United Nations Development Programme Drylands Development Centre United Nations Avenue, Gigiri P.O. Box 30552 Nairobi, 00100, Kenya www.undp.org/drylandscentre Email: [email protected]
Understanding Community Resilience: Findings from Community-Based Resilience Analysis (CoBRA) Assessments
Marsabit, Turkana and Kajiado counties, Kenya and Karamoja sub-region, Uganda Understanding Community Resilience: Findings from Community-Based Resilience Analysis (CoBRA) Assessments Analysis (CoBRA) from Community-Based Resilience Findings Understanding Community Resilience: Disclamer The views expressed in this publication are those of the authors and do not necessarily represent those of the United Nations, including UNDP, or its Member States.
Copyright © 2014 United Nations Development Programme Drylands Development Centre United Nations Avenue, Gigiri P.O. Box 30552, Nairobi, 00100, Kenya All rights reserved
April 2014
Layout Catherine Kimeu/UNON
Editor Catharine M. Way
Printing UNON, Publishing Services Section, Nairobi, ISO 14001:2004-certified D1 No.: 13 - 54467/700 Copies
This publication was funded by the European Commission Directorate General for Humanitarian Aid and Civil Protection. The views expressed herein can in no way be taken to reflect the official opinion of the donor.
Humanitarian Aid and Civil Protection Humanitarian Aid and Civil Protection
Understanding Community Resilience: Findings from Community-Based Resilience Analysis (CoBRA) Assessments
Marsabit, Turkana and Kajiado counties, Kenya and Karamoja sub-region, Uganda
Commissioned by UNDP Drylands Development Centre
Under the framework of the European Commission Directorate General for Humanitarian Aid and Civil Protection’s Drought Risk Reduction Action Plan United Nations Development Programme (UNDP) UNDP partners with people at all levels of society to help build nations that can withstand crisis, and drive and sustain the kind of growth that improves the quality of life for everyone. On the ground in 177 countries and territories, we offer global perspective and local insight to help empower lives and build resilient nations.
The UNDP Drylands Development Centre is a unique global thematic centre that provides technical expertise, practical policy advice and programme support for poverty reduction and development in the drylands of the world. The Centre’s work bridges between global policy issues and on-the-ground activities, and helps governments to establish and institutionalize the link between grassroots development activities and pro- poor policy reform. The main areas of focus are mainstreaming of drylands issues into national development frameworks; land governance; marking markets work for the poor; decentralized governance of natural resources; and drought risk management.
www.undp.org/drylandscentre
ii Understanding Community Resilience Table of contents
Acknowledgements...... iv Executive summary ...... v 1. Introduction...... 1 1.1 The CoBRA model...... 2 1.2 The CoBRA methodology...... 3 2. Field testing...... 4 2.1 Identification of field sites...... 4 2.2 Approach...... 4 2.3 Constraints and limitations of data collection ...... 7 3. Findings...... 9 3.1 Main hazards or shocks...... 9 3.2 Characteristics of a resilient community...... 9 3.3 Extent to which communities achieved resilience characteristics...... 13 3.4 What a resilient household looks like...... 15 3.5 Interventions that contributed to household resilience...... 16 3.6 How key informants achieved resilience...... 18 4. Summary of feedback from stakeholder consultations...... 23 5. Lessons learnt from CoBRA assessments in Kenya and Uganda...... 25 5.1 Benefits of the methodology...... 25 5.2 Limitations of the methodology ...... 27 5.3 Indicators of resilience...... 28 6. Conclusions and recommendations...... 31 6.1 Conclusions...... 31 6.2 Recommendations...... 33 6.3 Next steps...... 34 6.4 Suggested improvements to the CoBRA methodology...... 35 Annex 1: Community-Based Resilience Analysis Assessment Report Marsabit Central, Laisamis and Maikona districts...... 37 Annex 2: Community-Based Resilience Analysis Assessment Report Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts...... 65 Annex 3: Community-Based Resilience Analysis Assessment Report Kotido and Kaabong districts...... 89 Annex 4: Community-Based Resilience Analysis Assessment Report Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts...... 121
Table of contents iii Acknowledgements
This document is a synthesis of the results and the findings of four pilot community-based resilience analysis (CoBRA) assessments implemented in Kenya and Uganda.
We are indebted to the consultants who facilitated the CoBRA assessments and led the preparation of this publication: Courtenay Cabot Venton, Catherine Fitzgibbon, Laban MacOpiyo, Moses Ojota and Laz Ocira.
Special gratitude goes to all the government and non-governmental partners who proactively participated in the CoBRA exercises in each location and provided the context- specific inputs into the respective Annexes. Special mention goes to James Oduor, Paul Kimeu, Francis Koma, Julius Taigong’ and Benedict Musyoka at National Drought Management Authority in Kenya, and Solomon Elungat and Martin Odong at the Office of the Prime Minister in Uganda.
Our special thanks also go to the UNDP Country Offices in Kenya and Uganda who provided technical and administrative support throughout the entire CoBRA assessment processes. The authors are indebted to many people who, at different points in the production of this publication offered comments and valuable technical inputs to the report. Our special thanks to Oliver Wasonga, Jacob Park and Mark A. Constas, as well as other technical reviewers who provided comments at various stages of the CoBRA methodology developments.
Also acknowledged with thanks are Yuko Kurauchi and Francis Opiyo at UNDP Drylands Development Centre (DDC) for their insights and guidance in this publication process.
The UNDP DDC gratefully acknowledges the financial support provided by the European Commission Directorate General for Humanitarian Aid and Civil Protection to produce this document.
iv Understanding Community Resilience Executive summary
In the last few years, as natural disasters and other crises have pushed communities to the limits of their adaptation and coping capacity, ‘disaster resilience’ has emerged as a key goal for governments and other development and humanitarian stakeholders in the Horn of Africa. It is in this context that the United Nations Development Programme Drylands Development Centre initiated the Community-Based Resilience Analysis (CoBRA) project, with financial support from the European Commission Directorate General for Humanitarian Aid and Civil Protection (ECHO). The CoBRA methodology is one of the first practical analytical tools developed to identify indicators for measuring community resilience as part of ECHO’s wider Drought Risk Reduction Action Plan.
This report summarizes the findings of the first round of CoBRA field testing in four drought-prone locations in Kenya (Marsabit, Turkana and Kajiado counties) and Uganda (the Karamoja sub-region). It also reports on the outputs of subsequent validation sessions held in each of the four locations with local technical stakeholders and community representatives. Individual assessment reports for each location are attached as Annexes to this report.
CoBRA approach and objectives The CoBRA approach is largely qualitative, based on understanding resilience from a community perspective. It does not identify any preconceived components of resilience but rather allows communities to define it, assess their progress in achieving it, identify households that are more (or fully) resilient and specify the interventions they believe best build resilience.
The CoBRA methodology has four broad objectives:
1. Identify the priority characteristics of disaster resilience for a target community; 2. Assess the communities’ achievement of these characteristics at the time of the assessment and during the last crisis or disaster; 3. Identify the characteristics and strategies of disaster-resilient households; and 4. Identify the most highly rated interventions or services in building local disaster resilience.
A detailed CoBRA conceptual framework document and implementation guidelines have been developed to fully explain the model’s logic and methodology.
Methodology The CoBRA methodology uses participatory qualitative approaches – focus group discussions (FGDs) and key informant interviews (KIIs) (see Boxes 1 and 2). In each field site, 36 to 42 FGDs and KIIs were carried out by teams of facilitators. Following each assessment, feedback sessions were held with community representatives and local stakeholders to validate the findings.
Key findings Four full CoBRA assessments have been completed to date, three in Kenya, in partnership with the National Drought Management Authority, and one in Uganda, in partnership with the Office of the Prime Minister. The assessments have highlighted the following findings:
Executive summary v Box 1. Steps in conducting a focus group discussion
Step 1. Agree on the definition of resilience: What does a resilient community look like? What are the main hazards or shocks facing the community? Step 2. Identify resilience characteristics: What are the characteristics of a resilient community? Step 3. Prioritize resilience characteristics: What are the three most important characteristics of resilience in the community, ranked by importance? Step 4. Rate the community’s progress in attaining the priority resilience characteristics: On a scale of 0 to 10, to what extent has this community achieved each of these characteristics in the current period and in the last crisis period? Step 5. Identify interventions that have contributed to household resilience: What interventions have helped to enhance households’ level of resilience, and what additional/future interventions would help to build resilience further? Step 6. Identify the households in the community that have achieved (fully or partially) the resilience characteristics and list their common features and attributes.
Box 2. Questions posed in key informant interviews with resilient households
• What factors or characteristics have contributed to your household’s resilience? • How did your household become resilient? • Why do you think your family coped better with shocks and crises affecting the community? • What interventions do you think would best build wider resilience in this community?
• Throughout the locations, communities consistently highlighted several priority characteristics (described in approximately 20–30 statements) that identify a resilient community. Characteristics that were highly prioritized in all the assessment locations include: - Education: All children would be able to complete primary/secondary/tertiary school; - Water: The whole community would have access to sufficient, good-quality water at all times of the year; - Peace and security: The whole community would enjoy continual peace and security. Human health, productive livestock herds and farms, access to markets and credit, diversified incomes and roads were also highly rated in many of the locations. These characteristics came up consistently across different livelihood groups (pastoral, agropastoral and peri-urban) and age and gender groups, irrespective of the number and level of services and interventions provided in the assessment sites. However, there were prominent differences between locations in the priority assigned to these characteristics, depending on unique local ecological and socioeconomic conditions. • Resilient households were consistently described as households with greater income and assets built through diverse sources. The most common feature of resilient households noted through the FGDs and KIIs was their multiple sources of income, which tended to combine traditional on-farm activities (pastoral and crop farming) with off-farm income-generating activities (IGAs) that are less dependent on the weather, such as small businesses, trading or wage/casual labour. The few resilient households with single income sources were principally pastoralists with large herds. Resilient households also tended to have higher levels of education, and this was perceived to account partially for their better access to diverse sources of income. vi Understanding Community Resilience • Communities perceived they had attained extremely low levels of the identified resilience characteristics, even in times of no crisis or hazard. This perception underlines the very low levels of development and high chronic vulnerability experienced in these locations at all times, which are exacerbated by recurrent drought and other hazards. However, even within the same location, some livelihood groups are more vulnerable to drought hazards than others. • The most highly ranked characteristics of resilience often did not correspond with interventions supported by governments and their partners. Comparing the most highly rated local resilience-building characteristics with the portfolio of projects and interventions provided to the same communities often reveals a mismatch. For example, completing secondary and tertiary education was regularly cited as a key characteristic of resilience; however, support for expanding access to these services is rarely prioritized in government/NGO programming in these areas. Equally, communities did not always prioritize characteristics that should be part of any investment. For example, feedback sessions with local stakeholders consistently cited communities’ lack of priority for natural resource management, which stakeholders felt was a significant gap.
Benefits and limitations of the CoBRA approach The CoBRA methodology represents one of the very few practical analytical tools that has been developed and tested to measure resilience at the community level. It brings communities’ perspectives into resilience debates and provides communities with an opportunity to describe what resilience means in reality. The combination of community and household resilience characteristics identified through FGDs and KIIs helps to increase understanding of the term. It can provide a valuable basis upon which to develop a small set of locally specific indicators and to support strategic planning processes. A CoBRA assessment provides a substantial amount of information in a relatively short time frame and at significantly less cost than equivalent quantitative approaches used in the past. This is due to the proactive participation of local government and NGOs, which involves close collaboration and technical and logistical backstopping.
A CoBRA assessment is not a stand-alone measurement of resilience, and it does not offer a definitive quantitative measurement of the number or proportion of resilient households in a given community. The resilience attainment scores are perceptual and not statistically significant. Additionally, scores cannot be compared between locations, as they are derived from separate conversations, in different contexts. CoBRA examines the multidimensional nature of resilience, which is unlikely to be affected by any single intervention. Therefore it should not replace individual project or programme evaluations.
Developing indicators of resilience The conceptual framework of CoBRA is based on the premise that resilience can be measured in two ways:
1. Using an overall or universal measure or indicator(s) of resilience, which enables us to understand whether resilience is increasing, decreasing or staying the same. The almost uniform finding that resilient households have multiple sources of income and a strong asset base indicates there is potential for measuring absolute resilience. The findings from the CoBRA assessments confirm the hypothesis underpinning the conceptual framework – that households would define themselves as resilient when they were able to feed their families adequately every day and meet basic needs on a consistent basis both in stressful and ‘normal’ times without external relief. Being able
Executive summary vii to quantifiably measure this ‘ability to cope’ would then provide a universal indicator of resilience. 2. Using composite and contextually specific indicators of resilience, which enable us to understand how local drivers of resilience are expanding or contracting, and the impact of interventions on those drivers. The four assessments to date have consistently highlighted highly similar household characteristics of resilience. Community characteristics are more varied, but with strong commonalities. Linking these characteristics with quantitative indicators that could be tracked over time is the logical next step for the CoBRA methodology.
Implications for community-based resilience programming
• The priority characteristics of resilience identified in a target community through CoBRA can be used as a decision support tool to inform policy, planning and programme/ project processes at different levels. Importantly, the assessment highlights how these characteristics vary between different locations, groups or contexts, which can be used to help tailor resilience programming. • The characteristics identified and prioritized by the communities can in turn be used to identify keystone indicators of resilience for more systematic tracking. These indicators can be both universal and contextually specific, leading to a potential resilience threshold. The keystone indicators can be monitored over time to track changes in resilience characteristics quantitatively. • The CoBRA assessment provides data on the most highly rated interventions or services in building local resilience, which can inform community-based programming as well as wider policy and practice. Typically, top-rated interventions help to improve access to basic services – such as education, water, health care, roads and markets – as well as their quality and efficiency. Importantly, such interventions encompass both long- term and short-term investments and include factors that are not always immediately considered a part of disaster risk reduction (DRR) or development strategies. For the DRR approach to help affected populations become truly resilient and cope with future shocks and stresses independently, the risk reduction concept and strategy need to be broadened. • The study provides strong evidence that interventions that demonstrably increase income and diversify sources of income should be a core focus of any community- based programming. Interestingly, this is also the finding that had the most consistent support in the validation sessions with stakeholders. Source of income needs to be incorporated into any analysis to ensure that unsustainable activities are not included.
