemination of information, the reputation of the host Government, a perceived lack of productivity in delivering the response, and a perceived lack of control over running the operation locally.
Research proposition 3 The resilience of trust in a disaster management EOC is determined by the resilience criteria proposed by swift trust theory.
Although trust criteria used in the EOC were not in accordance with the trust criteria proposed by swift trust theory, this research still sought to test the applicability of swift trust theory’s resilience factors. The results showed that, although many of the swift trust resilience factors relied on attributes of a temporary system, only some of them were applicable in the context of the EOC. Regardless, the case study still provides a valuable contribution to understanding what factors could impact the resilience of trust within an EOC.
The results showed that as the Director (as the contractor) did not select all the personnel to be included within the EOC, this factor was not able to act as a substitute for interpersonal history between the respondents when forming trust. However, similarly to Hyllengren et al. (2011), the results showed that the reputation of the Director could influence trust formation, and thus also influence its resilience.
Given an EOC is comprised of both permanent and surge staff, it was shown that how the EOC (and a temporary system more generally) is formed can impact trust formation and the resilience of trust in that system. Particularly, the case study provides a contribution to Meyerson et al. (1996) by considering a unique type of temporary system built around a pre-existing group of permanent EOC employees; who perform new roles and form additional trusting relationships with surge staff.
Contrary to the Meyerson et al. (1996) theory of swift trust, the case study showed that optimism did not impact the resilience of trust. Swift trust theory would suggest that a respondent in the EOC would expect an operational problem in the EOC to impact their colleague and not themselves. However, those in positions of leadership in the EOC had a realistic expectation that they would be impacted by operational problems that arose during the response and recovery. Similarly, some respondents felt a shared responsibility for any operational problem in the EOC, and that the consequences of a problem should not be borne solely by their colleague. Therefore, the results challenge
295 the use of Weinstein (1980) as a theoretical foundation for optimism in swift trust; as it was concluded that, contrary to the author’s argument, a respondent’s degree of optimism did not correlate with their level of perceived control.
Although the literature shows that trusting disposition is important for trust formation (B. Barber, 1983; Gambetta, 2000; Hung et al., 2004; McKnight et al., 1998; Tatham & Kovács, 2010), the case study showed that it cannot be assumed that members of an EOC as a temporary system will start from a position of trust (i.e. act initially as if trust were present); a factor which Meyerson et al. (1996) argue can make trust more resilient. These results oppose the suppositions of swift trust theory (Meyerson et al., 1996). It was shown that some respondents did not have a trusting disposition due to a consideration for their personal vulnerability, or because they wanted to protect the EOC.
As per swift trust theory (Meyerson et al., 1996), the case study showed that a colleague’s assumed qualifications could impact the resilience of trust. In support of McKnight et al. (1998) and Alvinius et al. (2010), it was found that in this case study, a colleague’s parent organisation was sometimes used as a substitute for knowing that colleague’s qualifications. This however, contradicts the research of Zolin (2002), showing that organisational stereotyping can actually help trust form and make it more resilient.
The results observed that hedges did not impact the resilience of trust in the EOC. Rather, the results refine the concept of hedges as described by Meyerson et al. (1996) by showing that hedges do not always increase a person’s personal confidence, or reducing their level of team commitment. Furthermore, the case study also highlights the inadequacy of hedges, as described by Meyerson et al. (1996), in being able to mitigate against: damage to a person’s reputation, and them being perceived as untrustworthy; or organisational risks experienced by the EOC.
The results extend the work of Meyerson et al. (1996) by showing that the resilience of trust may be impacted by how competently a respondent can improvise during the operation. In support of G. R. Webb and Chevreau (2006), the case study showed that improvisation was necessary in the response and recovery (G. R. Webb & Chevreau, 2006). However, some respondents felt unable to improvise due to the reporting hierarchy within the EOC. Therefore, the results showed that a national response should
296 be approached, as Harrald (2006) describes, in a balanced and adaptive way; allowing both discipline and agility. Furthermore, the Standard Operating Procedures for a national operation must not only be up to date and deemed relevant by their users, but, as G. Webb (2004) argues, they also must incorporate improvisation as an accepted part of the response. This is an important finding relevant to the disaster management literature.
In agreement with Meyerson et al. (1996), Jarvenpaa and Leidner (1998) and Majchrzak et al. (2007), it was found that the resilience of trust could be impacted by the respondent’s ability to stay focussed on the tasks of the operation. However, it was also found that bypassing the reporting protocols of the EOC, in order to quickly produce response data, can lead to distrust. It is important, particularly for international organisations, to: follow due process, respect the procedures put in place by the EOC, and not undermine (implicitly or explicitly) the authority and control of the local leadership (Hilhorst, 2005).
The analysis of Research Proposition 3 also highlights an additional contribution of knowledge concerning the involvement of international disaster management personnel in a multi-national response. The case study found that international personnel sent to aid a response in a developing nation will generally be assumed to be qualified. However, this must not be exploited by any international organisation; it is in their best interests that their personnel are perceived to be competent (Scheper et al., 2006). If an international colleague is perceived to be incompetent, then this can lead to damaged relations and distrust in that organisation; and potentially the wider international community (Telford & Cosgrave, 2007). This is supported by the results of this case study research.
Research limitations This research was limited to the study of swift trust within the temporary system of an EOC. Thus, the results are contextualised to Fiji and trust formation between particular respondents involved in the response and recovery following TC Winston. It was not feasible to conduct a multi-case study analysis provided the unpredictability of natural disaster occurrence and the practical challenges of gaining access to, and conducting research in, an EOC. However, this does not detract from the validity and value of the results. Rather, the adoption of a single case study methodology meant that an in-depth understanding of the phenomenon of trust and its formation within an EOC was possible. The applicability of these findings to other EOCs responding to a natural disaster through further research is encouraged.
297 The research also recognises the limitations imposed by the use of social constructionism and interpretivism, as well as those occurring from applying a qualitative analysis subject to the researcher’s interpretation of the data (and relationships built through the interviews). Such limitations mean that the data, and conclusions drawn from it, are only representative of one interpretation of social reality. This limitation was mitigated, as far as possible, through considering reflexivity and any natural bias when approaching the research topic.
Finally, the research was also limited in its ability to observe personnel operating within the EOC during the initial stages of the response. This was simply due to the time required to gain permission to conduct research in such a dynamic and restricted location. Therefore, the interview was conducted with the respondents retrospectively considering trust formation. That said, one respondent stated that it would not have been possible to conduct interviews during the initial stages of the response, due to the intensive requirements of the operation, which led to the respondents’ lack of availability at that time.
Further research The case study provided an in-depth analysis of trust formation within the EOC after TC Winston. Particularly, it tested the applicability of the theory of swift trust in temporary systems as a basis from which to model trust formation within an EOC. Although it was found that swift trust was not applicable in the context of the EOC (as the trust criteria used by EOC respondents did not align with all swift trust criteria in the Meyerson et al. (1996) model), the case study highlighted multiple areas regarding swift trust theory, and trust formation and disaster management after a natural disaster, which could benefit from additional research.
7.6.1. Further research into swift trust in temporary systems Foremost, this research recommends that additional case studies of the applicability of swift trust in an EOC be undertaken. This is not to promote generalisability, but rather to promote a better understanding of the contextual and subjective nature of trust within a disaster management setting, and how local and international disaster management personnel build trust. Given that the EOC, in this case, was found to be a temporary system, further research should be undertaken to test whether the theoretical trust antecedents of swift trust theory exist within other EOCs subject to a national operation.
It is also recommended that further research be undertaken to consider the influence of other theoretical models of trust and how they could be combined with swift trust
298 theory to provide a more accurate picture of trust formation within an EOC. As an example, according to Parsons (1963), the EOC could be considered a small ‘social system’, where trust is formed by the consensual sum of the individuals within that system; as was demonstrated by the weight assigned by locals towards their international colleagues humbly integrating into the response. Similarly, Barber (1983) argues that trust is dependent upon cognitive and moral expectations; such as this expected humility. Further research should be undertaken to potentially refine swift trust theory using the theoretical characteristics of other pre-established sociological theories of trust.
Swift trust theory does not provide a timeframe for the applicability of swift trust. Further research should be conducted to ask, ‘how long is it before the trust forming in a temporary system is no longer ‘swift’?’ This is a necessary question given the unique trusting antecedents of swift trust and how they rely on the qualities of the temporary system. This enquiry should be conducted in reference to temporary systems other than an EOC, where all swift trust criteria apply. This is because it remains to be understood when trust formation can no longer be modelled by swift trust (where a temporary system has been established), but rather by a different theoretical approach.
Similarly, swift trust theory does not consider how much evidence-based information is required before trust is no longer swift trust. It is not clear where the tipping-point is between the use of categorical assumptions and evidence-based information; that is, where a person stops using swift trust and starts making an informed decision about trusting their colleague.
Further research is required to incorporate the social dynamics of trust into the contextual resilience factors of swift trust in a temporary system. Having approached the question of resilience through the paradigm of social constructionism and interpretivism, this research argues that to provide a holistic understanding of the resilience of trust, further consideration must be given to the dyadic relationship between the two people involved.
Swift trust theory does not consider a temporary system formed around a pre-existing core team and the impact of this on trust. Further research is required to understand how the composition and assembly of a temporary system can impact how trust forms within it.
299 7.6.2. Further research into trust formation and disaster management after a natural disaster……………………………….. It is recommended that further research be undertaken to understand, of the trust criteria employed by an individual, which are most valued when forming trust in an EOC after a natural disaster.
Further research is also required to identify whether the temporary authority of the head of the EOC, and their selection criteria for including a person in the EOC, can impact trust formation. Additionally, research is also required to explain how explicit the head of the EOC’s selection criteria need to be if they are to both: be accessible to the members of the temporary system, and act as a substitute for interpersonal history between the members of the EOC.
There remains a gap in understanding the degree to which the head of an EOC is able to actually manage team composition, and the subsequent impacts this management factor has on trust formation.
Although it was found that an EOC requires ingenuity and improvisation, there is a lack of research considering the impacts of these flexible processes on how the EOC operates. It is recommended that further research be conducted to understand if structural or procedural changes to the EOC during the implementation of the operation affect the productivity and efficacy of the centre.
The findings of this research suggest that the EOC may be regarded as an ‘informal organisation’. Thus, it is recommended that further research be undertaken to consider the applicability of new institutionalism to an EOC, and whether this theoretical approach can provide additional insights into trust formation within a disaster management context.
It is recommended that research be conducted to gauge the necessity and value of emergence in relation to the overall success of the TC Winston operation. Insights from the EOC’s post-disaster review should be drawn into this analysis.
Finally, in light of the flow on effects from OCHA’s breach of trust, it is recommended that further research be undertaken to consider how the ‘ripple effect’ and emotional contagion theory can impact trust formation in a disaster management context.
300 REFERENCES………………………
ABC-News. (2016). Cyclone Ula: Hundreds evacuated as first Pacific cyclone of 2016 slams into Tonga, heads for Fiji. Retrieved 03/2016, 2016, from http://www.abc.net.au/news/2016-01-02/cyclone-ula-batters-tonga/7064028 ABC/AFP. (2015). Tropical Cyclone Pam: Vanuatu's government criticises aid groups over poor coordination of disaster relief. http://www.abc.net.au/news/2015-03-19/vanuatu- criticises-lack-of-aid-coordination-after-cyclone-pam/6333676 Abrahams, J. (2001). Disaster management in Australia: The national emergency management system. Emergency Medicine, 13(2), 165-173. Al-Nammari, F., & Alzaghal, M. (2015). Towards local disaster risk reduction in developing countries: Challenges from Jordan. International Journal of Disaster Risk Reduction, 12, 34-41. Alexander, D. (1997). The study of natural disasters, 1977–97: some reflections on a changing field of knowledge. Disasters, 21(4), 284-304. Alexander, D. (2008). Emergency command systems and major earthquake disasters. Journal of Seismology and Earthquake Engineering, 10(3), 137. Aljazeera. (2016). Powerful earthquake rocks southern Taiwan. Retrieved 11/2016, from http://www.aljazeera.com/news/2016/02/earthquake-measuring-64-hits-southern- taiwan-160205213442324.html Alvinius, A., Danielsson, E., & Larsson, G. (2010). The inadequacy of an ordinary organisation: Organisational adaptation to crisis through Planned and Spontaneous links. International Journal of Organisational Behaviour, 15(1), 87-102. Anderson, W. A. (1970). Military organizations in natural disaster: established and emergent norms. American Behavioral Scientist, 13(3), 415-422. Andrabi, T., & Das, J. (2010). In aid we trust: Hearts and minds and the Pakistan earthquake of 2005. World Bank Policy Research Working Paper Series. Anglin, R. (2011). Chapter 4: Disaster Recovery Managers. In E. J. Blakely, E. L. Birch, R. V. Anglin & H. Hayashi (Eds.), Managing Urban Disaster Recovery: policy, planning, concepts and cases. Berkshire, UK: Crisis Response Publications. Aquilino, W. S. (1994). Interview mode effects in surveys of drug and alcohol use: A field experiment. Public opinion quarterly, 58(2), 210-240. Baier, A. (1986). Trust and antitrust. Ethics, 96(2), 231-260. Barber, B. (1983). The logic and limits of trust. USA: University of Toronto Press Inc.
301 Barber, R. (2015). One Size Doesn't Fit All - Tailoring the International Response to the National Need Following Vanuatu's Cyclone Pam. Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47(4), 644-675. Bates, D. C. (2002). Environmental refugees? Classifying human migrations caused by environmental change. Population and environment, 23(5), 465-477. Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The qualitative report, 13(4), 544-559. BBC, N. (2016). Fiji profile - timeline. http://www.bbc.com/news/world-asia-pacific-14919688 Becker, H., & Geer, B. (1957). Participant observation and interviewing: A comparison. Human organization, 16(3), 28-32. Benson, C., & Clay, E. J. (2004). Understanding the economic and financial impacts of natural disasters: World Bank Publications. Berg, J., Dickhaut, J., & McCabe, K. (1995). Trust, reciprocity, and social history. Games and economic behavior, 10(1), 122-142. Berkes, F. (2007). Understanding uncertainty and reducing vulnerability: lessons from resilience thinking. Natural hazards, 41(2), 283-295. Berry, R. S. (1999). Collecting data by in-depth interviewing. Birkmann, J., Buckle, P., Jaeger, J., Pelling, M., Setiadi, N., Garschagen, M., . . . Kropp, J. (2010). Extreme events and disasters: a window of opportunity for change? Analysis of organizational, institutional and political changes, formal and informal responses after mega-disasters. Natural hazards, 55(3), 637-655. Blakely, E. J., Birch, E. L., & Anglin, R. V. (2011). Introduction. In E. J. Blakely, E. L. Birch, R. V. Anglin & H. Hayashi (Eds.), Managing Urban Disaster Recovery: policy, planning, concepts and cases. Berkshire, UK: Crisis Response Publications. Blakely, E. J., Birch, E. L., Anglin, R. V., & Hayashi, H. (2011). Section Two: Social, Cultural and Economic Recovery. In E. J. Blakely, E. L. Birch, R. V. Anglin & H. Hayashi (Eds.), Managing Urban Disaster Recovery: policy, planning, concepts and cases. Berkshire, UK: Crisis Response Publications. Boin, A., & 't Hart, P. (2010). Organising for effective emergency management: Lessons from research. Australian Journal of Public Administration, 69(4), 357-371. Brewer, M. B. (1981). Ethnocentrism and its role in interpersonal trust. Scientific inquiry and the social sciences, 214, 231.