Recommendations The findings from the CoBRA assessments lead to a number of practical policy recommendations to spur innovative and entrepreneurial capacities of drought-affected dryland communities in the Horn of Africa.
1. CoBRA findings should be used to identify a locally validated list of resilience indicators that government authorities and other stakeholders can use to monitor progress and impact. Including communities’ definitions of resilience is important not only as a participatory exercise, but also as a means of understanding local contextual factors that drive or undermine resilience. CoBRA findings should also support the development of more streamlined monitoring frameworks for both national/regional (e.g., regional and country programming papers of the Intergovernmental Authority for Development’s Drought Disaster Resilience and viii Understanding Community Resilience Sustainability Initiative1) and local climate-resilient development planning processes. Stakeholders need to review current data sets to determine if and how indicators identified through the CoBRA process are already being collected. Data collection processes may need to be revised or standardized to ensure that appropriate information is collected regularly on certain topics, such as secondary school completion rates or peace and security. 2. Policy and planning frameworks need to refocus on activities that build and diversify incomes and assets in the drylands. The finding on the link between resilience and higher levels of income and assets derived from multiple sources has important implications for governments and donors seeking to understand how to allocate resources to best build resilience. Increasing access to credit, strengthening savings groups and improving commercial literacy and business skills are all key interventions in this regard. 3. Activities that support the achievement of community-identified characteristics of resilience need to be a key focus. The high ratings universally given to education, water, health and peace and security for community resilience means that these factors will drive change that enables households to develop multiple sources of income and make other positive changes. Education can link people to broader income- generation opportunities, especially off-farm activities such as wage labour. Access to credit can facilitate various IGAs and help diversify livelihoods. Water is key to both human and ecosystem health and economic development, contributing directly to livestock/agricultural production and productivity. These linkages are not necessarily systematically addressed in current policy and planning frameworks. 4. Policy and planning need to increase their emphasis on raising awareness of natural resource management, particularly the critical role natural resources can play in the drylands in the face of shocks and stresses. The CoBRA assessment demonstrates a key gap in communities’ awareness of the importance of natural resources and of maintaining the long-term health of local ecosystems that form the basis for their livelihoods. There is a risk that people will be more prone to engage in activities that may build diverse incomes that are unsustainable (such as charcoal production). 5. CoBRA results stress the need for coordinated and concerted action among actors at different scales – including hardware and software, long term and short term, and small and large investments. Some of the perceived priority characteristics – such as peace and security, secondary/tertiary education and roads – are not systematically integrated into development and humanitarian support. Ignoring these costly and long-term interventions and instead focusing on less costly investment may lead to false economies. Communities consistently highlighted interventions that enhance access to markets, savings and credit as highly beneficial for enhancing community/ household resilience, and these should be prioritized in the short term. Their success, however, may be handicapped if larger scale interventions continually fail to be developed in tandem. Greater support for a coordinated approach will be required to ensure resilience in the long term.
1 This initiative was developed based on the discussions and recommendations at the Joint Horn of Africa and East Africa Summit of Heads of State and Government held in Nairobi, Kenya on 9th September 2011, to enhance partnership and coordination and strengthen regional and national frameworks to end drought emergencies in the HoA.
Executive summary ix x Understanding Community Resilience 1. Introduction
CoBRA consultation meeting with Men in Laisamis, Marsabit County, Kenya
The Horn of Africa has long faced droughts partners carrying out full field testing of and other natural disasters. But in recent the approach in the Horn of Africa.) years, as the impacts of extreme events have begun to worsen, partially as a After providing a brief overview of the result of climate change, already stressed CoBRA model (this section) and field communities have been pushed to the limit work methodology (section 2), the report of their adaptation and coping capacity. outlines and analyses the key findings of Therefore, supporting communities to four full CoBRA assessments, which were become resilient in the face of disasters has undertaken between June and August emerged as a key goal for governments 2013 in three counties in Kenya (Marsabit, and development and humanitarian Turkana and Kajiado) and two districts of stakeholders in the region. As a consequence, Karamoja in Uganda (section 3). The report programmes and funding strategies are also provides some of the main feedback increasingly realigning themselves around generated during review and validation resilience-building objectives. In parallel, workshops attended by community several efforts are being made to develop representatives and technical stakeholders models that better define and measure from the local government and non- resilience. It is in this context that the UNDP governmental groups (section 4), which Drylands Development Centre developed took place at each of the four field sites the Community-Based Resilience Analysis between September and November 2013. (CoBRA) project, with financial support A national workshop held in Kampala from the European Commission Directorate in November 2013 also provided an General for Humanitarian Aid and Civil opportunity for in-depth validation of the Protection (ECHO). assessment findings from Uganda.
The CoBRA assessment methodology is The report points out that the findings of one of the first practical analytical tools the CoBRA assessments could inform wider developed to identify indicators for debate and learning on monitoring and measuring community resilience as part measurement of resilience at community, of ECHO’s wider Drought Risk Reduction national, sub-regional and regional levels. Action Plan. (A comprehensive explanation It also examines opportunities for CoBRA of the conceptual framework behind the to link to or inform other monitoring methodology is contained in the CoBRA and evaluation processes to improve Conceptual Framework and Methodology,2 effectiveness in measuring changes in and the CoBRA Implementation resilience (section 5). Finally, the report Guidelines3 provide guidance for local makes recommendations to spur use of
2 Available for download on the Drought Online at: http://www.disasterriskreduction.net/drought-online/cobra/en/ 3. Ibid.
Introduction 1 the model and on modifications that may • Which households are more resilient be required to improve the methodology and able to cope with shocks and (section 6). stresses? • What kinds of factors are affecting Detailed findings from the four individual their ability to cope? assessments are presented in Annexes 1–4. • How do communities score their attainment of these priority 1.1 The CoBRA model characteristics in a normal period and in a crisis period? The CoBRA model, presented in Figure 1, was developed through a literature review, The scoring exercise provides important stakeholder consultation and field testing data on community perceptions during 2012 and 2013. concerning their status and their progress towards resilience. The characteristics Over time, various factors – including can also be used to develop indicators policies, support, changes in context or to quantitatively assess resilience, using autonomous household adaptation and existing survey data. change – can influence the resilience of communities to shocks and stresses. Performing repeat assessments helps to Resilience level may be assessed based monitor not only trends in communities’ on how communities cope with and priorities in characteristics of resilience, but overcome various shocks and stresses: also their progress in attaining resilience those that are able to bounce back to their characteristics over time. A careful balance condition in the pre-crisis period, or even must be struck, however, between the need improve their situation, may be considered to maximize the accuracy and robustness resilient, while those that are collapsing of the findings and the need to maintain or are recovering but are worse off than the methodology as a cost-efficient, user- previously may not be resilient. friendly and practical tool.
To measure resilience and the impact The original conceptual framework for of interventions on resilience, baseline CoBRA emphasized the comparison information must be established. Doing of resilience attainment rates within so involves answering these fundamental communities between normal and crisis questions: periods (i.e., quantitative assessment). After a series of technical consultations, • What are the main characteristics of it became evident that there is greater resilience at community and household value in other aspects of the CoBRA levels? assessment findings, particularly the key
Figure 1. CoBRA model
Baseline CoBRA Time Repeat CoBRA assessment assessment Priority Priority characheristics of Risilience indicators characheristics of resilience Bounce back resilience • Community Direct interventions/services/support better • Community • Household • Household Informs External policy/political context Trends and Trends and Bounce back attainment attainment • Current/normal Households adaptation and change • Current/normal period Recover but period • Crisis period worse than • Crisis period before • Review resillience
Stresses and shocks indicators Collapse
2 Understanding Community Resilience 3. Identify the characteristics and strategies of disaster-resilient households; and 4. Identify the most highly rated interventions or services in building local disaster resilience.
The CoBRA methodology should not be perceived as a comprehensive or stand- alone tool for measuring resilience. The approach is largely a qualitative assessment tool, based on understanding resilience from a community perspective. It does not provide a quantitative measurement of the number or proportion of a population Cattle standing in a degraded field in Loitoktok, that has achieved resilience. The approach Kajiado County, Kenya aims to learn from positive experiences by identifying households perceived to factors, or priority characteristics, that be resilient – to understand what those contribute to building resilience to disaster households have or do differently that at community and household levels (i.e., enables them to cope better with shocks qualitative assessment). Such data have or stresses. been perceived as extremely critical to the process of identifying the resilience A key strength of the tool is the information impact indicators and prioritizing climate- it provides for further quantitative analysis. resilient policy, planning and programming As the evidence presented in this document decisions. demonstrates, the analysis facilitates the identification of a handful of characteristics 1.2 The CoBRA methodology of resilience. This information can then be used by other stakeholders to identify a set During development of the CoBRA of appropriate and manageable indicators methodology and implementation for monitoring over time. This qualitative guidelines, it became clear that little approach provides clarity and captures consensus exists among stakeholders as information on resilience pathways for to which components or characteristics communities that can’t easily be explained of disaster resilience are most important. by quantitative approaches. Consequently, the study team decided not to identify any particular components but rather to allow communities themselves to define resilience. This was particularly important because resilience covers such a wide range of activities and indicators. As a result, the methodology that emerged has four broad objectives:
1. Identify the priority characteristics of disaster resilience for a target community; 2. Assess the community’s achievement of these characteristics at the time of the assessment (generally carried out during a ‘normal’ period) and during A vehicle overturned in April 2013 by flush floods the last crisis or disaster; along river Turkwel, Turkana County, Kenya
Introduction 3 2. Field testing
Kajiado women during CoBRA assessment
All four CoBRA assessments were carried the two countries, and specific study sites out in line with the implementation were selected to ensure representation guidelines, which were developed based of different livelihood zones and levels on a series of consultations with a wide of intervention (see Table 1). Two sites, range of stakeholders. This was the first Marsabit and Turkana, are considered time the guidelines had been used in highly prone to drought, having practice, so they were adjusted and experienced two severe droughts in the improved as the assessments proceeded. last five years (2009 and 2011). In both These are described in more detail below. instances, the counties were declared highly food insecure and were the subject The field assessments were led by the of significant humanitarian response. By UNDP Drylands Development Centre contrast, the semi-arid Kajiado county in and government partners – the National Kenya and the Karamoja sub-region in Drought Management Authority (NDMA) Uganda are also prone to drought but to a in Kenya and the Office of the Prime much lesser extent. Kajiado has not been Minister in Uganda. The assessments considered highly food insecure for several were participatory – multi-agency teams years, and Karamoja was last the subject of comprising government, UN and NGOs a Famine Early Warning System Network operating in the respective field sites food security alert in July 2008. facilitated and participated in each assessment. 2.2 Approach 2.1 Identification of field sites The CoBRA methodology uses participatory qualitative approaches in the Field sites were chosen to represent a form of focus group discussions (FGDs) range of drought-affected locations in and key informant interviews (KIIs) to
Table 1. CoBRA field assessments Country County/sub-region Locations/districts Population* Assessment dates Marsabit Central, Laisamis and Kenya Marsabit 87,000 4–15 June 2013 Maikon Turkana North, South, Central, Kenya Turkana 179,000 24 June–6 July 2013 East, West and Loima Kajiado Central, Mashuru, Kenya Kajiado Loitoktok, Kajiado North and 214,000 19–30 August 2013 Isinya Uganda Karamoja Kotido and Kaabong 160,000 15–26 July 2013
*Approximate total population of the surveyed locations/districts
4 Understanding Community Resilience identify and prioritize the characteristics of that the 155 FGDs and 159 KIIs undertaken resilience. In each field site, data collection provide sufficient information for analysis was undertaken by three to four teams and learning. of four facilitators and one supervisor. Participating voluntarily, all the assessment Focus group discussion team members are based and working in the assessment locations with various FGDs are the main method for collecting governmental and non-governmental CoBRA assessment data. Typically, each organizations. Each team was given FGD has approximately 15 participants responsibility for undertaking 10–14 FGDs. who met the criteria of the sampling Each took an average of two to three hours frame developed as part of the facilitators’ to complete. training programme. During the training, trainees were asked to identify the most Prior to undertaking FGDs and KIIs in the appropriate approach to FGDs, to ensure field, facilitators and supervisors received that all views on resilience are heard. intensive training on the methodology. Facilitators repeatedly recommended that Training sessions took place in the main men, women and youth should participate town in each of the assessment sites. They in separate discussions, and the men were involved three to four days of exercises and often separated into older and younger one day of field testing, in which all trainees groups. For the purpose of the study, in undertook one FGD and one KII in pairs. view of the contexts of the assessment Afterward each team of two facilitators had locations, youth were defined as aged 15– individual debriefings with course trainers. 30. As local residents, facilitators were also Test FGD report sheets were included responsible for mobilizing participants for in the final survey results if they were FGDs by making links with elders, women’s deemed of adequate quality. Supervisors groups and other networks in each were responsible for supporting facilitators location to identify suitable participants. by maintaining quality and consistency in Participation was entirely voluntary. data collection and data entry during the field work. At the start of each FGD, facilitators were encouraged to spend some time Purposive sampling for FGDs and KIIs explaining the purpose and rationale of was used in each location. First, disaster- the CoBRA assessment to the community affected regions were selected. Then participants. This issue was discussed in these were systematically narrowed down depth during the facilitators’ training, as to livelihood zones and intervention the explanations are essential to managing levels that were sampled randomly, while participants’ expectations about receiving ensuring adequate representation of men, a direct intervention or benefit as a result women and youth. There is no statistical of their participation. This is discussed basis for the number of FGDs undertaken further in the limitations section. in each location or the various livelihood types or age and gender groups; the Another important issue, covered in detail number of FGDs and KIIs completed in the facilitators’ training, was developing was a factor of staff resources and time a consistent and locally understandable available. While the study is not statistically definition of resilience. It is a highly technical significant (and therefore cannot be term, and a direct translation does not exist transposed to the wider population), there in most local languages. Consequently, was clear consistency and repetitiveness facilitators were requested individually in the answers provided, suggesting that to translate the technical terms into local further sampling of similar populations languages in simple terms using words would consolidate findings. It is believed and examples that ordinary people could
Field testing 5 Box 3. Plain language definition of resilience in Turkana, Kenya (left) and Karamoja, Uganda
easily understand. These contextualized 5. Identification of resilient households in descriptions or definitions of resilience that community and the characteristics were then refined by the facilitator of these households; trainees and translated back into English 6. Trends in resilience – whether the and the local languages (see Box 3). This number of households considered exercise ensured that all facilitators were resilient is increasing, decreasing or using similar terms and definitions when staying the same; explaining resilience as a concept to focus 7. Recent factors/interventions that have group participants and key informants. improved resilience for some (or all) of the households; and After the explanatory session, the main 8. Future or additional interventions that focus group discussion proceeded by participants feel would further build taking community participants through the resilience. following topics: Assessment team leaders and local 1. The main crises or hazards affecting facilitators jointly selected the focus group that community; locations and the composition of the group 2. A set of outcome statements in each site, using population data. The describing the characteristics of a locations and people selected reflected a resilient community in their context; mix of the following criteria: 3. Ranking of statements to identify the most important ones; • Livelihood zones – pastoral, 4. Scoring of the community’s progress in agropastoral and peri-urban, etc.; attaining the priority characteristics in • High and low intervention areas – both the current (ideally normal) and locations that were relatively well crisis periods; served and those poorly served by basic services, infrastructure and other interventions; and
6 Understanding Community Resilience • Age and gender – people who 2.3 Constraints and limitations represented a range of the population of data collection in terms of age and gender. The balance was agreed following Raised expectations discussions with the trainee facilitators Initially focus groups expected that as part of the training. the teams of facilitators were there to undertake some form of needs assessment Key informant interviews that would potentially result in a benefit or intervention for that community. To manage In each location, one to four KIIs were inaccurate assumptions, expectations and undertaken with households perceived as biases, a community mobilization exercise resilient by the focus group participants. In was undertaken in most locations prior to some cases, resilient households were also the FGDs to explain the purpose of the identified through discussions with chiefs CoBRA assessment and clarify that no or other senior local residents. The KII is direct interventions were likely as part of a semi-structured interview that solicits it. An effort was also made to combine details on: facilitators from different agencies and sectors on teams. This helped avoid bias • Household composition, education towards the interests of specific agencies level and livelihood/economic activity and sectors in the participants’ responses. of each member of the household; • Factors that have contributed to the Scoring the attainment of resilience household’s resilience; characteristics • Pathways to resilience, i.e., steps taken by the household to become resilient; After the resilience characteristics were • Actions or strategies the household identified and ranked, focus group took to cope with recent shocks and participants were asked to score their crises affecting that community; and level of attainment of the top-ranked • Interventions and support that would characteristics. Facilitators were trained best assist others in their community to undertake this exercise by asking to become more resilient. participants to score on a scale from 0 to 10, with 10 being the complete Table 2 summarizes the FGDs and KIIs attainment of that characteristic and 0 an undertaken in the four assessment areas. absolute lack of that characteristic. This Further details of sample locations are was done to ensure a uniform approach to included in the assessment reports measuring the level of attainment for the (Annexes 1–4). dynamic and diverse set of characteristics, including the characteristics unrelated to Table 2. FGDs and KIIs undertaken in four households, such as peace and security, CoBRA assessments and more qualitative characteristics, such as access to health care services. Some Assessment Number of Number of inconsistency was found in how people site FGDs KIIs understood and scored the characteristics. Marsabit 41 41 Furthermore, the scoring of characteristics was sometimes skewed towards lower Turkana 42 42 values, and it required more probing from Karamoja 36 40 facilitators to bring the values to seemingly realistic levels. It was also difficult for Kajiado 36 36 communities to objectively define the scale Total 155 159 or level of attainment, which presented a challenge in most cases to both facilitators and respondents.
Field testing 7 Reluctance to being identified as Changing livelihoods resilient In Karamoja, livestock holdings per In some locations individuals identified by household were reported to have fallen focus group participants as members of dramatically over the years, mainly resilient households refused to accept this because of cattle raids, recurrent drought, designation. They were concerned that livestock diseases and increasing accepting this label might lead to high settlement of nomadic populations in the expectations for peer support from other region. In addition, government initiatives community members or result in their to introduce ox ploughs have increased exclusion from development programmes the acreage devoted to crop farming. or assistance. The assessment team took Thus, some communities that used to this sensitivity into consideration and be pastoral have lost grazing lands to strived to identify resilient households in subsistence crop farming and appear to a discrete manner while explaining the be transitioning to crop farming, despite purpose and implications of the exercise the unreliability of rainfall. This made it fully to the individuals and focus group difficult for the facilitators to identify the participants. appropriate livelihood category in some places. Consequently limited weight can be given to analysis by livelihood zone in that particular assessment in Uganda.
Women participants during CoBRA FGD in Laisamis district of Marsabit County, Kenya
8 Understanding Community Resilience 3. Findings
The Maasai cattle’s trekking to the grazing fields in Kajiado County, Kenya
This section summarizes the consolidated prices, reduced coping capacity and a findings from the CoBRA field work slow/limited humanitarian response. conducted in the four sites. The findings • In Karamoja, the drought of 2010–2011 are presented in the following categories: was referred to most frequently and agreed upon as the main crisis period. • What are the main hazards or shocks • In Kajiado, communities viewed facing the communities? drought as the most significant • What are the characteristics of a contributor to livestock losses and the resilient community? single most important factor limiting • To what extent has the community their resilience capacity. Communities achieved those characteristics? reported the most recent drought, of • What does a resilient household look 2009, as the main crisis period. like? • What interventions contributed to Other hazards or crises reported included household resilience, and what additional floods, conflicts, livestock diseases and interventions would best build resilience? human diseases. • How did key informants achieve and maintain resilience? 3.2 Characteristics of a resilient community 3.1 Main hazards or shocks Focus group participants were asked to state The main hazard reported in all the FGDs in all as many characteristics as they could think of four sites was drought. All sites also reported to describe a resilient community. Typically that the timing of the CoBRA assessment each group provided 15 to 20 statements. coincided with a relatively ‘normal’ period The participants were then requested to rank with minimal shocks or stress. and score the statements by importance. Each member was given six beans to rank • In Marsabit, the drought of 2010–2011 the three most significant statements (three was identified as the most recent crisis beans for the most significant statement, two period. for the second and one for the third) in terms • In Turkana, communities reported of priority for building resilience. The bean that, while past droughts had been scores were then totalled for each statement. longer, the drought of 2010–2011 was For ease of comparison, the statements were the most severe, and its impacts were grouped into the five sustainable livelihood exacerbated by extremely high food framework (SLF) categories.4
4 SLF presents the main factors that affect people’s livelihoods and the typical relationships between them. It identifies five core asset categories or types of capital upon which livelihoods are built: financial, human, natural, physical and social. For further details on SLF, please refer to UK Department for International Development (DFID), Sustainable Livelihoods Guidance Sheets (London, DFID, 1999).
Findings 9 Table 3. Highly ranked community resilience statements SLF Resilience characteristic Full statement category Financial • Access to credit • People would have good access to affordable credit and would be saving money (through banks, microfinance institutions and community savings and credit). • Productive farms • Farmers would be more productive and profitable (i.e., they would have inputs like quality tools, oxen, fertilizers and improved knowledge of good farming practices). • Employment • There would be many opportunities for jobs and other forms of paid employment through government, factories or other businesses. • Diversified IGAs • Many households would be involved in other IGAs such as small businesses and trading. • Livestock herds • Pastoralists would have herds large enough to sustainably support their families. • Pasture and fodder • There would be sufficient pasture (or fodder) for livestock at all times of the year. • Health care for livestock • The community would have access to high-quality and affordable animal health services, including veterinary services and vaccinations, whenever they need them. Human • Education • All children would be able to complete primary/secondary/ tertiary education. • Food security • All households would be able to feed themselves well every day. • Health care for humans • The community would have access to quality and affordable basic health care locally. Natural • Natural resource • Local rangelands and other natural resources would be well management managed so they do not become degraded over time. Physical • Access to markets • The community would have easy access to markets to buy goods and sell their produce. • Irrigation • Farmers would be irrigating land to improve the production of crops for consumption and sale. • Roads • There would be good-quality roads to the community. • Sanitation • Everyone would have good sanitation. • Shelter • Everyone would live in good-quality housing. • Telecommunications • There would be a reliable mobile phone network to all communities all the time. • Water for humans • The whole community would have access to sufficient, good- quality water at all times of the year. • Water for livestock • Livestock would have access to sufficient water at all times of the year. Social • Peace and security • The whole community would enjoy continual peace and security.
Table 3 lists the characteristics most characteristics of a resilient community. In commonly cited across the four study sites. most focus groups, statements addressing It is important to note that there was little education often referred to and ranked variation; the statements/characteristics secondary and tertiary levels of education were ranked similarly in all sites. (Each Annex as well as primary. In addition, while contains a complete list of the statements communities tended to have access to identified and the corresponding scores, water in some form, the FGDs stressed the providing some context.) consistent availability of clean water as the main characteristic. In relation to peace and Overall, education, water and peace and security, many communities felt their overall security were the most highly ranked situation was better at the time of the
10 Understanding Community Resilience Table 4. Top three resilience characteristics by gender and age group
Gender/age Marsabit Turkana Karamoja Kajiado group
Women Peace and security Education Productive farms Education Education Diversified IGAs Education Water for humans Water for humans Water for humans Livestock herds Health care for humans
Men Peace and security Education Productive farms Education Water for humans Peace and security Peace and security Water for humans Education Water for humans Education Health care for humans
Youth Education Education Education Roads Peace and security Access to markets Access to markets Education* Water for humans Access to credit Access to credit Water for humans* Employment* *These three characteristics received the same scores. assessment. However, many focus groups In almost all livelihood zones, education prioritized the statement with the caveat featured as a prominent characteristic that any deterioration of peace and security of resilience. Some notable differences could have an immediate and substantial include: impact on overall community resilience. • Pastoral areas consistently prioritized Analysis by gender and age water for human consumption, which is logical given that they are generally Statements were also analysed according least likely to be located close to to gender and age groups (see Table 4). permanent water sources. Access to livestock inputs was also highly rated While some similarities in priority when considered as a group: livestock statements can be observed among herds, water, pasture and health the groups, there were also some key inputs, etc. differences: • Peri-urban groups in Marsabit and Karamoja tended to mention peace • Women consistently mentioned and security more than the other education and water for human groups. Many of the peri-urban groups consumption as priority resilience were made up of pastoralists who were characteristics. formerly more mobile and had larger • Men tended to focus on peace and herds but have become sedentary as security, education and water for a result of insecurity as well as lack of human consumption. water and pasture, livestock diseases, • Youth focused on education, but etc. they were often more concerned with characteristics that build income Analysis by intervention level and wage opportunities, including access to markets, access to credit, In each assessment site, survey sites were employment and roads. also selected to represent a range of core and peripheral locations. Before the Analysis by livelihood group field work, all sublocations (Kenya) and sub-counties (Uganda) in the target area Table 5 summarizes the top resilience were mapped to identify those with high, characteristics by livelihood group. medium and low access to services and
Findings 11 Table 5. Top three resilience characteristics by livelihood group
Livelihood Marsabit Turkana Karamoja Kajiado group Agropastoral Peace and security Peace and security Productive farms Water for humans Water for humans Education Education Education Education Food security Health care for Health care for humans humans
Pastoral Education Education Productive farms Education Peace and security Irrigation Livestock Health care for Water for humans Water for humans Peace and security humans Water for humans
Agricultural - - Productive farms Education Access to credit Irrigation Education Access to markets
Peri-urban Peace and security - Peace and security Education Education Education Roads Water for humans Access to credit Productive farms
Urban - Education - Roads Diversified IGAs Access to markets Water for humans Education
Fishing - Education - - Water for humans Diversified IGAs
Note: Empty cells indicate the livelihood group was not interviewed in that location.
interventions such as higher level basic mentioned as key characteristics of services, roads, mobile phone network resilience for both high and medium coverage, cash transfers, market access, intervention areas. etc. (see Table 6).5 • By contrast, low intervention areas, while also ranking education highly, • Access to water for human consumption put more emphasis on human health and education were consistently and to some extent peace and security.
Table 6. Top three resilience characteristics by intervention level
Intervention Marsabit Turkana Karamoja Kajiado level
High Peace and security Education Productive farms Roads Education Diversified IGAs Education Education Water for humans Water for humans Health care for Water for humans humans
Medium Peace and security Education Productive farms Water for humans Education Water for humans Education Education Water for humans Diversified IGAs Livestock Roads
Low Water for humans Education Productive farms Education Peace and security Peace and security Education Health care for Education Health care for Health care for humans humans humans Water for livestock
5 A full list services/interventions mapped is provided in each of the individual assessment reports in the Annexes.
12 Understanding Community Resilience No significant patterns emerged in the resilience characteristics identified by the different intervention level groups. All groups gave priority to characteristics generally identified with education. Peace and security and health were mentioned more regularly by the low intervention groups, which may reflect the fact that as insecurity increases the provision of many basic services decreases, as does NGO activity. This highlights the critical role insecurity plays in perpetuating an area’s low intervention status.
3.3 Extent to which CoBRA facilitators in Kotido going through the training sessions communities achieved resilience characteristics Analysis - all respondents Table 7 provides the overall average Focus group participants were asked to scores of the perceived attainment rates score the extent to which they had achieved for the top-ranked community resilience their priority characteristics of resilience. characteristics in normal/crisis periods They scored each statement twice: first for in the four assessment sites. The top- the current period (agreed to be a normal ranked statements were not the same in period6) and second for the last significant every location; cells are left blank where crisis period. The scores are ranked on a that statement was not relevant to the scale from 0 to 10, with 10 being perfect location. Clearly, the scores represent the attainment of that characteristic (for focus group participants’ context-specific example, the entire community has access perceptions, which are not statistically to sufficient safe water at all times during a significant and should not be compared calendar year), and 0 being no attainment among the assessment sites. However, (no one in the community has access to they do provide a useful sense of how the sufficient safe water at all times of the community members perceive different calendar year). aspects of their resilience. This can be especially critical for measuring less This section does not include the analysis quantitative factors such as peace and by gender because focus groups were security. asked to rank the attainment of resilience characteristics for the entire community. Peace and security and water for livestock Therefore any differences between men are the characteristics of resilience that and women in the same community would had the highest attainment scores in the be based on perceptions. For further current period for most locations. This is details on the results per SLF category likely due to favourable climatic patterns in and by livelihood groups and level of the recent past and relative social stability. interventions, please refer to the reports Other highly scored statements included for each location in Annexes 1 through 4. pasture and fodder (Marsabit) and water for livestock (Marsabit, Karamoja and Kajiado).