302 Bui, T., Cho, S., Sankaran, S., & Sovereign, M. (2000). A framework for designing a global information network for multinational humanitarian assistance/disaster relief. Information Systems Frontiers, 1(4), 427-442. Bull-Kamanga, L., Diagne, K., Lavell, A., Leon, E., Lerise, F., MacGregor, H., . . . Reid, H. (2003). From everyday hazards to disasters: the accumulation of risk in urban areas. Environment and Urbanization, 15(1), 193-204. Burke, C. S., Sims, D. E., Lazzara, E. H., & Salas, E. (2007). Trust in leadership: A multi-level review and integration. The Leadership Quarterly, 18(6), 606-632. Burr, V. (1995). An introduction to social constructionism: London: Routlegde. Campbell-Lendrum, D., & Corvalán, C. (2007). Climate change and developing-country cities: implications for environmental health and equity. Journal of Urban Health, 84(1), 109- 117. Carter, W. N. (2008). Disaster management: A disaster manager's handbook. Cassar, A., Healy, A., & Von Kessler, C. (2011). Trust, risk, and time preferences after a natural disaster: experimental evidence from Thailand. Castillo, M., & Carter, M. (2011). Behavioral responses to natural disasters. Unpublished manuscript. Chandrasekhar, D. (2010). Understanding stakeholder participation in post-disaster recovery (case study: Nagapattinam, India). University of Illinois at Urbana-Champaign. Retrieved from https://www.ideals.illinois.edu/bitstream/handle/2142/16018/1_Chandrasekhar_Divy a.pdf Chandrasekhar, D. (2012). Digging deeper: participation and non-participation in post-disaster community recovery. Community Development, 43(5), 614-629. doi: 10.1080/15575330.2012.730538 Chapman, K. (2011). Call for more sanctions against Fiji. http://www.stuff.co.nz/national/politics/5283034/Call-for-more-sanctions-against-Fiji Charvériat, C. (2000). Natural disasters in Latin America and the Caribbean: An overview of risk. http://www.econstor.eu/bitstream/10419/88044/1/idb-wp_434.pdf Chokshi, A., Seyed, T., Marinho Rodrigues, F., & Maurer, F. (2014). ePlan multi-surface: A multi- surface environment for emergency response planning exercises. Paper presented at the Proceedings of the Ninth ACM International Conference on Interactive Tabletops and Surfaces. Clark, D. (2000). World Urban Development: Processes and Patterns at the End of the Twentieth Century. Geography, 85(1), 15-23. doi: 10.2307/40573371
303 Clifford, N., Cope, M., Gillespie, T., & French, S. (2016). Key methods in geography: Sage. Coleman, J. S. (1990). Foundations of social theory. USA: Harvard University Press. Collins, A. E. (2009). Disaster and development: Routledge. Comfort, L. K. (2005). Risk, security, and disaster management. Annu. Rev. Polit. Sci., 8, 335-356. Comfort, L. K., & Kapucu, N. (2006). Inter-organizational coordination in extreme events: The World Trade Center attacks, September 11, 2001. Natural hazards, 39(2), 309-327. Conzo, P. (2014). Trust and Cheating in Sri Lanka: The Role of Experimentally-Induced Emotions about Tsunami. Working Paper. University of Turin. Department of Economics and Statistics: Cognetti de Martiis. Retrieved from http://www.csef.it/WP/wp355.pdf Coppola, D. P. (2006). Introduction to International Disaster Management. Burlington: Burlington : Elsevier Science. Corstange, D. (2009). Sensitive questions, truthful answers? Modeling the list experiment with LISTIT. Political Analysis, 45-63. Cortazzi, M. (2014). Narrative analysis (Vol. 12): Routledge. CRED, C. f. R. o. t. E. o. D. (2014). Where are the population affected by natural disasters located? In CRED (Ed.), CRED Crucnch (36 ed.). Belgium: CRED. Creswell, J. W. (1994). Research design: Quantitative and qualitative approaches: Thousand Oakes: Sage Publication. Creswell, J. W. (2012). Qualitative inquiry and research design: Choosing among five approaches: Sage publications. Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods approaches: Sage publications. Creswell, J. W., & Clark, V. L. P. (2007). Designing and conducting mixed methods research. Crisp, C. B., & Jarvenpaa, S. L. (2013). Swift trust in global virtual teams. Journal of Personnel Psychology, 12(1), 45-56. Cudworth, E. (2003). Environment and society: Psychology Press. Curnin, S., & Owen, C. (2013). Obtaining information in emergency management: a case study from an Australian emergency operations centre. International Journal of Human Factors and Ergonomics, 2(2-3), 131-158. Curnin, S., Owen, C., Paton, D., Trist, C., & Parsons, D. (2015). Role clarity, swift trust and multi- agency coordination. Journal of Contingencies and Crisis Management, 23(1), 29-35. Currao, T. J. (2009). A new role for emergency management: fostering trust to enhance collaboration in complex adaptive emergency response systems. Monterey, California. Naval Postgraduate School. Retrieved from
304 http://calhoun.nps.edu/bitstream/handle/10945/4460/09Dec_Currao.pdf?sequence= 1 Cutter, S. L., Barnes, L., Berry, M., Burton, C., Evans, E., Tate, E., & Webb, J. (2008). A place-based model for understanding community resilience to natural disasters. Global environmental change, 18(4), 598-606. David, M., & Sutton, C. D. (2011). Social research: An introduction: Sage. Davies, M., Guenther, B., Leavy, J., Mitchell, T., & Tanner, T. (2009). Climate change adaptation, disaster risk reduction and social protection: complementary roles in agriculture and rural growth? IDS Working Papers, 2009(320), 01-37. Delhey, J., & Newton, K. (2003). Who trusts?: The origins of social trust in seven societies. European Societies, 5(2), 93-137. Demiroz, F., & Kapucu, N. (2012). The role of leadership in managing emergencies and disasters. European Journal of Economic and Political Studies, 5(1), 91-101. DeWalt, K. M., & DeWalt, B. R. (2011). Participant observation: A guide for fieldworkers: Rowman Altamira. Dilley, M. (2005). Natural disaster hotspots: a global risk analysis (Vol. 5): World Bank Publications. Drabek, T. E., & McEntire, D. A. (2003). Emergent phenomena and the sociology of disaster: lessons, trends and opportunities from the research literature. Disaster Prevention and Management: An International Journal, 12(2), 97-112. Drnevich, P., Ramanujam, R., Mehta, S., & Chaturvedi, A. (2009). Affiliation or Situation: What Drives Strategic Decisionmaking in Crisis Response? Journal of Managerial Issues, 216- 231. Dunn, K. (2005). Interviewing. In I. Hay (Ed.), Qualitative Research Methods in Human Geography
Melbourne: Oxford University Press. ECB. (2007). Building Trust in Diverse Teams. London: The Emergency Capacity Building Project. El-Masri, S., & Tipple, G. (2002). Natural disaster, mitigation and sustainability: the case of developing countries. International Planning Studies, 7(2), 157-175. Elo, S., Kääriäinen, M., Kanste, O., Pölkki, T., Utriainen, K., & Kyngäs, H. (2014). Qualitative content analysis: a focus on trustworthiness. Sage Open, 4(1), 2158244014522633. Elo, S., & Kyngäs, H. (2008). The qualitative content analysis process. Journal of advanced nursing, 62(1), 107-115. EM-DAT. (2015). EM-DAT Database - Advanced Search. Retrieved 12/03/2015, from EM-DAT http://www.emdat.be/advanced_search/index.html
305 Fahy, M. J. (2012). Understanding Swift Trust to Improve Interagency Collaboration in New York City. Monterey, California. Naval Postgraduate School. Retrieved from http://calhoun.nps.edu/bitstream/handle/10945/17362/12Sep_Fahy_Michael.pdf?seq uence=1 Fay, M., Ghesquiere, F., & Solo, T. (2003). Natural disasters and the urban poor. https://www.wdronline.worldbank.org/bitstream/handle/10986/10374/356570ENGLI SH0Oct031321Nat1Dis1EN.pdf?sequence=1 Fehr, E. (2009). On the economics and biology of trust. Journal of the European Economic Association, 7(2-3), 235-266. Fernandez, C. (2016). Quake strikes northeast India, Bangladesh; 11 dead, nearly 200 hurt. Retrieved 11/2016, 2016, from http://news.trust.org//item/20160104013241- etuqz/?source=jtOtherNews3 Ferrin, D. L., & Gillespie, N. (2010). Trust differences across national-societal cultures: much to do, or much ado about nothing? In M. N. Saunders, D. Skinner, G. Dietz, N. Gillespie & R. J. Lewicki (Eds.), Organizational Trust: A Cultural Perspective (pp. 42-86): Cambridge University Press. Retrieved from http://wiki.thomasfrivold.com/images/a/a7/Organizational_Trust.pdf#page=66. Fleming, D. A., Chong, A., & Bejarano, H. D. (2014). Trust and reciprocity in the aftermath of natural disasters. The Journal of Development Studies, 50(11), 1482-1493. Flick, U. (1992). Triangulation revisited: strategy of validation or alternative? Journal for the theory of social behaviour, 22(2), 175-197. Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative inquiry, 12(2), 219-245. Flyvbjerg, B. (2011). Case Study. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage handbook of qualitative research: Sage. Force, P. (2003). Self-interest before Adam Smith: A genealogy of economic science (Vol. 68): Cambridge University Press. Galletta, A. (2013). Mastering the semi-structured interview and beyond: From research design to analysis and publication: NYU Press. Gambetta, D. (2000). Can we trust trust?Trust: Making and breaking cooperative relations (Vol. 2000, pp. 213-237). Oxford: University of Oxford. Retrieved from https://www.researchgate.net/profile/Diego_Gambetta/publication/255682316_Can_ We_Trust_Trust_Diego_Gambetta/links/004635314c0b7d3da3000000.pdf. Gibbs, G. R. (2008). Analysing qualitative data: Sage.