6 Communities in all areas considered the situation at the time of the assessments to be ‘normal’ except for the Kaabong district in Karamoja sub-region. Some focus groups at that site felt that the situation at the time of the assessment was bad. However, the crisis period to which they referred in scoring was a more intense flood crisis that took place the previous year.
Findings 13 Table 7. Community attainment of resilience characteristics – three highest/lowest-ranked statements
Current Crisis Current Crisis Current Crisis Current Crisis Characteristic Marsabit Turkana Karamoja Kajiado
Overall average 3.7 2.4 2.5 1.3 4.8 1.9 3.5 2.3
Three highest scoring statements
Pasture and fodder 6.7 2.1 ------
Peace and security 6.1 4.1 3.3 1.8 7.1 2.2 4.5 4.2
Water for livestock 5.9 2.9 - - 5.2 1.6 4.4 1.4
Livestock herds - - 3.2 2.0 - - - -
Natural resource - - 3.1 0.7 - - - - management
Education - - - - 5.6 2.2 - -
Access to markets - - - - 5.2 2.3 - -
Access to credit ------4.4 3.2
Three lowest scoring statements
Access to credit 1.0 0.5 1.4 0.8 3.7 1.3 - -
Productive farms 2.0 0.0 ------
Telecommunications 2.1 1.1 ------
Irrigation - - 0.9 0.6 - 2.5 1.6
Roads - - 1.8 1.1 - - 2.1 1.4
Livestock herds - - - 3.6 3.8 - -
Employment - - - 4.0 1.3 2.1 1.3
Note: For each location and period, only the three top-ranked and bottom-ranked characteristics are listed.
While access to credit was scored highly in When the levels of resilience achieved in Kajiado (implying this was a characteristic a normal period and a crisis period are that most of the community had attained), compared, several points can be noted: it was one of the lowest scoring factors for the three other sites. This is likely because • In most locations, communities did of the proximity of Kajiado county to the not perceive that they had progressed capital city of Nairobi and the presence of towards attaining their priority many microfinance institutions in the area. characteristics of resilience (i.e., score Other characteristics that received a low mostly less than 5 out of 10 even in score from two of the four sites include a normal period). This highlights the irrigation (Turkana and Kajiado), roads chronic underlying vulnerability of all (Turkana and Kajiado) and employment these communities. (Karamoja and Kajiado). Characteristics that • It is not possible to compare the scores received a low score from one site included of one site with those of another to productive farms, telecommunications and assess relative levels of resilience. First, livestock (herd size), representing highly as the assessments in each site were varied sectoral conditions in the various conducted independently, the scores contexts. cannot be seen as relative to each
14 Understanding Community Resilience other. Further, because each site had Hence its communities are more variations in its priority characteristics, heavily affected by shocks and crises. the communities were not always ranking the same set of indicators. For 3.4 What a resilient household example, in terms of food security and looks like other human development indicators, Kajiado is relatively better off than the Focus group participants were asked to other sites, but its resilience scores describe the characteristics of households were similar to the other areas. This that were more resilient compared to may demonstrate the context-specific others, that is, the households that had and dynamic nature of resilience attained many or all of the resilience concepts: while some characteristics characteristics prioritized. The top are universally and persistently viewed characteristics of a resilient household, as important (such as education), cited consistently by focus groups across the priority of other characteristics all four locations, were: changes over time as the members of a community transform their lives and • Having a member with employment or livelihoods. An example is the recent wage labour; shift of livelihoods from pastoralism • Having a business or other IGAs less to agriculture and other alternative dependent on the weather; livelihoods in Kajiado. • Having a large herd; and • Having a large farm. Nonetheless, there are interesting comparisons to be made in the differing These are all largely related to gaps between the normal and crisis diversification of risk, in the form of either periods. Karamoja communities felt alternative or reliable forms of income or they had attained the highest level of significant assets (such as herds and land) resilience overall, but proportionately that allow a family to absorb or mitigate the their scores dropped more significantly impacts of shocks and stresses. The section than elsewhere during a crisis on interviews with key informants contains period. There are several possible more detail on the interplay between the explanations for these results. One is factors that propel a household towards that communities are better able to greater resilience. withstand shocks in Kajiado as a result of their higher level of development Focus groups were further questioned and access to markets. Conversely, as about whether the number of resilient communities in Turkana already have households was increasing, decreasing such a low level of resilience, it is hard or staying the same. Table 8 shows for the scores to fall further during a the average proportion of focus group shock. Additionally, while Turkana is members citing increasing resilience. As chronically underserved in all sectors, above, not every group was interviewed during crises it receives relatively in every location, and therefore blank cells comprehensive levels of humanitarian are left where the data were not gathered. support in terms of food aid, cash Some of the findings include: transfers, water tankering, etc. This in turn may make the community • Women tended to be more optimistic feel more supported in crisis times, than men and youth, with the exception even though only for the short term. of Turkana, where youth were the most Karamoja, by comparison, is not so optimistic. chronically underserved but may not • There was little consistency across have such comprehensive and timely livelihood groups. In Marsabit and response programming in place. Kajiado, over 75 percent of the pastoral
Findings 15 Table 8. Proportion of households citing increasing resilience
Marsabit Turkana Karamoja Kajiado
Overall average 60% 46% 52% 63%
Gender/age
Women 69% 40% 62% 73%
Men 56% 43% 40% 46%
Youth 60% 58% 47% 67%
Livelihood groups
Agropastoral 38% 9% 67% 60%
Pastoral 76% 25% 10% 73%
Agricultural - - 50% 86%
Peri-urban 60% - 75% 0%
Urban - 18% - 33%
Fishing - 100% - -
Intervention level
High 17% 59% 54% 50%
Medium 81% 36% 50% 67%
Low 71% 25% 50% 75%
Note: All figures below 50 percent are highlighted. Empty cells indicate the assessment did not focus on that livelihood group in that location.
respondents felt that resilience was By contrast, the reasons for decreasing increasing, whereas in Turkana and resilience were largely associated with Karamoja 25 percent or less expressed concurrent drought, conflict, disease, that optimism. Agropastoral groups limited resources and lack of employment/ were very positive in Karamoja and income-generation opportunities. Kajiado, but there was much more pessimism in Marsabit and Turkana. 3.5 Interventions that The proportion of urban and peri- contributed to household urban groups sensing an increase in resilience similarly ranged from none resilience to three quarters of the respondents. • Groupings by intervention levels also Communities were asked to list all the showed no clear patterns. services and interventions they had benefited from in the last two to five years. The main explanation given by A reasonably wide range of sectoral and respondents for increasing resilience public, non-governmental and private related to the positive spiral of improved interventions was mentioned. These access to education in the assessment included water, education, livestock locations: more educated children lead to restocking, cash transfers, health better access to diversified IGAs, they felt, services, mobile phone coverage, inputs which in turn results in higher household to productive farms, roads and other and community prosperity and a more livestock support. From this long list, each empowered community. focus group was asked to identify jointly
16 Understanding Community Resilience the three current or previous interventions cited interventions (combining present that had been most beneficial in building and future) are summarized in Table 9. their resilience, and to explain why. The table shows the repeated and clear Groups were also asked to list the priority given to water, education and three additional interventions they felt health interventions. These interventions would best build their resilience. Many reflect the high ranking given to these communities restated interventions similar factors as characteristics of resilience to those mentioned in the first list, with the by all focus groups. Water interventions justification that the current provision or were prioritized for obvious reasons, scale of intervention was too limited and particularly for improving food security should be expanded. The most commonly and livelihoods. These included any
Table 9. Resilience-building interventions most commonly cited by focus groups
Type of intervention Marsabit Turkana Karamoja Kajiado Total
Education Bursaries and scholarships; construction 38 40 33 46 157 or renovation of school facilities, including boarding facilities
Water Water source improvement, improved 48 35 33 33 149 storage capacity
Health care Improvements to health services, staffing 20 26 29 32 107 or facilities
Inputs to productive farms Irrigation, greenhouses, oxen, agricultural 16 12 25 19 72 extension services, etc.
Restocking Programmes restocking livestock, 28 15 10 3 56 particularly with drought-resilient breeds or animals such as camels
Access to credit or other forms of 12 17 18 7 53 business support
Roads 12 9 9 15 45
Cash transfers Transfer of resources targeted at the most vulnerable populations such as the 18 20 - - 38 chronically food insecure (e.g. hunger safety net programmes) and children (e.g. child sponsorship programmes)
Other livestock support Livestock markets, health services, fodder 10 3 5 7 25 for production, etc.
Mobile phone coverage 15 - 1 2 18
Markets Safe and constant access to markets for - 4 5 9 18 buying/selling goods Note: Empty cells indicate the interventions were not cited in those locations.
Findings 17 interventions that expanded water • Pathways to resilience; sources and water storage facilities, such • Ability to cope with recent shocks and as tanks at household level or water pans hazards; and at community level. Education was seen • Priority interventions recommended as a benefit in itself and one that would by resilient households. also lead to improved life chances, such as employment for children. Interventions Composition and characteristics of such as scholarships, bursaries and resilient households boarding schools were regularly cited as important for ensuring that children The KII record sheet records the age, gender, completed higher levels of education. education level and economic activity of all members of the household interviewed. Health interventions were also perceived The size of resilient households varied as critical. Household resilience can be considerably from 1 to 14 members, with seriously undermined by the illness of a an average household size of 6 members household member, leading to a significant for each site (Table 10). The table also loss of productive time and income, shows the average educational attainment especially when health facilities and of members of resilient households and services are lacking or costly. Agricultural the average literacy rate for the area. interventions also ranked highly in three of Although the average literacy rate is the four sites. not a directly equivalent comparison, it suggests that resilient households have 3.6 How key informants significantly higher education levels than achieved resilience average. Additionally, the majority of school-age children in these households The second major form of data collection were reported to be in full-time education. for a CoBRA assessment is key informant In areas where very low proportions of the interviews undertaken with members of population have education, those with households identified as resilient. Between education are far better placed to access 36 and 42 KIIs were undertaken in each any job opportunities that arise. Education assessment site, providing a total of 159 also improves commercial and financial interviews for analysis. KIIs followed a semi- literacy, enabling households to engage structured interview format that examined better with markets and develop IGAs. the following four areas: All 159 key informants had household • Composition and characteristics of the member(s) engaged in one or more of the household; following activities:
Table 10. Size and education levels of resilient households
Location Marsabit Turkana Karamoja Kajiado
Average household size 6.0 5.7 6.2 6.4
Percent of households with at least one member who 74% 69% 66% 89% has completed primary school
Percent of households with at least one member who 48% 43% 34% 69% has completed secondary school or higher
Average literacy rate for the area7 26% 18% 21% 55%
7 This is not a directly comparable statistic as it refers to individuals, not households; however, recent data on education completion rates per household are not easily available.
18 Understanding Community Resilience • Wage employment or casual labour; These findings strongly reinforce the • Business or petty trade; defining characteristics of resilient • Livestock raising; households cited by focus group • Agricultural production; or participants. These characteristics were the • Fishing (only in Turkana). only findings that were uniformly accepted by the participants in the validation Table 11 shows the percentage of workshops as valid in each assessment households benefiting from different income site. sources. In all sites, the great majority of resilient households had multiple income Interestingly, a 2013 study by the Kenya sources; only in Marsabit did the proportion National Bureau of Statistics and Society of such households fall below 90 percent. for International Development highlighted Marsabit was also the only site where a the clear link between individuals’ significant number of resilient households educational attainment and participation relied exclusively on livestock, consisting of in the labour market in Kenya: pastoralists with large herds. The table also shows that the diversified income sources “People with no education in Kenya do not replace their traditional agricultural are 1.7 times more likely to have no or pastoral activities but are in addition to work than people with secondary them. Turkana is the only site where more education or above. Employment of the resilient households interviewed for pay is higher for individuals benefited from diversified livelihood income with a secondary education in than from on-farm or fishing income. both rural (21.3 percent) and urban (43.2 percent) areas, even though In all assessment sites, the diversification employment for pay in urban areas is of income sources emerged as the key twice the employment for pay in rural characteristic of resilient households. areas. Overall, individuals living in Most were supplementing traditional on- urban areas who have no education farm activities with wage and/or business are twice as likely to be without work incomes. Table 11 shows the percentage as their rural counterparts. Higher of key informants who had multiple income educational attainment is associated sources and then breaks down the type of with lower participation in agricultural income relied upon. activities.”8
Table 11. Income sources of resilient households
Total / Location Marsabit Turkana Karamoja Kajiado average
Total number of key informants 41 42 40 36 159 Percent with multiple income sources 75% 90% 98% 100% 91% Percent with agricultural, pastoral or 80% 78% 95% 98% 87% fishing income Percent with income-earning activities/ 48% 83% 80% 86% 74% small business income Percent with wage or casual labour 51% 45% 53% 58% 52% income
Note: Figures do not add to 100% because each household can name multiple income sources.
8 Kenya National Bureau of Statistics and Society for International Development, Exploring Kenya’s Inequality: Pulling Apart of Pooling Together? (Nairobi, Ascent Limited, 2013).
Findings 19 resilient households were in a positive spiral of income growth and asset accumulation.
Good livestock management, particularly timely destocking and restocking of herds, was repeatedly mentioned by resilient households in pastoral areas. The proceeds from timely sales of livestock were used to support the remaining livestock (for example, to purchase water, fodder, health care); pay for the households’ basic needs, such as food and school fees; and restock after the crisis period. This ensures that household herd sizes can ‘bounce Traders and camels in a sale yard in Kakuma, Turkana back’ relatively quickly following a shock, County, Kenya including drought.
Pathways to resilience Over one quarter (27 percent) of When respondents were asked how respondents mentioned the role of saving they became (and remained) resilient, and loans in expanding their income or responses were consistent in all assessment income sources. This was most common areas. Virtually all respondents cited their in Kajiado and least in Karamoja. The multiple income sources as the reason majority of households were involved in for their resilience. In particular, resilient some form of savings and credit group households often mentioned non-farm structures, while others borrowed from income sources, which are generally less family or friends. In Karamoja, dowry dependent on rain and thus less affected payments were mentioned as an income by drought. used to start businesses or grow herds.
Another regularly repeated theme was A few respondents cited their good the use of one income source to expand household management or their business- or improve others and build assets. For mindedness as factors supporting their example, households with a wage earner resilience (see Box 4). Two respondents or business regularly explained that cited support they received from NGO income from either of these sources had projects as factors in making them resilient. been saved and used to start or grow One household was given a camel as part businesses, expand livestock herds or of a restocking project and another was invest in agricultural production. The value a beneficiary of a voucher redemption of education in enhancing incomes was project.9 also repeatedly mentioned. Again, resilient households noted that having multiple Ability to cope with recent shocks incomes and more assets (especially and hazards bigger herds) enabled the households to keep more children in school for longer. The majority of key informants referred Typically, these households could afford to drought as the major hazard they had to sell livestock to pay school fees, which faced. Most indicated that they were better increased employment chances. Hence placed to cope with this crisis than others
9 The beneficiary was a trader in the project whereby the most food insecure were issued with vouchers for a certain locally available food item – milk, meat or fish. The vouchers could be redeemed with local traders who were in turn paid by the project. The increased and stable demand assisted in growing the traders’ business.
20 Understanding Community Resilience Box 4. Examples of interviewees’ pathways to resilience
Marsabit: “I knew in the middle of the year it was going to be bad so I sold five goats and used the money to hire a vehicle to take the rest of my herd to where there was pasture.” “I went to the mission school; the nuns gave me a scholarship to complete secondary. Then I got a job with an NGO. My wife is a teacher. We also trade goats.” Karamoja: “I was the only son in a family with four sisters. We got cows as bride price for each one so I inherited a large herd. I sell animals each year to pay for school fees. My oldest now has a job in Kampala and also sends money for schooling.” Kajiado: “After finishing secondary school I got a job as a farm manager for three years. I used the knowledge I gained to start my own farm growing vegetables. With the money I made I bought more land to expand production and build a shop. I also now rent some of the land.” “My husband died and left me a cow. We now have three, so we always have milk. I worked as a farm labourer to pay school fees, my oldest got bursary from CDF to complete secondary. Now she has a job and pays school fees for my other children.” Turkana: “I sold one camel to set up a shop. I also have a herd of 100 goats and a son working in Lodwar who sends money.” “When goats started dying due to PPR in 2008, my family sold 30 goats and started diversifying into business.” due to their additional income sources On the whole, it was clear that resilient and assets. Those with wage and business households coped better with drought than income noted that these income sources others due to their higher and more diverse were not so affected by drought and incomes. These are a result of longer- therefore could be relied upon through term adaptation and investments during these periods. Timely sale of livestock and beyond drought crisis periods, such was repeatedly mentioned by pastoral as keeping children in school or starting households as a coping strategy as well as businesses in non-drought periods. Good a longer term route to resilience (i.e. longer herd management, including the timely term adaptation). Moving livestock to sale of livestock, was the most significant areas of good grazing and water was also short-term or immediate coping strategy mentioned by households with livestock. that took place during or before a hazard period for pastoral groups. Other coping strategies that were mentioned with less frequency included Priority interventions the use of savings (sometimes through savings or credit groups or structures); Key informants were asked for the three loans from friends and family; credit from most important interventions to improve traders; and reduction of expenditures and their communities’ resilience. Table 12 lists consumption. the four intervention areas most commonly cited by each of the four assessment sites. A minority of respondents (approximately 10 percent) mentioned specific The commonality of interventions cited interventions as useful coping factors. across and between assessment sites shows These included cash/food for work a clear preference for interventions that schemes; cash transfers such as the increase productive assets and business hunger safety net programme in Marsabit skills, and hence income. Although many and Turkana; training from NGOs; peace or all of these interventions were also forums; and water tankering. mentioned by focus group participants,
Findings 21 Table 12. Resilience-building interventions most commonly cited by key informants
Ranking Marsabit Turkana Karamoja Kajiado
1st Livestock production Access to credit, Agricultural Access to credit, inputs business support production inputs business support
2nd Access to credit, Irrigation Education Good livestock business support management practices
3rd Water Education Access to credit, Education business support
4th Agricultural Water Water Agricultural production inputs production inputs
key informants generally rated them be taken by households themselves rather higher. This is understandable, as the than externally delivered interventions. majority of these households have small This reflects a much greater level of self- businesses and a more business-oriented reliance at this site, which has not had the approach to their on-farm activities. same history of NGO and governmental humanitarian and development It is also notable that resilient households programmes. Consequently many did not mention interventions related idiosyncratic actions or interventions were to humanitarian response, such as mentioned in this area alone, such as restocking and cash transfer programmes. “people should practice rotational grazing Interestingly, in Kajiado, many respondents techniques” and “store acacia seeds for referred to actions that should and could fodder”.
22 Understanding Community Resilience 4. Summary of feedback from stakeholder consultations Health facility in Kajiado County, Kenya
The findings of the CoBRA analyses were In all the review and feedback workshops, presented to local stakeholders, both participants largely agreed that the ranked community representatives and technical resilience characteristics resonated well stakeholders, across all the sites where the with the reality in these communities assessment was conducted in both Kenya and, in general, the characteristics and Uganda: prioritized by the communities were what the stakeholders expected. They also • 3–4 September in Marsabit, Marsabit acknowledged that a common thread of county; resilience characteristics emerged across • 2–3 October in Lodwar, Turkana all locations, even though the surveyed county; communities were diverse in terms of • 29–31 October in Kotido and Kaabong nationality, age, ethnicity, religion, sex and in Karamoja sub-region; and socioeconomic status. • 6–7 November in Kajiado, Kajiado county. The three most commonly cited and relatively equally prioritized resilience The assessment results from Kotido and characteristics were in the areas of: Kaabong districts were also presented to stakeholders at the national level in • Education; Kampala, Uganda, on 28 November 2013. • Water for humans; and • Health care for humans. These sessions proved to be a critical process in the CoBRA methodology as Peace and security also emerged as a they encouraged dialogue on community critical priority for resilience, particularly in and household resilience among a diverse the northern Kenya assessments (Marsabit and dynamic set of stakeholders, both and Turkana). The feedback sessions those drawn from the community and emphatically validated the two most highly technical experts.10 They also helped ranked resilience-building interventions in establish a common local standpoint all assessment locations, those involving on the issues under investigation while the education and water sectors. cultivating ownership of the assessment results. This section highlights some of the At the same time, all the feedback sessions main feedback points on the assessment noted that the communities failed to fully outcomes, putting the findings in context. capture and appreciate the contribution
10 To solicit participants for the technical workshops, NDMA sent invitations to relevant government, UN and NGO partners operating in the location, including all the organizations participating in the assessment. Participants in the community workshops were chosen to represent different livelihood, gender/age and intervention groups; some but not all had participated in FGDs/KIIs.
Summary of feedback from stakeholder consultations 23 of natural resources management to In all the feedback sessions, participants local resilience-building, and the roles agreed with the accuracy of the and importance of natural resource characteristics of resilient households in management were largely subdued both terms of education level and diversified in priority resilience characteristics and income sources combined with resilience-building interventions. This was pastoralism/farming and other IGAs. partly due to lack of awareness; given that these resources have always been present, In all cases, participants emphasized the they are taken for granted. Also, there has important role the assessment results been limited community ownership of local would play as a planning support tool, given natural resources across the assessment that they largely represent community sites. views, perspectives and aspirations on resilience-building. In particular, the Many feedback sessions also raised stakeholders made the following specific a concern that the livestock sector recommendations: was undervalued and did not feature as significantly as it should in the • Continue to support pastoralists as a assessments, given the proportion of the principal livelihood group, increasing population whose livelihoods depend the focus on services and interventions entirely or partially on this sector. This for improving livestock production and could be attributed to the fact that the productivity. livestock-related characteristics were • Develop opportunities to diversify disaggregated during the analysis (such livelihood options, with a next step as large herds, access to livestock inputs of mapping out these opportunities and water for livestock) to allow for more and presenting them by relevance to in-depth insights and additional detail on locations. potential resilience-building interventions • Disaggregate and investigate key in the sector. There is therefore a need characteristics more closely, such to aggregate the livestock-related as education as a critical factor for characteristics so that pastoralism receives resilience, and make an effort to sufficient priority for building resilience. determine what has hindered or improved educational development Participants in the feedback sessions and what areas require greater expressed caution with regard to the investment (quality, enrolment levels, resilience attainment scores, given that etc.). they are based on perceptions, which can • Identify ‘keystone’ indicators of be arbitrary and heavily influenced by resilience based on the assessment externalities. For example, participants results for continuous monitoring noted, the focus groups might have of progress. For example, are there been influenced by current seasonal a number of indicators that can be conditions (such as whether the session monitored independently as key took place in the wet season or the dry markers of resilience? These could then season). Consequently, the feedback be monitored more effectively without sessions concluded that even though introducing the other variables that these scores are useful for understanding could dilute or introduce noise into the community perceptions, they may not be analysis. Similarly, it would be useful to fully objective. Thus they are not useful for compare CoBRA indicators to existing purposes of comparison and should be data sets, such as poverty indicators, complemented by other quantitative data. Household Economy Analysis data, etc.
24 Understanding Community Resilience 5. Lessons learnt from CoBRA assessments in Kenya and Uganda Young men during the CoBRA FGD’s in Magadi, Kajiado County, Kenya
5.1 Benefits of the are contextual, varying by location, methodology although with many commonalities. It seems that household characteristics The CoBRA methodology represents one represent what resilience looks like of the very few practical tools that have as an end state or impact, whereas been developed and tested to measure community characteristics highlight resilience at the community level. The the key outcomes required to reach process of developing and testing it has this end state. Hence education level highlighted several benefits: is clearly linked with improving and diversifying household income. • It allows the community to define • It provides a basis for identifying resilience. CoBRA does not attempt quantitative impact and outcome to pre-define the components indicators. The community and or indicators of resilience. The household characteristics of resilience participatory nature of the approach provide a valuable basis upon which to brings communities’ perspectives into develop locally specific indicators. By the debate, giving them an opportunity using the communities’ prioritization to describe what resilience means to of issues affecting or defining them as a community and as individual resilience, stakeholders can avoid households. The characteristics being drawn into measuring indicators and issues that have emerged have in every livelihood category. These often diverged significantly from the quantitative indicators may already be priorities of government bodies and a part of data collection/monitoring other development stakeholders. processes, or they could be added • It combines community and household into them. It is not recommended characteristics: The combination that a CoBRA assessment be used to of community and household generate another major quantitative characteristics provides a holistic household survey. Ideally, by drawing understanding of what resilience looks on a relatively small set of indicators (5 like in reality. It also helps in clarifying to 10), local resilience can be measured the links between household and and tracked more systematically over community resilience. For example, time. In addition, CoBRA findings household resilience is closely linked can support identification of the most to financial security in terms of higher appropriate indicator to measure income and a diversified asset base. priority resilience characteristics. These characteristics of household For example, many local authorities resilience are emerging as consistent collect exhaustive data on education. and universal in all locations. However, However, where governments do community characteristics of resilience not provide universal access to
The CoBRA approach 25 secondary education, they may not planning processes, as they represent be systematically collecting data on community views on key issues. The the percentage of households with at communities’ ranking of the most least one member having completed useful interventions and services for secondary education. building resilience provides valuable • It is relatively quick and inexpensive input in prioritizing local investment to undertake. A CoBRA assessment decisions. One of the main advantages provides a substantial amount of of CoBRA stressed by the pilot users information in a relatively short period is that it can be undertaken without of time and at significantly less cost than technical expertise (although actors equivalent quantitative approaches. working in development, DRR and This is due to the participatory approach, community-based resilience-building which involves collaboration with local will clearly be important to the process). governmental and non-governmental The generic nature of the information organizations, which also provide generated means it can be equally technical and logistical backstopping useful to all government/NGO or support. Table 13 summarizes the development and humanitarian actors direct costs of undertaking the four in a community. Consequently there CoBRA assessments. are significant opportunities to share • It provides a good basis for strategic costs if organizations work together planning. CoBRA findings make a to undertake a CoBRA assessment in useful contribution to local strategic their sites.
Table 13. Direct costs of CoBRA assessments undertaken to date
Number Number Number Number of Total Region Budget breakdown (US$) of districts of FGDs of KIIs partners (US$)
Training 2,200
Travel 7,300 Marsabit 3 41 41 17 14,200 Security 400
Mobilization 4,300
Training 3,000
Travel 5,100 Turkana 6 42 42 26 16,000 Security 700
Mobilization 7,200
Training 1,900
Travel 3,200 Karamoja 2 36 40 30 12,600 Security 700
Mobilization 6,800
Training 1,500
Travel 1,000 Kajiado 5 36 36 21 9,700 Security 0
Mobilization 7,200
26 Understanding Community Resilience 5.2 Limitations of the • Resilience attainment scores are methodology not comparable between locations. Resilience scores cannot be compared Resilience is a highly complex concept. It is between locations, as they are inherently multidimensional and dynamic, derived from separate conversations which is why so few practical measurement in different contexts and therefore tools have been developed. Nonetheless do not relate to each other in any it has limitations: systematic way. This is illustrated by the equally low attainment scores for • A CoBRA assessment is not a stand- Kajiado and Marsabit, despite Kajiado alone measurement of resilience. being seen as much more resilient to A CoBRA assessment should draw drought than Marsabit by most other from and add to existing monitoring, resilience metrics. It is also not clear measurement and assessment whether resilience attainment scores processes. It provides a community’s in a community can be tracked over perspective on resilience and useful time with any consistency. The priority qualitative information on the characteristics are largely a function of important characteristics of resilience; the long-term visions and perspectives however, it does not offer a definitive of the communities and hence are quantitative measurement of resilience expected to remain unchanged for in terms of clearly identifying how some time. However, as communities many communities or households in are constantly transforming their lives an area have achieved some or all of and livelihoods, a repeat CoBRA these characteristics. assessment in the same sites after • Resilience attainment scores are several years might also result in the perceptual and subject to change. selection of a different set of priority The resilience attainment scores are characteristics. As demonstrated based on perceptions and are not by the cases of Kenya between arid statistically significant, so they cannot regions (e.g., Marsabit and Turkana) be extrapolated to a wider audience. and semi-arid regions (e.g., Kajiado), In the pilot it was extremely difficult it seems likely that as characteristics to assess how communities scored receive more attention over years, the current/normal period. Some they may become less of a priority. communities focused on the current • CoBRA cannot be used to evaluate season rather than an average for the individual programmes or services. whole year, which has implications for Many agencies are looking for seasonally variable characteristics such resilience assessment tools that can be as water availability or security. The used to measure the degree to which scores should rather be understood as their particular programme or support a key reference to foster deliberations influenced a community’s resilience. on specific livelihood/age/gender CoBRA cannot respond to this groups and sectoral resilience demand, and it is questionable, given characteristics. During the review and the nature of resilience, whether any feedback sessions, questions such tool can. CoBRA findings reinforce the as why certain groups score certain multidimensional nature of resilience characteristics as they do often led and the need for interventions to unique contextualized discussions, involving several sectors and agencies, which provided rich information for often at scale. Consequently, CoBRA decision-making. These tend to be cannot replace individual project or overlooked in broad countywide or programme evaluations. district-wide planning, implementation and monitoring processes.