306 Golafshani, N. (2003). Understanding reliability and validity in qualitative research. The qualitative report, 8(4), 597-606. Goodman, R. A. (1981). Temporary systems: Professional development, manpower utilization, task effectiveness, and innovation: Praeger Publishers. Goodman, R. A., & Goodman, L. P. (1976). Some Management Issues in Temporary Systems: A Study of Professional Development and Manpower-The Theater Case. Administrative Science Quarterly, 494-501. Gubrium, J. F., & Holstein, J. A. (2002). Handbook of Interview Research - Part I Forms of Interviewing. In J. F. Gubrium & J. A. Holstein (Eds.), Handbook of Interview Research: Sage. Guha- Sapir, D., Vos, F., Below, R., & Ponserre, S. (2011). Annual disaster statistical review 2010. Centre for Research on the Epidemiology of Disasters. Guha-Sapir, D., Hoyois, P., & Below, R. (2014). Annual Disaster Statistical Review 2013: The Numbers and Trends. In CRED (Ed.). Brussels: CRED. Guha - Sapir, D., Hoyois, P., & Below, R. (2016). Annual Disaster Statistical Review 2015: The Numbers and Trends. In CRED (Ed.). Brussels. Guion, L. A., Diehl, D. C., & McDonald, D. (2001). Conducting an in-depth interview: University of Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, EDIS. Gurenko, E. N. (2004). Catastrophe risk and reinsurance: a country risk management perspective: World Bank Publications. Haas, J. E., Kates, R. W., & Bowden, M. J. (1977). Reconstruction following disaster Reconstruction following disaster: US The Massachusetts Institute of Technology. Haddow, G., Bullock, J., & Coppola, D. P. (2017). Introduction to emergency management: Butterworth-Heinemann. Hannigan, J. (2012). Disasters Without Borders: The International Politics of Natural Disasters. United Kingdom: Polity Press. Harrald, J. R. (2006). Agility and discipline: Critical success factors for disaster response. The Annals of the American Academy of Political and Social Science, 604(1), 256-272. Hay, I. (2000). Qualitative research methods in human geography. Hayashi, H. (2011). Chapter 2: Recovery plans and planning. In E. J. Blakely, E. L. Birch, R. V. Anglin & H. Hayashi (Eds.), Managing Urban Disaster Recovery: policy, planning, concepts and cases. Berkshire, UK: Crisis Response Publications. Helmke, G., & Levitsky, S. (2004). Informal institutions and comparative politics: A research agenda. Perspectives on politics, 2(4), 725-740.
307 Henderson, L. J. (2004). Emergency and disaster: pervasive risk and public bureaucracy in developing nations. Public Organization Review, 4(2), 103-119. Herr, R., & Bergin, A. (2011). Our near abroad - Australia and Pacific islands regionalism. In A.-A. s. p. institute (Ed.). Hewitt, K. (1998). Regions of risk: a geographical introduction to disasters. Hilhorst, D. (2005). Dead letter or living document? Ten years of the Code of Conduct for disaster relief. Disasters, 29(4), 351-369. Hochrainer, S. (2009). Assessing the macroeconomic impacts of natural disasters: are there any? World Bank Policy Research Working Paper Series, Paper 4968. Hosmer, L. T. (1995). Trust: The connecting link between organizational theory and philosophical ethics. Academy of management review, 20(2), 379-403. Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative health research, 15(9), 1277-1288. Hung, Y.-T. C., Dennis, A. R., & Robert, L. (2004). Trust in virtual teams: Towards an integrative model of trust formation. Paper presented at the System Sciences, 2004. Proceedings of the 37th Annual Hawaii International Conference on. Hunter, L. M. (2005). Migration and environmental hazards. Population and environment, 26(4), 273-302. Hyllengren, P., Larsson, G., Fors, M., Sjöberg, M., Eid, J., & Kjellevold Olsen, O. (2011). Swift trust in leaders in temporary military groups. Team Performance Management: An International Journal, 17(7/8), 354-368. IFRC. (2018). Types of disasters: Definition of hazard. Retrieved 10/10/18, from https://www.ifrc.org/en/what-we-do/disaster-management/about- disasters/definition-of-hazard/ İsbir, E. G., & Genç, F. N. (2008). The Role of International Organizations in Disaster Response: The 1999 Marmara Earthquake Case. AMME IDARESI DERGISI, 41(3), 73-97. Janssen, M., Lee, J., Bharosa, N., & Cresswell, A. (2010). Advances in multi-agency disaster management: Key elements in disaster research. Information Systems Frontiers, 12(1), 1-7. Jarvenpaa, S. L., & Leidner, D. E. (1998). Communication and trust in global virtual teams. Journal of Computer-Mediated Communication, 3(4), 0-0. Jones, E. C., & Faas, A. (2016). Social Network Analysis of Disaster Response, Recovery, and Adaptation: Butterworth-Heinemann. Kahn, M. E. (2005). The death toll from natural disasters: the role of income, geography, and institutions. Review of economics and statistics, 87(2), 271-284.
308 Kapucu, N. (2008). Collaborative emergency management: better community organising, better public preparedness and response. Disasters, 32(2), 239-262. Kapucu, N., Arslan, T., & Demiroz, F. (2010). Collaborative emergency management and national emergency management network. Disaster Prevention and Management: An International Journal, 19(4), 452-468. Kapucu, N., & Garayev, V. (2011). Collaborative decision-making in emergency and disaster management. International Journal of Public Administration, 34(6), 366-375. Katoch, A. (2006). The responders'cauldron: The uniqueness of international disaster response. Journal of International Affairs, 59(2), 19. Kellenberg, D., & Mobarak, A. M. (2011). The economics of natural disasters. Annu. Rev. Resour. Econ., 3(1), 297-312. Kellenberg, D. K., & Mobarak, A. M. (2008). Does rising income increase or decrease damage risk from natural disasters? Journal of urban economics, 63(3), 788-802. Kellett, J., & Peters, K. (2014). Dare to prepare: taking risk seriously (pp. 20). London: Overseas Development Institute. Kendra, J. M., & Wachtendorf, T. (2003). Elements of resilience after the world trade center disaster: reconstituting New York City's Emergency Operations Centre. Disasters, 27(1), 37-53. Khan, M. R., & Rahman, M. A. (2007). Partnership approach to disaster management in Bangladesh: a critical policy assessment. Natural hazards, 41(2), 359-378. Kim, B. (2001). Social constructivism. Emerging perspectives on learning, teaching, and technology, 1(1), 16. King, N., Cassell, C., & Symon, G. (1994). Qualitative methods in organizational research: A practical guide. The Qualitative Research Interview, 17. Knack, S., & Keefer, P. (1997). Does social capital have an economic payoff? A cross-country investigation. The Quarterly journal of economics, 1251-1288. Kohlbacher, F. (2006). The use of qualitative content analysis in case study research. Paper presented at the Forum: Qualitative Social Research. Kramer, R. M., & Cook, K. S. (2004). Trust and distrust in organizations: Dilemmas and approaches: Russell Sage Foundation. Kramer, R. M., & Lewicki, R. J. (2010). Repairing and enhancing trust: Approaches to reducing organizational trust deficits. The Academy of Management Annals, 4(1), 245-277. Krefting, L. (1991). Rigor in qualitative research: The assessment of trustworthiness. American journal of occupational therapy, 45(3), 214-222.