The CoBRA approach 27 5.3 Indicators of resilience
The identification of appropriate indicators to measure resilience is a key debate among stakeholders working in the area of resilience-building. The conceptual framework for this study is based on the premise that resilience can be measured in two ways:
1. A universal measure or indicator(s) of resilience supports understanding of whether resilience is increasing, decreasing or staying the same; and 2. Composite and contextually specific Women participating during CoBRA assessment in indicators of resilience support Moroto, Uganda understanding of how local drivers of resilience are expanding or contracting, One of the most suitable monitoring and the impact of interventions on tools tracking resilience is the Household those drivers. Economy Approach (HEA). Using detailed household-level data to compare A universal measure of resilience conditions in a reference year to those in the current or modelled year, it assesses the The almost uniform evidence of resilient impacts of changes on households’ ability households having multiple sources of to meet a set of defined minimum survival income and a strong asset base indicates and livelihoods protection requirements. It there is potential for measuring absolute collects data on all sources of household resilience. Clearly, resilience involves income and assets, as well as information several dimensions – so, for example, the on how these various sources expand drivers of change that allow a family to have and contract in response to shocks and multiple sources of income may include stresses. education (linked with broader income- generating opportunities such as wage For the purpose of analysis, this study labour), access to credit (to facilitate income proposes to link the ‘resilience threshold’ generation and diversified livelihoods) and to the ‘livelihoods protection threshold’, sufficient water (to support a livestock herd which represents the total income required or improve agricultural output). to sustain local livelihoods (see Figure 2). To be disaster resilient, a household (or These findings confirm the hypothesis community) should be able to maintain a underpinning the conceptual framework: sufficient level of income and production that households would define themselves above the livelihood protection threshold as resilient if they were able to feed their during both normal periods and crisis families adequately every day and meet periods to meet the minimum required basic needs on a consistent basis in times expenditure and consumption. In the HEA, of shock and stress as well as in ‘normal’ this threshold consists of multiple sources times without external relief. Quantifying of income and assets sufficient for the this ability to cope, i.e., resilience family to ensure survival, maintain access threshold, would then provide a universal to basic services and sustain livelihoods.11 indicator of resilience.
11 Note that the CoBRA conceptual framework contains more detail on the HEA.
28 Understanding Community Resilience In the case of the most common climate- hold in all livelihood zones, echoing the related shocks (such as drought and findings from the CoBRA assessment. To flood), households generally experience this end, the HEA, described in greater a reduction in income and production, detail in the conceptual framework, may largely related to a decline in weather- provide a potential tool for modelling the dependent activities, including rain-fed ability of households to access enough agriculture and livestock production. income and assets to meet basic needs, in Households could be considered resilient both shock and ‘normal’ times. This would if they have other sources of income and allow for quantitative tracking of resilience production or some form of contingency levels using existing data sets. buffer exceeding the expected losses arising in a crisis period, and if they can Indicators for components of resume levels of income and production in resilience a timely manner after the crisis period. At the same time, the CoBRA assessment This finding is supported by other efforts to also provides a wealth of information on the measure resilience. For example, a recent development of indicators for components study12 by Save the Children assessed of resilience. As mentioned in the previous consolidated data sets for the HEA and section, the four assessments to date Cost of Diet (the minimum amount of have consistently highlighted the same money a household must spend to meet its household characteristics of resilience. energy, macronutrient and micronutrient Community characteristics are more varied, requirements using locally available foods) but with strong commonalities. Linking in an attempt to find common trends these characteristics with quantitative around food security, assuming that a food- indicators that could be tracked over time secure household can cope with disaster is the logical next step for the CoBRA risk. The patterns suggest that households methodology. Theoretically, this would that can manage change have multiple provide more information on the number sources of income and access to wage of households that are resilient and trends labour, and their livestock is an essential in the factors driving resilience. source of cash income. These findings
Figure 2. Links between resilience threshold and livelihoods protection threshold
140 Livelihoods protection threshold 120 Survival threshold
100
80 Crops Livestock 60 Labour Remittances 40 Charcoal sales Minimum food requirement 20 Other basic needs Percent of household food requirements Percent School fees 0 Income and production Income and production Minimum required normal year crisis year expenditure and consumption
12 Save the Children, Livelihoods at the Limit: Food Security in a Changing World, 2013.
The CoBRA approach 29 Table 14 provides an example of impact results. For each indicator the table and outcome indicators that could be highlights some of the challenges and used to track resilience more quantitatively issues associated with this approach. in Marsabit based on CoBRA assessment
Table 14. Mapping CoBRA characteristics of resilience using quantitative indicators: Marsabit, Kenya
Marsabit CoBRA assessment findings
Top 3 community characteristics of Top 3 household characteristics of resilience resilience
• Peace and security • A household member has employment/wage labour • Education to secondary/tertiary • Household has a business or IGA(s) level • Household has a large herd (at least 200 shoats and 50 camels/ • Water for humans cattle)
Indicators Challenges/issues
Potential impact indicators The income level (X) would need to be agreed. The national figure • Percent of households with used to calculate the proportion in absolute poverty would likely be annual income above X too low. Measures such as the ‘livelihoods protection’ threshold used in HEA could be used. The figure represents total income required to meet all basic needs, achieve an acceptable minimal living standard and sustain livelihoods in the medium to longer term. This would be locally defined drawing from local cost data.
• Percent of households with off- This indicator is rarely if ever measured, but it could be extrapolated farm or non-farm income sources from national census or household budget surveys. Such data are • Percent of households with rarely collected more often than once per decade and rarely include multiple source of income a large sample size from sparsely populated drought-affected areas. activities Data are also unlikely to identify unsustainable or damaging income sources that need to be excluded, such as charcoal production or child labour.
• Percent of households with Pastoral households often understate livestock figures, which are livestock holding of more than hard to verify. Those with large herds may have low incomes, as 200 shoats the majority of their wealth is stored as livestock. Therefore it may still be difficult to understand the overlap between the suggested income and asset indicators mentioned to get a final figure on the percentage that has achieved ‘resilience’.
• Potential outcome indicators There are few areas with ongoing systematic data collection on peace »» Number of incidents of and security. Gaining consensus on a single proxy indicator such conflict as the one shown is likely to be difficult as there is no single widely »» Percent of households accepted indicator on peace and security. Consequently an index of enjoying peace and security several indicators may be required. Agencies would need to consider how and when these data should be collected as part of standard data collection processes.
»» Percent of households with at There is a heavy focus on the collection of data on primary least one member completing school attendance and completion rates in line with Millennium secondary school Development Goal and national targets. Census and household budget surveys may collect this information but not with sufficient frequency or sample size.
»» Percent of households with Data on water sources rarely capture information in this way. access to sufficient water all However, continuous access to water is likely to be a major factor year affecting household productivity.
30 Understanding Community Resilience 6. Conclusions and recommendations Children during the CoBRA assessment community awareness in Karamoja, Uganda
6.1 Conclusions are prominent differences between assessment areas in the priority given to Key findings these characteristics, depending on local ecological and socioeconomic conditions. Four full CoBRA assessments have been completed to date. Overall assessment • Resilient households are consistently findings have highlighted the following: described as those that have greater income and assets built through • Communities consistently highlight diverse sources. This is emerging several priority characteristics to as the potential basis for ‘universal’ describe a resilient community. The indicators of resilience. The most common characteristics that were common feature of resilient households highly prioritized in all the assessment noted through the FGDs and KIIs is locations include: their multiple income sources, which - Education: All children would be able tend to combine traditional on-farm to complete primary/secondary/ activities (pastoral and crop farming) tertiary school; and other on- or off-farm IGAs such - Water: The whole community would as wage labour and small business. have access to sufficient safe water The resilient households with single at all times of the year; and income sources have either a large - Peace and security: The whole herd or a large farm. Access to credit, community would enjoy continual savings groups, village savings and peace and security. lending associations, and training on alternative livelihoods were all stressed There are also some prominent differences as key resilience-building interventions in ranking of these characteristics to be promoted further. depending on local ecological and • Communities perceive their socioeconomic conditions. attainment levels of resilience characteristics as extremely low, Human health, productive livestock even in times of no crisis or hazard. herds/farms, access to markets and This underlines the very low levels of credit, diversified incomes and roads are development and high vulnerability also highly rated. These characteristics experienced in these locations, come up consistently across different and the very negative community livelihood groups (pastoral, agropastoral, perceptions of resilience. Drought peri-urban, etc.) and age and gender and other hazards merely exacerbate groups, irrespective of the number and these vulnerabilities. However, some level of services/interventions provided characteristics and locations are more in the assessment sites. However, there vulnerable to drought hazards than
Conclusions and recommendations 31 others. In Kajiado, for example, levels • The priority characteristics of resilience of development are relatively high, identified in a target community yet the people score themselves very through CoBRA can be used as a poorly on resilience. This is believed critical decision support tool to inform to relate to the following factors: policy, planning and programme/ (1) Regardless of their development project processes at different levels. status, communities always feel they The assessment highlights how these lack certain aspects of resilience, characteristics vary between different which they assign a low attainment locations, groups or contexts, which score relative to other factors; (2) can be used to help tailor resilience Kajiado suffers from a chronic lack programming. of some basic services; and (3) rural/ • The characteristics identified and pastoral groups were the focus of the prioritized by the communities can surveyed population in Kajiado, and in turn be used to identify keystone they experience low development indicators of resilience for more indicators compared with the wider systematic tracking. The keystone urban population in the towns just indicators can be both universal and outside Nairobi. contextually specific, leading to a • The most highly ranked characteristics potential resilience threshold. The of resilience do not often correspond priority resilience characteristics and with priority interventions by their attainment scores collected governments and their partners. through the initial CoBRA assessment When the most highly rated may serve as the baseline, which can characteristics affecting resilience also be used to identify keystone are compared with the portfolio of indicators that can be monitored over projects and interventions provided to time through repeat monitoring to see the same communities, there is often how communities perceive changes in a mismatch. For example, completing these characteristics. secondary and tertiary education is • The CoBRA assessment provides data regularly cited as a key characteristic of on the most highly rated interventions resilience, but support for expansion of or services in building local resilience, access to education is rarely prioritized which can also inform community- in programming in these areas. based programming and wider Equally, communities do not always policy and practice. As part of the prioritize characteristics that should assessment, communities and resilient be part of any investment in resilience. households are asked to prioritize For example, in feedback sessions, the interventions they believe best local stakeholders consistently cited build resilience. Unsurprisingly, the communities’ failure to prioritize interventions tend to mirror the priority natural resource management, which characteristics of resilience. Typically, stakeholders saw as a key gap. top-rated interventions are improved access to, quality of and efficiency Implications for community-based in education, water, human health, roads and markets. Importantly, such resilience programming interventions include both long-term The findings from the CoBRA assessment and short-term investments and factors can be used to support DRR and resilience that are not always considered a part programming in a number of very practical of DRR and/or wider development ways. These can be summarized in relation strategies. to the CoBRA model presented in Figure 1, as well as the four objectives of the For the DRR approach to help affected CoBRA assessment outlined previously: populations become truly resilient and
32 Understanding Community Resilience cope with future shocks and stresses Development’s Drought Disaster independently, the risk reduction Resilience and Sustainability Initiative) concept and strategy need to be and local climate-resilient development broadened. planning processes. Stakeholders • The study provides strong evidence need to review current data sets that interventions that demonstrably to determine if and how indicators increase and diversify income should identified through the CoBRA process be a core focus of any community- are already being collected. Data based programming. The feedback collection processes may need to be consultation strongly and consistently revised or standardized to ensure that supported the principle that resilience is appropriate information is collected a factor of income and assets. However, regularly on certain topics, such as it is critical to incorporate the sources of secondary school completion rates or income into this analysis. Diversification peace and security. of income can lead to increased reliance 2. Policy and planning frameworks need on unsustainable activities, such as to refocus on activities that build charcoal production, that may increase and diversify incomes and assets resilience in the short term but are not in the drylands. The finding on the suitable for the long term and could link between resilience and higher ultimately undermine resilience. In levels of income and assets derived general, resilient households undertake from multiple sources has important a combination of on-farm and off-farm implications for governments and IGAs. donors seeking to understand how to allocate resources to best build 6.2 Recommendations resilience. Increasing access to credit, strengthening savings groups and The findings from the CoBRA assessments improving commercial literacy and lead to a number of practical policy business skills are all key interventions recommendations to spur innovative and in this regard. entrepreneurial capacities of drought- 3. Activities that support the achievement affected dryland communities in the Horn of community-identified characteristics of Africa. of resilience need to be a key focus. The high ratings universally given to 1. CoBRA findings should be used to education, water, health and peace and identify a locally validated list of security for community resilience mean resilience indicators that government that these factors will drive change authorities and other stakeholders that enables households to develop can use to monitor progress and multiple sources of income and make impact. Including communities’ other positive changes. Education definitions of resilience is important can link people to broader income- not only as a participatory exercise, generation opportunities, especially but also as a means of understanding off-farm activities such as wage labour. local contextual factors that drive or Access to credit can facilitate various undermine resilience. IGAs and help diversify livelihoods. Water is key to both human and CoBRA findings should also support ecosystem health and economic the development of more streamlined development, contributing directly to monitoring frameworks for both livestock/agricultural production and national/regional (i.e., regional and productivity. These linkages are not county programming papers of the necessarily systematically addressed in Intergovernmental Authority for current policy and planning frameworks.