309 Kreimer, A. (2001). Social and economic impacts of natural disasters. International Geology Review, 43(5), 401-405. Kreps, D. M. (1990). Game theory and economic modelling Retrieved from http://www.oxfordscholarship.com/view/10.1093/0198283814.001.0001/acprof- 9780198283812-chapter-3 Krippendorff, K. (2012). Content analysis: An introduction to its methodology: Sage. Krippendorff, K. (2013). Content analysis: An introduction to its methodology (Third Edition ed.): Sage. Kumar, A., & Paddison, R. (2000). Trust and collaborative planning theory: The case of the Scottish planning system. International Planning Studies, 5(2), 205-223. Kusumasari, B., Alam, Q., & Siddiqui, K. (2010). Resource capability for local government in managing disaster. Disaster Prevention and Management: An International Journal, 19(4), 438-451. Lalonde, C. (2004). In search of archetypes in crisis management. Journal of Contingencies and Crisis Management, 12(2), 76-88. Lam, C., Lin, M. R., Tsai, S. H., & Chiu, W. T. (2010). A pilot study of citizens' opinions on the Incident Command System in Taiwan. Disasters, 34(2), 447-469. Lee, R. M. (1993). Doing research on sensitive topics: Sage. Lewicki, R. J., & Bunker, B. B. (1995). Trust in relationships: A model of development and decline: Jossey-Bass. Lewicki, R. J., McAllister, D. J., & Bies, R. J. (1998). Trust and distrust: New relationships and realities. Academy of management review, 23(3), 438-458. Lewicki, R. J., & Tomlinson, E. C. (2003). Trust and trust building. Beyond intractability, 305-315. Lewicki, R. J., Tomlinson, E. C., & Gillespie, N. (2006). Models of interpersonal trust development: Theoretical approaches, empirical evidence, and future directions. Journal of management, 32(6), 991-1022. Lewis, J. D., & Weigert, A. (1985). Trust as a social reality. Social forces, 63(4), 967-985. Lin Moe, T., & Pathranarakul, P. (2006). An integrated approach to natural disaster management: public project management and its critical success factors. Disaster Prevention and Management: An International Journal, 15(3), 396-413. Lindell, M. K. (2013). Disaster studies. Current Sociology, 61(5-6), 797-825. doi: 10.1177/0011392113484456 Lindell, M. K., Perry, R. W., & Prater, C. (2007). Introduction to emergency management. Hoboken, N.J. : Chichester: Hoboken, N.J. : Wiley, Chichester : John Wiley distributor.
310 Linnerooth-Bayer, J., Mechler, R., & Hochrainer, S. (2011). Insurance against Losses from Natural Disasters in Developing Countries. Evidence, gaps and the way forward. IDRiM Journal, 1(1), 59-81. Luhmann, N. (1979). Trust and Power (T. Burns & G. Poggi Eds.): John Wiley & Sons Ltd. Luhmann, N. (2000). Familiarity, confidence, trust: Problems and alternatives. Trust: Making and breaking cooperative relations, 6, 94-107. MacDuffie, J. P. (2011). Inter-organizational trust and the dynamics of distrust. Journal of International Business Studies, 42(1), 35-47. Majchrzak, A., Jarvenpaa, S. L., & Hollingshead, A. B. (2007). Coordinating expertise among emergent groups responding to disasters. Organization Science, 18(1), 147-161. Marjanovic, P., & Nimpuno, K. (2003). Living with risk: Toward effective disaster management training in Africa. In A. Kreimer, M. Arnold & A. Carlin (Eds.), Building Safer Cities: The future of disaster risk (3 ed., pp. 14). Washington: The World Bank. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of management review, 20(3), 709-734. Mayring, P. (2014). Qualitative content analysis: theoretical foundation, basic procedures and software solution Retrieved from http://www.ssoar.info/ssoar/bitstream/handle/document/39517/ssoar-2014- mayring-Qualitative_content_analysis_theoretical_foundation.pdf?sequence=1 McAllister, D. J. (1995). Affect-and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of management journal, 38(1), 24-59. McEntire, D. A. (2004). Development, disasters and vulnerability: A discussion of divergent theories and the need for their integration. Disaster Prevention and Management: An International Journal, 13(3), 193-198. McEntire, D. A. (2007). Local emergency management organizations Handbook of disaster research (pp. 168-182). USA: Springer. McKnight, D. H., & Chervany, N. L. (1996). The meanings of trust. McKnight, D. H., Cummings, L. L., & Chervany, N. L. (1998). Initial trust formation in new organizational relationships. Academy of management review, 23(3), 473-490. Mendonca, D., Beroggi, G. E., & Wallace, W. A. (2001). Decision support for improvisation during emergency response operations. International Journal of Emergency Management, 1(1), 30-38. Meyerson, D., Weick, K. E., & Kramer, R. M. (1996). Swift trust and temporary groups. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 166-195): Sage Publications.
311 Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: A sourcebook. Beverly Hills: Sage Publications. Militello, L. G., Patterson, E. S., Bowman, L., & Wears, R. (2007). Information flow during crisis management: challenges to coordination in the emergency operations center. Cognition, Technology & Work, 9(1), 25-31. Miller, R. L., & Brewer, J. D. (2003). The A to Z of social research: a dictionary of key social science research concepts: Sage. Mirza, M. M. Q. (2003). Climate change and extreme weather events: can developing countries adapt? Climate policy, 3(3), 233-248. Mishra, A. K. (1996). Organizational responses to crisis Trust in Organizations. Frontiers of theory and research (pp. 261-287). California: Sage. National-Emergency-Operation-Center. (2016a). National Emergency Operation Center - Tropical Cyclone Winston - Situation Report 38 of 29/02/2016. National-Emergency-Operation-Center. (2016b). National Emergency Operation Center - Tropical Cyclone Winston - Situation Report 46 of 02/03/2016. National-Emergency-Operation-Center. (2016c). National Emergency Operation Center - Tropical Cyclone Winston - Situation Report 83 of 15/03/2016. NDMP. (1995). Fiji - National Disaster Management Plan. Fiji. Noji, E. K. (2005). Disasters: introduction and state of the art. Epidemiologic reviews, 27(1), 3-8. Noor, K. B. M. (2008). Case study: A strategic research methodology. American journal of applied sciences, 5(11), 1602-1604. Noran, O. (2011). Towards a collaborative network paradigm for emergency services Adaptation and value creating collaborative networks (pp. 477-485): Springer. O'Brien, G., O'Keefe, P., Rose, J., & Wisner, B. (2006). Climate change and disaster management. Disasters, 30(1), 64-80. O'Brien, W. R. (2011). Temporary Network Development Capability In High Velocity Environments: A Dynamic Capability Study Of Disaster Relief Organizations. Osborne, M. J. (2004). An introduction to game theory (Vol. 3): Oxford University Press New York. Oxfam-New-Zealand. (2016). Cyclone Ula strikes Tonga: Oxfam poised to respond. Retrieved 11/2016, 2016, from https://www.oxfam.org.nz/news/cyclone-ula-strikes-tonga- oxfam-poised-respond Packendorff, J. (1995). Inquiring into the temporary organization: new directions for project management research. Scandinavian journal of management, 11(4), 319-333.
312 Papadopoulos, T., Gunasekaran, A., Dubey, R., Altay, N., Childe, S. J., & Fosso-Wamba, S. (2016). The role of Big Data in explaining disaster resilience in supply chains for sustainability. Journal of Cleaner Production. Parsons, T. (1963). On the concept of influence. Public opinion quarterly, 27(1), 37-62. Paton, D., & Flin, R. (1999). Disaster stress: an emergency management perspective. Disaster Prevention and Management: An International Journal, 8(4), 261-267. Patterson, O., Weil, F., & Patel, K. (2010). The role of community in disaster response: conceptual models. Population Research and Policy Review, 29(2), 127-141. Pelling, M. (2003). The vulnerability of cities: natural disasters and social resilience: Earthscan. Pelling, M., Özerdem, A., & Barakat, S. (2002). The macro-economic impact of disasters. Progress in Development Studies, 2(4), 283-305. Pelling, M., & Uitto, J. I. (2001). Small island developing states: natural disaster vulnerability and global change. Global Environmental Change Part B: Environmental Hazards, 3(2), 49- 62. Pérez-Nordtvedt, L., O’Brien, R., & Rasheed, A. A. (2013). What Are Temporary Networks and When Are They Useful? Group & Organization Management, 38(3), 392-421. Perry, M. (2007). Natural disaster management planning: A study of logistics managers responding to the tsunami. International Journal of Physical Distribution & Logistics Management, 37(5), 409-433. Perry, R. W. (1995). The structure and function of community emergency operations centres. Disaster Prevention and Management: An International Journal, 4(5), 37-41. Perry, R. W. (2003). Emergency operations centres in an era of terrorism: Policy and management functions. Journal of Contingencies and Crisis Management, 11(4), 151- 159. Perry, R. W. (2004). Disaster exercise outcomes for professional emergency personnel and citizen volunteers. Journal of Contingencies and Crisis Management, 12(2), 64-75. Pettit, S. J., & Beresford, A. K. (2005). Emergency relief logistics: an evaluation of military, non- military and composite response models. International Journal of Logistics: Research and Applications, 8(4), 313-331. PIR. (2006). Police Commissioner Hughes, others removed. http://www.pireport.org/articles/2006/12/07/police-commissioner-hughes-others- removed-0 Pirson, M., & Malhotra, D. (2011). Foundations of organizational trust: what matters to different stakeholders? Organization Science, 22(4), 1087-1104.