Conclusions and recommendations 33 4. Policy and planning need to increase community characteristics of resilience. their emphasis on raising awareness Further work needs to be done, ideally of natural resource management, with a wider group of stakeholders, particularly the critical role natural to identify proxy indicators (in each resources can play in the drylands in country or sublocation context) that the face of shocks and stresses. The would best statistically measure their CoBRA assessment demonstrates a attainment. In the Horn of Africa many key gap in communities’ awareness efforts are under way to develop and of the importance of natural resources monitor resilience frameworks, and and of maintaining the long-term few have successfully implemented health of local ecosystems that form any monitoring systems. CoBRA- the basis for their livelihoods. There is identified indicators could be used to a risk that people will be more prone feed into this process. to engage in activities that may build • Use CoBRA assessment findings to diverse incomes that are unsustainable identify and analyse existing data sets to (such as charcoal production). examine the links between household 5. CoBRA results stress the need for and community characteristics. Once coordinated and concerted action a set of resilience indicators has been among actors at different scales – identified, existing data sets can be including hardware and software, examined to ascertain the statistical long term and short term, and small relationship between household and and large investments. Some of the community level indicators. Effectively perceived priority characteristics – this would enable the verification such as peace and security, secondary/ and triangulation of the hypothesis tertiary education and roads – are emerging from CoBRA findings – that not systematically integrated into improvements in community-level development and humanitarian characteristics of resilience will result support. Ignoring these costly and in an increase in the proportion of long-term interventions and instead ‘resilient’ households as defined by focusing on less costly investment may indicators around income and asset lead to false economies. Communities wealth and diversity. consistently highlighted interventions • Continue to roll out CoBRA assessments that enhance access to markets, to improve the methodology. CoBRA savings and credit as highly beneficial is not designed to use statistically for enhancing community/household significant, quantitative data collection resilience, and these should be processes, for two reasons. First, the prioritized in the short term. Their cost would be prohibitive and would success, however, may be handicapped limit the ability of agencies to engage if larger scale interventions continually in the process. Second, resilience is fail to be developed in tandem. made up of a variety of quantitative Greater support for a coordinated and qualitative factors, and therefore approach will be required to ensure a purely quantitative assessment resilience in the long term. would not capture the full picture of characteristics of resilience. However, 6.3 Next steps it could be helpful to design a more systematic research sample in a given • Work to improve the mapping of location, using control and sample CoBRA-identified characteristics of communities, to test whether the resilience with standard indicators findings of CoBRA are corroborated from existing data sets. The CoBRA by a more statistically significant assessments to date have identified approach, such as by focusing on a a repeated set of household and single livelihood zone, sublocation
34 Understanding Community Resilience or parish type. In the latter case resilience characteristics have some an assessment of 35 to 40 FGDs value, particularly in highlighting the would likely provide a representative underlying lack of many services and sample of the population. It may also amenities and the areas that are most be possible to examine the impact or least affected by hazards. However, (retrospectively) of some forms of they should not be the focus of the interventions, particularly long-term CoBRA methodology, in part because multisectoral approaches such as attempting to establish a single score for community managed disaster/drought resilience in any absolute sense may be risk reduction when compared with impossible given the contextual nature control villages in the same area. of the concept. It is recommended Each additional CoBRA assessment to instead concentrate on measuring improves our understanding of the communities’ perceptions of trends methodology and how it can be in the key characteristics of resilience. improved or revised to become more An example would be the extent effective. In particular there is a need to which they feel that a particular to consider undertaking smaller scale characteristic, such as access to water, CoBRA assessments. has improved, declined or stayed the • Consider how often to undertake same over a specific period of time. repeat CoBRA assessments. The The pilot assessments revealed that CoBRA assessment is designed to communities consistently score their measure changes in resilience over achievement of priority characteristics time. It is not recommended that the very low, but it is not clear if this low assessment be repeated in the same score is higher or lower than previously. set of sublocations every year. Instead, Measuring trends would help to clarify it may be possible to monitor change whether communities see themselves using proxy indicators as they are on a positive or negative resilience developed. Given the longer term and pathway despite ongoing hazards and multisectoral nature of resilience and shocks. the interventions the CoBRA highlights as most valuable, it may be some years Therefore, once the group has before a repeat assessment would prepared a short list of top- offer significant additional information. ranked resilience statements, it is However, repeat monitoring is likely to recommended that participants be be particularly important after a crisis asked, “Over the last five years, has period, to measure perceived change your community’s attainment of this in response to a crisis. characteristic improved, worsened or stayed the same?” Answers should be 6.4 Suggested improvements to coded 1–5 as follows: the CoBRA methodology 1. Significantly better 2. Somewhat better The following recommended improvements 3. Stayed the same / no difference emerged from a comprehensive review of 4. Somewhat worse the results of the four CoBRA assessments 5. Significantly worse undertaken in 2013. They apply to use of the methodology in the Horn of Africa and This question would be asked before beyond. participants are asked to score their current levels of attainment from • Reduce the focus on resilience 0–10. Clearly this modification should ‘scores’. The scores provided by be tested and revised depending on communities on attainment of key results.
Conclusions and recommendations 35 • Improve the identification of key sites for FGDs and KIIs include a resilience indicators. The CoBRA balance of ‘high’ and ‘low’ intervention assessment reports completed to date areas. This mapping of interventions is make recommendations addressing done with local facilitators as part of indicators that agencies could consider the CoBRA training. The interventions collecting to track progress on key mapped to date have tended to resilience characteristics. If successive focus on ‘development’ oriented CoBRA assessments highlight similar services, such as health and education characteristics, particularly the facilities and roads. It is suggested to household characteristics, it might be include in intervention mapping some useful to work with other stakeholders humanitarian interventions, such as to identify standard proxy indicators commercial/slaughter destocking or for each. The challenges outlined in water tankering, that were provided Table 14 should be considered by during the last crisis. This might enable wider groups of stakeholders, ideally better analysis of changes in resilience led by the appropriate government scores between normal and crisis agency. periods. • Revise the graphics used to depict • Widen the list of interventions selected the characteristics identified by communities to include private by communities. The CoBRA and community sources. The long list implementation guidelines include of interventions given by focus group a set of standard photos or graphics participants tends to focus heavily on that attempt to represent the NGO and government services and characteristics they mention. These programmes. This is understandable are very useful in helping communities given that FGDs are generally with high proportions of illiterate undertaken by teams comprised of staff members to ‘bean score’ their priority from these organizations. However, our characteristics. However, some understanding of the value of support photos can be misleading or poorly and changes emerging from the understood by communities, so it private sector and within community may be easier to use simpler graphics sources is more limited. It is not clear to represent some characteristics, how facilitators can support groups such as community empowerment or to consider these more fully without diversified livelihoods. being too suggestive and thereby • Expand the selection of survey skewing the list. More consideration sites to include those receiving and testing of different facilitation humanitarian as well as developmental techniques should take place. interventions. The selected survey
36 Understanding Community Resilience Annex 1 Community-based Resilience Analysis Assessment Report Marsabit Central, Laisamis and Maikona districts Marsabit county, Kenya
Commissioned by UNDP Drylands Development Centre
Under the framework of the European Commission Directorate General for Humanitarian Aid and Civil Protection’s Drought Risk Reduction Action Plan
Marsabit Central, Laisamis and Maikona districts 37 Contents
1. Introduction...... 39 2. Approach...... 39 2.1 Characteristics of field site...... 39 2.2 Data collection ...... 42 2.3 Constraints and limitations of data collection process ...... 42 3. Findings...... 43 3.1 Main hazards or shocks ...... 43 3.2 Characteristics of a resilient community...... 43 3.3 Extent to which the community has achieved resilience...... 48 3.4 What a resilient household looks like...... 51 3.5 Interventions that contributed to household resilience...... 52 3.6 How key informants achieved resilience ...... 54 4. Summary of feedback from local consultations...... 56 5. Conclusions and recommendations...... 58 5.1 Conclusions...... 58 5.2 Recommendations...... 59
38 Understanding Community Resilience 1. Introduction This report outlines findings of CoBRA Marsabit assessment. It also incorporates The comprehensive Community-Based the key feedback and consolidated Resilience Analysis (CoBRA) assessment inputs generated at review and validation was undertaken in three districts of Marsabit workshops for the draft assessment report. county, Kenya, from 3 to 15 June 2013. These were conducted for community The exercise was carried out with financial representatives on 3 September and for support from the European Commission local government and non-governmental Directorate General for Humanitarian Aid technical stakeholders working in the area and Civil Protection under the framework on 4 September. of its Drought Risk Reduction Action Plan. This assessment builds on an initial field A detailed explanation of the conceptual trial of the draft CoBRA tool undertaken in framework underpinning the methodology late 2012. is contained in the CoBRA Conceptual Framework and Methodology document.1 The multi-agency assessment was jointly led by the United Nations Development 2. Approach Programme (UNDP) Drylands Development Centre (DDC) and the Government of Kenya’s National Drought Management 2.1 Characteristics of field site Authority (NDMA). Logistical and practical The field assessment was conducted in support was provided by Food for the three districts (Marsabit Central, Laisamis Hungry’s (FH) Kenya office. A wide range of and Maikona) of the newly formed Marsabit international and local NGOs operating in county, in northern Kenya, bordering the area also participated in the assessment Ethiopia. The population of these randomly by providing local staff as facilitators to sampled districts represents 61 percent undertake CoBRA training and fieldwork. of the total county population, which is A list of agencies participating in CoBRA estimated at 291,166.2 Within these 3 training and/or field data collection is sampled sub-districts, 17 administrative included in Appendix 1. sampling locations were selected for the assessment (Figure 1), reflecting pastoral, CoBRA assessment has four broad agropastoral and urban/peri-urban objectives: livelihood zones. The sampling frame listing the locations where focus group 1. Identify the priority characteristics of discussions (FGDs) and key informant resilience for a target community; interviews (KIIs) were undertaken is shown 2. Assess the communities’ achievement in Table 1. of these characteristics at the time of the assessment and during the last Marsabit was chosen for CoBRA crisis/disaster; assessment because it is one of the most 3. Identify the characteristics and arid and drought-affected areas of Kenya.3 strategies of resilient households; and It remains one of the poorest counties 4. Identify the most highly rated in the nation, with 76 percent of the interventions or services in building population living under the poverty line, local resilience. according to the Kenya National Bureau of
1 www.youblisher.com/p/784724-Community-Based-Resilience-Analysis-CoBRA-Conceptual-Framework-and-Methodology/ 2 Kenya National Bureau of Statistics, Population and Housing Census 2009. 3 Kenya Ministry of State for Development of Northern Kenya and other Arid Lands, National Policy for the Sustainable Development of Northern Kenya and Other Arid Lands, 2012.
Marsabit Central, Laisamis and Maikona districts 39 Figure 1. Surveyed locations in Marsabit county
Surveyed locations
Statistics.4 Education and health indicators are predominantly semi- or fully mobile are unsatisfactory, and the county has pastoralists with some agropastoralists limited infrastructure. Substantial portions on the mountain around Marsabit town of the population are either food-insecure and the hilly areas of Moyale. The first or chronically hungry. Just 26 percent of tarmac road, to Merille, in the far south of the people can read and write.5 the county, was completed only in the last couple of years. With a surface area of 70,961 square kilometres, Marsabit is a sparsely In common with much of northern Kenya, populated county with an average density Marsabit has experienced repeated of four persons per square kilometre. It is droughts in recent years. The most predominantly inhabited by the pastoral recent was in 2011, when the Kenyan communities of the Rendille, Gabra and government6 classified Marsabit’s status Borana. The population is concentrated in as ‘emergency’, with 77 percent of the the higher and slightly less arid areas around population identified as needing food Marsabit town and Moyale. The people aid. Drought impact is exacerbated by
4 Kenya National Bureau of Statistics and Society for International Development, Exploring Kenya’s Inequality: Pulling Apart or Pooling Together? (Nairobi, Ascent Limited, 2013). 5 Kenya Commission of Revenue Allocation, County Fact Sheets, December 2011. 6 Kenya Food Security Steering Group, 2011/2012 Short Rains Season Assessment Report, January 2012.
40 Understanding Community Resilience Table 1. FGDs and KIIs undertaken for CoBRA Marsabit assessment
No. of No. of District Location Population Livelihood zone FGD composition FGDs KIIs
Females (all ages); Dakabaricha 4,240 Peri-urban 2 1 mixed youth
Male elders; Hula Hula 2,880 Agropastoral 2 4 mixed youth;
Females (all ages); Karare 3,365 Agropastoral 2 2 mixed youth
Marsabit Females (all ages); Songa 3,079 Peri-urban 2 2 Central mixed youth
Male elders; Dirib Gombo 4,220 Agropastoral 2 2 mixed youth;
Male elders; Sagante 4,737 Agropastoral 2 2 female elders
Male elders; Jaldesa 3,452 Agropastoral 2 2 females (all ages)
Male elders; Hurri Hills 4,008 Agropastoral 3 2 females (all ages)
Male elders; Pastoral/ Maikona 7,253 3 females (all ages); 3 Peri-urban mixed youth
Male elders; Maikona Bubisa 4,811 Pastoral 2 2 female elders
Male elders; Shura 1,204 Pastoral 2 2 females (all ages)
Male elders; Turbi 4,321 Pastoral 3 females (all ages); 3 mixed youth
Male elders; Pastoral/ female elders; Korr 13,012 4 4 Peri-urban males (all ages); females (all ages)
Males (all ages); Ngurunit 8,293 Pastoral 2 2 female youth
Pastoral/ Male elders; Laisamis 6,423 2 2 Laisamis Peri-urban mixed youth
Male elders; Logologo 5,144 Pastoral 2 2 Females (all ages)
Male elders; Lontolio 2,423 Pastoral 2 2 female youth
Females (all ages); Merille 4,263 Pastoral 2 2 mixed youth
Totals 87,128 41 41
Marsabit Central, Laisamis and Maikona districts 41 chronic inter-clan and ethnic insecurity place along with 41 KIIs. Table 1 summarizes and conflict. The conflict in Marsabit is the number and locations of FGDs and KIIs partly linked to scarce resources such as undertaken for this assessment. pasture for grazing and water, but it is also attributed to ethnic, tribal and political FGD and KII data were captured using tensions between tribes and groups, all preconfigured Excel spreadsheets. For exacerbated by the impacts of frequent analysis, the Excel data were reformatted drought. into statistical analysis software such as SPSS Statistics and recoded when required 2.2 Data collection (such as by classifying some categories of variables together). Complete details of the methodology used to undertake a CoBRA assessment are 2.3 Constraints and limitations of included in the main CoBRA assessment data collection process report7 and implementation guidelines. Data were collected using qualitative Dealing with raised expectations methods, in the form of FGDs and KIIs. Initially, most focus groups thought the teams of facilitators were there to Field work was undertaken by a team of undertake some form of needs assessment 12 facilitators and 3 supervisors. The team that might result in a benefit or intervention was divided into three groups of five for the community. In order to manage people each, with a supervisor and two inaccurate assumptions, expectations and pairs of facilitators. Each group covered biases, a community mobilization exercise one of the three districts surveyed. Each was undertaken in most locations prior to FGD was facilitated by two staff from the the FGDs to explain the purpose of CoBRA participating government departments assessment and clarify that no direct and NGOs. The facilitators and interventions in their community were supervisors, who were familiar with the likely as part of it. Furthermore, an effort local context, jointly selected the locations was made to maintain the multi-agency and compositions of focus groups, nature of the assessment by combining using statistical data and criteria such as facilitators from different agencies and livelihood zones and level of interventions. sectors. This arrangement was particularly Whenever possible, separate FGDs were helpful in the data collection process as organized for men, women and youth many communities were not used to being (aged 15–30, both sexes) to ensure that interviewed by representatives of both views on resilience from different gender government agencies and NGOs. and age groups would be adequately captured in the discussions. Each group Scoring the attainment of resilience had around 15 participants. characteristics Following the identification and ranking of In each focus group location, one to four resilience characteristics, FGD participants KIIs were undertaken with households were asked to score how far they had perceived to be resilient, as identified attained the top-ranked characteristics. by focus group participants during their Before the field assessment, facilitators discussions. In some cases, the households were trained to undertake this exercise were also identified as resilient through by asking the participants to score on a conversations with local chiefs or other scale of 0–10, with 10 being the perfect senior residents. A total of 41 FGDs took or ideal situation and 0 an absolute lack of
7 UNDP-DDC, Understanding Community Resilience: Findings from Community-Based Resilience Analysis Assessments – Marsabit, Turkana and Kajiado counties, Kenya, and Karamoja sub-region, Uganda (Nairobi, UNDP-DDC, 2014 [in press]).