313 Ponterotto, J. G. (2005). Qualitative research in counseling psychology: A primer on research paradigms and philosophy of science. Journal of counseling psychology, 52(2), 126. Popa, C. L. (2005). Initial trust formation in temporary small task groups: Testing a model of swift trust. Kent State University. Porta, R. L., Lopez-De-Silane, F., Shleifer, A., & Vishny, R. W. (1996). Trust in large organizations: National Bureau of Economic Research. Punch, K. F. (2013). Introduction to social research: Quantitative and qualitative approaches: sage. Quarantelli, E. L. (1988). Disaster crisis management: A summary of research findings. Journal of management studies, 25(4), 373-385. Quarantelli, E. L. (1997). Ten criteria for evaluating the management of community disasters. Disasters, 21(1), 39-56. Quarantelli, E. L. (1998). Major criteria for judging disaster planning and managing their applicability in developing countries. Quarantelli, E. L. (1999). The disaster recovery process: What we know and do not know from research. Quarantelli, E. L. (1999). Disaster related social behavior: Summary of 50 years of research findings. Quarantelli, E. L. (2000). Disaster planning, emergency management and civil protection: The historical development of organized efforts to plan for and to respond to disasters. Quarantelli, E. L. (2003). A half century of social science disaster research: Selected major findings and their applicability. Quarantelli, E. L. (2005). Catastrophes are different from disasters: some implications for crisis planning and managing drawn from Katrina. Understanding Katrina: Perspectives from the social sciences. Quinn Patton, M. (2002). Qualitative Research and Evaluation Methods: California: Sage Publications Inc. Raddatz, C. (2007). Are external shocks responsible for the instability of output in low-income countries? Journal of Development Economics, 84(1), 155-187. Raju, E. (2013). Exploring Disaster Recovery: Stakeholder Interfaces, Goals and Interdependencies. Lund University. Retrieved from https://lup.lub.lu.se/search/publication/4173419 Rasinski, K. A., Willis, G. B., Baldwin, A. K., Yeh, W., & Lee, L. (1999). Methods of data collection, perceptions of risks and losses, and motivation to give truthful answers to sensitive survey questions. Applied cognitive psychology, 13(5), 465-484.
314 Rasmussen, T. N. (2004). Macroeconomic implications of natural disasters in the Caribbean: International Monetary Fund. Reliefweb. (2016). Tropical Cyclone Winston - Feb 2016. Retrieved 10/2016, 2016, from http://reliefweb.int/disaster/tc-2016-000014-fji Reuveny, R. (2007). Climate change-induced migration and violent conflict. Political geography, 26(6), 656-673. Revi, A., Satterthwaite, D. E., Aragón-Durand, F., Corfee-Morlot, J., Kiunsi, R. B. R., Pelling, M., . . . Solecki, W. (2014). Urban areas. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, K. L. Ebi, Y. O. Estrada, R. C. Genova, B. Girma, E. S. Kissel, A. N. Levy, S. MacCracken, P. R. Mastrandrea & L. L. White (Eds.), (pp. 535-612). Cambridge, United Kingdom, New York, USA. Riessman, C. K. (2005). Narrative analysis. Narrative, memory & everyday life, 1-7. Riff, D., Lacy, S., & Fico, F. (2014). Analyzing media messages: Using quantitative content analysis in research: Routledge. Ritchie, B. (2008). Tourism disaster planning and management: From response and recovery to reduction and readiness. Current Issues in Tourism, 11(4), 315-348. Roberts, P. S., Ongaro, E., Massey, A., Holzer, M., & Wayenberg, E. (2011). 8. Dispersed federalism: Regional governance for disaster policy. Policy, Performance and Management in Governance and Intergovernmental Relations: Transatlantic Perspectives, 114. Ronan, K. R., Crellin, K., Johnston, D. M., Finnis, K., Paton, D., & Becker, J. (2008). Promoting child and family resilience to disasters: Effects, interventions, and prevention effectiveness. Children Youth and Environments, 18(1), 332-353. Roosli, R., O’Keefe, P., & Mydin, M. A. O. (2013). Evolution of Disaster Planning and Housing in Malaysia: A Reviews. World Applied Sciences Journal, 21(7), 945-959. Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross- discipline view of trust. Academy of management review, 23(3), 393-404. Ryan, M. (2013). Planning in the emergency operations center. Technological forecasting and social change, 80(9), 1725-1731. Saldaña, J. (2013). The coding manual for qualitative researchers: Sage. Sanderson, D., & Jones, P. (2017). Urban development in the Pacific. In P. Thomas & M. Keen (Eds.), Development Bulletin (Vol. 78): Australian National University.
315 Saunders, M. N., Dietz, G., & Thornhill, A. (2014). Trust and distrust: Polar opposites, or independent but co-existing? Human Relations, 67(6), 639-665. Schaubroeck, J., Lam, S. S., & Peng, A. C. (2011). Cognition-based and affect-based trust as mediators of leader behavior influences on team performance. Journal of Applied Psychology, 96(4), 863. Scheper, E., Parakrama, A., & Patel, S. (2006). Impact of the tsunami response on local and national capacities: Tsunami Evaluation Coalition. Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of organizational trust: Past, present, and future. Academy of management review, 32(2), 344-354. Seidman, I. (2013). Interviewing as qualitative research: A guide for researchers in education and the social sciences: Teachers college press. Shafritz, J., Ott, J., & Jang, Y. (2015). Classics of organization theory: Cengage Learning. Sharma, A., Surjan, A., & Shaw, R. (2011). Overview of urban development and associated risks. In R. Shaw (Ed.), Climate and Disaster Resilience in Cities (Vol. 6, pp. 1-16). Shelter-Cluster, F. (2016). Tropical Cyclone Winston. http://reliefweb.int/sites/reliefweb.int/files/resources/20160911_scf_tcwinston_shelt er_response_overview_v7.pdf Shirley, W. L., Boruff, B. J., & Cutter, S. L. (2012). Social vulnerability to environmental hazards Hazards Vulnerability and Environmental Justice (pp. 143-160): Routledge. Simo, G., & Bies, A. L. (2007). The role of nonprofits in disaster response: An expanded model of cross-sector collaboration. Public Administration Review, 67(s1), 125-142. Simons, H. (2009). Case study research in practice: SAGE publications. Smith, W., & Dowell, J. (2000). A case study of co-ordinative decision-making in disaster management. Ergonomics, 43(8), 1153-1166. Spradley, J. P. (2016). Participant observation: Waveland Press. Ssu-Jui, Y., & Huang, F. (2016). Body of last victim of apartment collapse in Taiwan found. Retrieved 02/2017, from http://focustaiwan.tw/news/asoc/201602180031.aspx Stake, R. E. (1995). The art of case study research: Sage. Stenchion, P. (1997). Development and disaster management. Australian Journal of Emergency Management. Stephens, H. W., & Grant, G. O. (2001). A New use for an Old Model: Continuity of Government as a Framework for Local Emergency Managers. In A. Farazmand (Ed.), Handbook of Crisis and Emergency Management (pp. 283-292): Marcel Dekker, Inc. Stephenson Jr, M. (2005). Making humanitarian relief networks more effective: operational coordination, trust and sense making. Disasters, 29(4), 337-350.
316 Stone, A. A., Bachrach, C. A., Jobe, J. B., Kurtzman, H. S., & Cain, V. S. (1999). The science of self- report: Implications for research and practice: Psychology Press. Straffin, P. D. (1993). Game theory and strategy (Vol. 36). Washington DC: The Mathematical Association of America. Strömberg, D. (2007). Natural disasters, economic development, and humanitarian aid. The Journal of Economic Perspectives, 199-222. Sundermann, L., Schelske, O., & Hausmann, P. (2013). Mind the risk - A global ranking of cities under threat from natural disasters. In U. Leimbacher (Ed.), (pp. 33). Zurich: Swiss Re. Swithern, S. (2014). Global Humanitarian Assistance Report 2014. In GHA (Ed.), (pp. 150): GHA. Tatham, P., & Kovács, G. (2010). The application of “swift trust” to humanitarian logistics. International Journal of Production Economics, 126(1), 35-45. Tatham, P., & Kovács, G. (2012). Developing and maintaining trust in hastily formed relief networks. In G. Kovács & K. M. Spens (Eds.), Relief supply chain management for disasters: humanitarian aid and emergency logistics: IGI Global. Taylor, S. J., Bogdan, R., & DeVault, M. (2015). Introduction to qualitative research methods: A guidebook and resource: John Wiley & Sons. Telford, J., & Cosgrave, J. (2007). The international humanitarian system and the 2004 Indian Ocean earthquake and tsunamis. Disasters, 31(1), 1-28. The-World-Factbook. (2016). Fiji. Retrieved 10/2016, from CIA https://www.cia.gov/library/publications/the-world-factbook/geos/fj.html Thévenaz, C., & Resodihardjo, S. L. (2010). All the best laid plans… conditions impeding proper emergency response. International Journal of Production Economics, 126(1), 7-21. Törnqvist, E., Sigholm, J., & Nadjm-Tehrani, S. (2009). Hastily formed networks for disaster response: Technical heterogeneity and virtual pockets of local order. Paper presented at the Proceedings of the 6th International ISCRAM Conference, Gothenburg, Sweden. Tran, M. (2006). Fiji's history of coups. https://www.theguardian.com/world/2006/dec/05/fiji.travel1 Tremblay, M. A. (1957). The key informant technique: A nonethnographic application. American Anthropologist, 59(4), 688-701. Tseng, C.-P., & Chen, C.-W. (2012). Natural disaster management mechanisms for probabilistic earthquake loss. Natural hazards, 60(3), 1055-1063. Turner III, D. W. (2010). Qualitative interview design: A practical guide for novice investigators. The qualitative report, 15(3), 754.
317 Turoff, M., Chumer, M., Van de Walle, B., & Yao, X. (2004). The design of a dynamic emergency response management information system (DERMIS). JITTA: Journal of Information Technology Theory and Application, 5(4), 1. Twigg, J. (2004). Disaster risk reduction: mitigation and preparedness in development and emergency programming: Overseas Development Institute (ODI). Uitto, J. I. (1998). The geography of disaster vulnerability in megacities: a theoretical framework. Applied geography, 18(1), 7-16. UN-HABITAT. (2006). The state of the world's cities report 2006/2007; 30 years of shaping the habitat agenda. In U. N. H. S. Programme (Ed.). UK, USA: UN-HABITAT. UN. (2005). Post-Disaster Recovery: Guidelines and Good Practices (E. Chamber, Trans.) (pp. 9): United Nations. UN. (2012). World Urbanization Prospects - the 2011 revision (highlights) (D. o. E. a. S. Affairs & P. Division, Trans.) (pp. 50). New York: United Nations. UN. (2014). Disaster Resilience Scorecard for Cities (pp. 56). United Nations. UN. (2015). GDP and its breakdown at current prices in US Dollars. Retrieved 03/2015, from Natinoal Accounts Main Aggregates Database - UN http://unstats.un.org/unsd/snaama/dnlList.asp UNDP. (2004). Reducing disaster risk: a challange for development (U. N. D. Programme, Trans.) (pp. 161). New York: United Nations. UNEP/UNISDR. (2013). Multiple Risk - Mortality Risk Index: UNEP/UNISDR. UNISDR. (2013). Global Assessment Report on Disaster Risk Reduction 2013. United Nations: United Nations. UNISDR. (2017). Terminology - Disaster. Retrieved 10/10/18, from https://www.unisdr.org/we/inform/terminology#letter-d United Nations, D. o. E. a. S. A., Population Division. (2014). World Urbanization Prospects: The 2014 Revision. Retrieved 03/2015, from United Nations http://esa.un.org/unpd/wup/DataQuery/ UNOCHA. (2016). Tropical Cyclone Winston - Response and Flash Appeal - Final Summary. http://reliefweb.int/sites/reliefweb.int/files/resources/FINAL%20- %20TC%20WINSTON%20Flash%20Report_130616.pdf Van Huyck, J. B., Battalio, R. C., & Walters, M. F. (1995). Commitment versus discretion in the peasant-dictator game. Games and economic behavior, 10(1), 143-170. Van Teijlingen, E. R., & Hundley, V. (2001). The importance of pilot studies.
318 Veszteg, R. F., Funaki, Y., & Tanaka, A. (2014). The impact of the Tohoku earthquake and tsunami on social capital in Japan: Trust before and after the disaster. International Political Science Review. Walker, P. (1992). Foreign military resources for disaster relief: An NGO perspective. Disasters, 16(2), 152-159. Waugh, W. L. (1999). Living with hazards, dealing with disasters: An introduction to emergency management: ME Sharpe. Waugh, W. L., & Streib, G. (2006). Collaboration and leadership for effective emergency management. Public Administration Review, 66(s1), 131-140. Webb, G. (2004). Role improvising during crisis situations. International Journal of Emergency Management, 2(1-2), 47-61. Webb, G. R., & Chevreau, F.-R. (2006). Planning to improvise: the importance of creativity and flexibility in crisis response. International Journal of Emergency Management, 3(1), 66- 72. Weick, K. E. (1993). The collapse of sensemaking in organizations: The Mann Gulch disaster. Administrative Science Quarterly, 628-652. Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of personality and social psychology, 39(5), 806. Weiss, R. S. (1995). Learning from strangers: The art and method of qualitative interview studies: Simon and Schuster. Wenger, D. E., Dykes, J. D., Sebok, T. D., & Neff, J. L. (1975). It's a matter of myths: An empirical examination of individual insight into disaster response. Mass Emergencies, 1(1), 33-45. Weston, C., Gandell, T., Beauchamp, J., McAlpine, L., Wiseman, C., & Beauchamp, C. (2001). Analyzing interview data: The development and evolution of a coding system. Qualitative sociology, 24(3), 381-400. White, M. P., & Eiser, J. R. (2006). Marginal trust in risk managers: Building and losing trust following decisions under uncertainty. Risk Analysis, 26(5), 1187-1203. Whiting, L. S. (2008). Semi-structured interviews: guidance for novice researchers. Nursing Standard, 22(23), 35-40. Wiethoff, C. M., & Lewicki, R. J. (2005). Trust and distrust in work relationships: A grounded approach. Paper presented at the IACM 18th Annual Conference. Wildman, J. L., Thayer, A. L., Rosen, M. A., Salas, E., Mathieu, J. E., & Rayne, S. R. (2012). Task types and team-level attributes: Synthesis of team classification literature. Human Resource Development Review, 11(1), 97-129. Williams, M. (2000). Interpretivism and generalisation. Sociology, 34(2), 209-224.
319 Williams, M. (2001). In whom we trust: Group membership as an affective context for trust development. Academy of management review, 26(3), 377-396. Wishart, J. P. (2008). Fostering partnership in humanitarian aid and disaster relief: DTIC Document. WorldBank, T. (2015a). Country and Lending Groups. Retrieved 03/2015, 2015, from http://data.worldbank.org/about/country-and-lending-groups WorldBank, T. (2015b). How does the World Bank classify countries? Retrieved 17/03/2015, 2015, from https://datahelpdesk.worldbank.org/knowledgebase/articles/378834-how- does-the-world-bank-classify-countries WorldBank, T. (2015c). Urban Population (% of total). Retrieved 03/2015, from The World Bank http://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS/countries/1W?display=map WorldBank, T. (2016). Fiji. Retrieved 10/2016, from World Bank http://data.worldbank.org/country/fiji Yin, R. K. (2009). Case study research: Design and methods (Vol. 5): Sage. Yin, R. K. (2017). Case study research and applications: Design and methods: Sage publications. Yoder-Bontrager, D. (2014). Nongovernmental organizations in disaster and coordination: A complex adaptive systems view. University of Delaware. Retrieved from http://media.proquest.com/media/pq/classic/doc/3633715991/fmt/ai/rep/NPDF?_s= wJWSR7OUwCzS6S1oDsJEKrU2HBU%3D Zainal, Z. (2007). Case study as a research method. Jurnal Kemanusiaan, 9. Zolin, R. (2002). Swift trust in hastily formed networks: DTIC Document. Zolin, R., & Gibbons, D. E. (2014). How Emergent Roles and Structures Create Trust in Hastily Formed Interorganizational Teams. Sage Open, 4(2). Zucker, L. G. (1983). Organizations as institutions. Research in the Sociology of Organizations, 2(1), 1-47.
320 APPENDICES………………………..
Appendix A – The EOC organisation chart; staff movement board; and vision, mission and values
Appendix B – An explanation of Fijian culture
Appendix C – Final interview questionnaire
321 Appendix A – The EOC organisation chart; staff movement board; and vision, mission and values
322
323
324 Appendix B – An explanation of Fijian culture
Fiji has a strong cultural history that is ingrained within the social fabric of the country. A part of this cultural history is the unique relationships that Indigenous Fijians (iTaukei) have with each other. When meeting someone for the first time, Fijians will introduce themselves and include within this where they are from and their lineage (EOC 16). There are six types of relationships that are based on two primary conditions:
1. Lineage – as stated, Fijians form a part of their identity through their lineage. This is generally on the paternal side, and goes back as far as their oldest generations.
2. Province and confederacy – Fiji is broken up into 14 provinces, which culminate into 3 confederacies (Burebasaga, Kubuna, and Tovata). The province that you are ‘from’ is based upon your lineage. As such, if you were born in a province in the Burebasaga confederacy, but your paternal lineage traces back to a different province in the Tovata confederacy, then that is your province and confederacy for the sake of forming cultural relationships.
The six types of cultural relationships that iTaukei have with each other are characterised by expected levels of respect and behaviour. Also importantly, there is no case where an iTaukei does not have a cultural relationship with another Indigenous Fijian. As such, everyone is related by one of the following relationships:
Kai, Mataqali (Mata), Tovata – these three relationships are grouped because they have similar respect and behaviour expectations. These relationships are similar to that of a brother, sister, aunt, uncle, etc. If someone is your Kai, Mata, or Tovata, they are to be respected, and your behaviour is to match this level of respect.
Naita – this relationship is more ‘relaxed’ than someone being your Kai, Mata, or Tovata. If someone is your Naita, then they are to be respected and cared for, but the relationship is also characterised by a ‘fun’ dynamic between the two people.
Tauvu (Tau) – this appears to be the most commonly used relationship; having been mentioned by multiple people within the EOC. If someone is your Tau, then they are someone that you can tease and make fun of in a demeaning manner. EOC 11 provided an example – if explaining something to a colleague who was their Tau, and their Tau did not understand, then they would say back, “see, that’s the problem of not doing studies when you are supposed to.” (EOC 11)
325 The Tau relationship also extends to people being able to take property from one another without question. Although this was historically around livestock or property, now it is limited to small items; particularly when in the workplace. Taus are also able to tell each other what to do and compliance is expected. Again, this relationship is not stretched to where requests or items taken are unreasonable. Examples of having a Tauvu relationship in the EOC could be a colleague taking someone’s lunch, or their pen, or asking them to book a car for them; and there is a cultural understanding that this is acceptable (unlike in a Western culture where this may be perceived as being rude). It is important to remember also that in a Tau relationship, whatever you do to your Tau can be reciprocated – which helps control the extent to which this relationship is utilised.
Dreu – this is a unique type of relationship that exists only between a small number of provinces (when compared against the other cultural relationships). The expected behaviour sits between that of someone being your Tauvu and Naita.
These relationships place the local respondents in a unique position when considering working in the EOC. The locals have to determine how much they let culture influence their behaviour and language in the workplace. Some respondents valued professionalism and they didn’t take advantage of these culture relationships Conversely, some took culture very seriously and used their Tau relationships to achieve tasks. As an example, the iTaukei had to determine the balance between someone being their superior in the workplace and also their Tau; and therefore how to relate to this person.
Considering these cultural relationships and how they influenced trust forming within the EOC, it was clear that the local respondents were able to relate to each other on a cultural level that their international colleagues could not. And even though there was no ‘hard and fast’ rule for how to integrate these relationships within the EOC, the locals could choose how much they wanted to use these cultural relationships to help form trust. For a local respondent, this does not mean that they trusted a local more than an international colleague; however no local had led me to believe that locals were trusted less than international colleagues. And this provided the key insight into the role of culture in influencing trust within the EOC. These cultural relationships provided a local respondent with additional evidence when deciding whether to trust another local colleague; whereas an international colleague did not have this as a possibility. Furthermore, regardless of whether the relationship was one of respect or joking banter, the relationships were not valued differently in their significance. That is, all cultural
326 relationships provided a link through which a local could identify themselves with respect to another iTaukei. Going further, EOC 16 stated that these cultural relationships, “all have the elements of trust within.” Therefore, locals simply had more information from which they could draw upon when establishing trust; and it was up to them whether they wanted to utilise this or not when operating in the EOC.
327 Appendix C – Final interview questionnaire
1. Can you explain your role here at the EOC?
2. How does this office work? Is it all about preparing for the daily briefings? Are there other major deliverables that the office has to deliver as a team or is it all individual work from there on?
3. And what area do you work under?
4. Do you know what everyone in the office does?
5. When with your colleagues, what did you interact about?
6. If you have a problem that you needed help with, who would you go to?
a. If they answer their manager - What if the question was not to do with your manager’s area of work, would you still go to them?
b. If they answer someone else – Ok, why not go through your manager?
7. Only ask the Director- Many people said that they reported to you during the operation and would come to you if they needed help. Did you find this to be the case? Did you think that you were having to micro-manage every decision? Was this your choice? Why was this the case?
8. How many of your interactions were related to your work? How did you interact with them, phone, email, person to person?
9. Would you consider any of your colleagues to also be a personal or close friend?
10. Do you consider trust between your colleagues important when working in the EOC?
11. What advantages do you think trust provides when working with your colleagues on the response and recovery?
12. How well would you have to know a colleague you had never met before trusting them when working in the EOC?
13. Only ask the local respondents - Do you have a cultural relationship (Mata, Kai, Tovata, Naita, Dreu or Tauvu) with anyone else in the office? Is there anyone who you are not related to in one of these ways?
328 a. Of the people you had not met nor worked with before during the response, does your cultural link to this person help you to trust them? Can you explain?
b. Do these cultural links apply to your international colleagues also? If not, are you quicker to trust a local colleague instead of an international colleague because of these cultural links?
14. Do you think the involvement of the military in the response affects trust within the EOC?
15. Do you expect to work with any of your current colleagues again in the future? If so, who and under what circumstances?
16. Given the chance of future interaction (or if this is limited), do you believe that people behave in a trusting manner because their reputation is at stake?
17. Do you consider how others perceive you and your reputation when deciding how to behave?
18. Of the local colleagues you had not met nor worked with before, how do you judge whether to trust them?
19. Would you use the same criteria in judging whether to trust another local person outside of the EOC?
20. Of the international colleagues you had not met nor worked with before, how do you judge whether to trust them?
21. Would you use the same criteria in judging whether to trust another international person outside of the EOC?
22. Of the local colleagues you had not met nor worked with before, do you know their professional qualifications? Do you feel they are qualified to do their job?
23. Of the international colleagues you had not met nor worked with before, do you know their professional qualifications? Do you feel they are qualified to do their job?
24. If you knew a colleague was here to work in a specific area, but was doing something that you thought was outside of their expertise as part of the response, would that affect how you work with them in the future?
329 25. Do people generally behave the way that you expect them to or does the EOC create an environment where you have to adapt to potentially unexpected behaviour of others?
26. If they answer that they have had to adapt – does having to adapt to the unexpected behaviour of others affect how you will approach working in any recovery offices in the future. If so, in what ways?
27. How much do you rely on your colleagues to do your job (if they need a prompt, refer them to their Network Mapping Worksheet)? Can you elaborate on this?
28. In general, if someone fails to do a task properly, does that just affect them or does it impact the wider team's performance?
29. Considering (their contractor), does their reputation help you trust the other colleagues whom you have not met or worked with before that also report to them? (If true) What about your colleagues in the wider EOC?
30. Considering (the person who they put as their manager – assuming it is not the same as the contractor) does their reputation help you trust the other colleagues whom you have not met or worked with before in the EOC that also report to them? (If true) What about your colleagues in the wider EOC?
31. Considering what can go wrong with working in the EOC, would you say you are; less likely, as likely, or more likely than your colleagues to be detrimentally affected?
32. When problems arise that affect the operation of the EOC, are solutions allowed to be improvised, or is there a strict procedure to follow?
33. Were you aware of the signage around the office talking about the EOC values – with one of those values being trust? Does this effect how you behave in the EOC?
34. Only ask the local respondents – what three things can international colleagues do to gain your trust?
35. Only ask the international respondents – what three things do you think your local colleagues think you should do to gain their trust?
330