42 Understanding Community Resilience attainment of resilience characteristics. The • How did key informants achieve and purpose was to adopt a uniform approach maintain resilience? for measuring the attainment level of the diverse set of characteristics, including 3.1 Main hazards or shocks the non-household-related characteristics, such as peace and security, and the more The main hazard reported in all the FGDs qualitative characteristics, such as health was drought. Communities viewed drought care services. Some lack of consistency hazard as the most significant contributor was found in scoring, which had to be to livestock losses and the single most corrected. Where data were unclear, they important factor limiting their resilience were excluded to ensure that only accurate capacity. Communities also reported the data were used in the final analysis. most recent drought of 2010–2011 as the main ‘crisis’ period to be referred to for Reluctance to being identified as CoBRA assessment. It was the most severe, ‘resilient’ although some of the past droughts lasted In some locations, individuals identified by for a longer period than the 2010–11 FGD participants as ‘resilient’ households drought. The impacts of the 2010–2011 refused to accept this designation. They drought were exacerbated by extremely were often concerned that accepting high food prices, reduced coping capacity might lead to high expectations for peer and a slow and limited humanitarian support from other community members or response. Communities further indicated result in their exclusion from development/ that conditions had improved since the last humanitarian programmes or assistance. drought. The current period was perceived The assessment team took this sensitivity principally as a ”normal” period, with a few into consideration and strived to identify communities stating that it was a “good” perceived ‘disaster-resilient households’ or “above normal” period. (This process in a discrete manner, while explaining the involved comparing a current period – here purpose and implications of the exercise represented by the preceding 12 months fully to the identified individuals and focus – to the long-term climatic normal for the group participants. specific location under assessment.)
3. Findings Other hazards or crises that were reported included political and resource-based This section reports on the summarized conflict, largely clan-based raiding of findings from CoBRA field work conducted livestock among the three major ethnic during the month of June 2013 in Marsabit. groups. Competition over scarce grazing Specifically, the findings are presented fields, water resources and pasture has according to the following categories: escalated inter-ethnic animosity, often resulting in violent conflicts, which are • What are the main hazards or shocks predatory in nature and very destructive. facing the communities? To a limited extent, the communities also • What are the characteristics of a reported livestock and human diseases as resilient community? hazards. • To what extent has the community achieved resilience characteristics? 3.2 Characteristics of a resilient • What does a resilient household look community like? • What interventions contributed to Focus group participants were asked household resilience? to describe what they viewed as the • What additional interventions would characteristics of a resilient community. The best build resilience? data are first presented for the entire set of
Marsabit Central, Laisamis and Maikona districts 43 respondents, to give an overall picture of Table 2 lists the most highly ranked the most highly rated characteristics. This characteristics used to describe a resilient is followed by an analysis by category of community within each of the five SLF respondent, gender/age, livelihood group categories according to the ‘bean scores’. and level of intervention in the community, (Note that many more characteristics/ which are used to disaggregate findings statements were included in the ranking, and identify differences across groups. but were given low scores and hence are not reported here.) The full list of Analysis – all respondents statements with scores is provided in Focus group participants were asked Appendix 2. to identify and rank statements used to describe a resilient community. Each Figure 2 shows the highest scored member was given six beans to rank the characteristics used to describe a resilient three most significant statements in terms community by all focus group participants of priority for building resilience, with three in the three districts. Figure 3 presents the beans for the most significant statement, total bean score under each of the five SLF two for the second most significant and categories. one for the third. The bean scores were then totalled for each statement. For Peace and security received the highest ease of comparison, the statements were bean score followed by education and grouped into the five sustainable livelihood water for human use. In terms of the framework (SLF) categories.8 SLF categories, statements addressing
Table 2. Community ranking of resilience characteristics by SLF category
Top-ranking resilience characteristics SLF category Total bean score* (according to ‘bean scores’)
Financial Productive farms (114) 554 Diversified income-generating activities (IGAs) (102) Livestock herds (84) Pasture and fodder (73) Employment (68) Access to credit (61)
Human Education (608) 1,032 Health care for humans (232) Food security (145)
Natural Natural resource management (NRM) (83) 105
Physical Water for humans (598) 1,657 Telecommunications (247) Access to markets (232) Roads (149) Water for livestock (146) Shelter (103)
Social Peace and security (792) 829
* Total bean score does not match the sum of the bean scores in the middle column because the middle column reflects only the top-ranking characteristics while the total in the right column reflects all the characteristics discussed in each category.
8 The sustainable livelihood framework presents the main factors that affect people’s livelihoods and typical relationships between them. It identifies five core asset categories or types of capital upon which livelihoods are built: financial, human, natural, physical and social. For further details, please refer to: UK Department for International Development (DFID), Sustainable Livelihoods Guidance Sheets (London, DFID, 1999).
44 Understanding Community Resilience Figure 2. Top-ranking resilience characteristics – all respondents
900 800
700
600
500
400 300
200
100 0
NRM Roads Shelter
Education
Employment Food security Livestock herds Diversified IGAs Access to credit Productive farms Water for humans Access to markets Peace and security Water for livestock Pasture and fodder Telecommunications
Health care for humans
Figure 3. Top-ranking resilience characteristics by SLF category – all respondents
1800 Others Peace and security 1600 Shelter Water for livestock 1400 Roads Access to markets 1200 Telecommunications Water for humans 1000 NRM Food security 800 Health care for humans Education 600 Access to credit
400 Employment Pasture and fodder
200 Livestock herds Diversified IGAs
0 Productive farms Financial Human Natural Physical Social physical capital were ranked most highly, Analysis by gender and age followed by human, social and financial The priority resilience statements were scores. Natural capacity received the also analysed by gender and age group, lowest number of statements and overall through groups comprised of women, men bean scores. and youth (aged 15–30, both sexes). The
Marsabit Central, Laisamis and Maikona districts 45 Figure 4. Top-ranking resilience categories by gender/age group 100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Women Men Youth
bean score allocated to each of the five • Women placed the greatest emphasis SLF categories was taken as a percentage on physical characteristics of resilience of the total bean score. The results are (for example, water) reflecting the fact represented in Figure 4 to demonstrate that women are the ones responsible the differing priorities that men, women for collecting water for household use. and youth place on resilience statements. • Men placed greater emphasis on financial statements of resilience Further to this, the most highly ranked (for example, agricultural inputs and resilience characteristics by gender/age pasture and fodder for livestock) group are presented in Table 3. compared to women and youth. • Youth placed far greater value on human The data suggest the following points: capital, which includes statements on education and health care. They also • All three groups consistently ranked placed far less emphasis on physical education, peace and security, and capital compared to men and women. water for human use as the most important characteristics of resilience. Analysis by livelihood group Resilience statements were also analysed Table 3. Top-ranking resilience characteristics for a range of livelihood groups: by gender/age group* agropastoral, pastoral and peri-urban. Figure 5 illustrates the differences in bean Gender/ Top 3 resilience Total score allocation according to the five SLF age group characteristics score** categories among the three livelihood Women Peace and security (291) 1,589 groups. Education (266) Water for humans (177) Further to this, the most highly ranked Men Peace and security (339) 1,598 resilience characteristics by livelihood Water for humans (257) group are presented in Table 4. Education (183) The data suggest the following points: Youth Education (119) 490 Peace and security (82) Water for humans (47) • Agropastoral groups placed significant emphasis on physical characteristics * Results from the mixed focus groups that do not fit the categories of male, female and youth were excluded from the analysis. of resilience, with water and access **Total score does not match the sum of the scores in the middle to markets being two of the most column because the middle column reflects only the top-ranking commonly cited characteristics in this characteristics while the total in the right column reflects all the characteristics discussed in each category. category.
46 Understanding Community Resilience Figure 5. Top-ranking resilience categories by livelihood group
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Agropastoral Pastoral Peri-urban
Table 4. Top-ranking resilience characteristics urban groups already have the greatest by livelihood group access to physical infrastructure. • All three groups consistently cited Livelihood Top 3 resilience Total education, peace and security, and group characteristics score* water for human use as the most Agropastoral Peace and security (371) 1,592 important characteristics of resilience Water for humans (311) overall. Education (162) Analysis by intervention level Pastoral Education (366) 2,085 The three districts assessed comprise Peace and security (302) Water for humans (221) between 6 and 11 administrative locations. Consultation with field teams, the National Peri-urban Peace and security (119) 500 Drought Management Authority and other Education (80) government line departments was used to Water for humans (65) map accessibility to and presence of the * Total score does not match the sum of the scores in the middle following basic services and interventions column because the middle column reflects only the top-ranking in each location: characteristics while the right column reflects all the characteristics discussed in each category. • Number and level of health facilities; • Pastoral households also ranked • Tarmac road; physical characteristics very highly, • Other main road; but they also put strong emphasis on • Mobile phone coverage; human characteristics, compared to • Well-functioning livestock market; agropastoral households. This may • Savings and credit programmes; reflect the generally lower provision • Number and level of education of education and health services in the facilities; more remote pastoral areas. • Banks; and • Peri-urban households put less • Bank agents/M-Pesa.9 emphasis on physical characteristics, instead ranking social and human Interventions that are universally provided characteristics most highly when in all the locations, such as food aid and compared to the other two groups. water interventions, or provided only at Again this may reflect the fact that peri- very low scale, such as for less than 500
9 M-Pesa is a mobile phone-based money transfer and microfinancing service offered by Safaricom and Vodacom in Kenya and Tanzania. (M stands for mobile and pesa means money in Kiswahili.)
Marsabit Central, Laisamis and Maikona districts 47 beneficiaries, were excluded from this Table 5. Top-ranking resilience characteristics mapping exercise. The locations were then by intervention level divided into three even groups: Intervention Top 3 resilience Total level characteristics score* • The bottom third, representing low intervention areas, which have up High Peace and security (233) 1,199 to 8 categories of interventions Education (165) Water for humans (128) implemented in their area; • The middle third, representing Medium Peace and security (316) 1,583 medium intervention areas, which have Education (226) Water for humans (225) 9 or 10 categories of interventions implemented in their area; and Low Water for humans (245) 1,395 • The top third, representing high Peace and security (242) intervention areas, which have Education (217) between 11 and 14 categories of * Total score does not match the sum of the scores in the middle column because the middle column reflects only the top-ranking interventions implemented in their characteristics while the right column reflects all the characteristics area. discussed in each category.
Figure 6 shows the differences in bean example agricultural inputs, pasture score allocation according to the five and fodder, and employment. SLF categories among the three location • All three groups consistently cited groups with different service/intervention education, peace and security, and levels. water for human use as the most important characteristics of resilience. Further to this, the most highly ranked resilience statements by intervention level 3.3 Extent to which the community are presented in Table 5. has achieved resilience The data suggest the following points: Focus group participants were asked to score the extent to which they had • All three intervention groups achieved their priority characteristics consistently placed the greatest of resilience. They were asked to score emphasis on physical characteristics of each statement twice, first for the current resilience. period (agreed to be a normal period) • Notably, the high intervention areas and second for the last significant crisis also put more weight on financial period (almost universally identified as the characteristics of resilience, for drought period of 2010–2011).
Figure 6. Top-ranking resilience categories by level of intervention 100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% High Medium Low
48 Understanding Community Resilience As with the previous section, the findings characteristics in the current period, and are presented first for all respondents and the red line represents the last crisis period. are then disaggregated by subgroups. The scores are ranked on a scale from 0 This section does not include analysis by to 10, with 10 being perfect attainment of gender because focus groups were asked that characteristic (for example, the entire to rank the attainment of the resilience community has access to sufficient, good characteristic statements for the entire quality water at all times in the current/ community, and therefore any differences crisis period), and 0 being no attainment between men and women would be based (no one in the community has access to on perceptions. sufficient, good quality water at all times in the current/crisis period). Table 6 presents the scores by SLF category for the top-ranked characteristics Social characteristics of resilience have of resilience. Figure 7 illustrates the the highest score (6.1). In other words, spider diagram of perceived resilience communities ranked social characteristics characteristics attainment rates per each as the area in which they had attained the SLF category. The outer ring represents a greatest degree of resilience. This category perfect or ideal score for all statements in was dominated by responses related to that SLF category. The blue band shows peace and security, and the high ranking communities’ average attainment of those is because these communities have gone
Table 6. Community attainment of resilience by SLF category – top-ranked characteristics
Current period Crisis period SLF category Top-ranked characteristics score score
Overall average 3.7 2.4
Financial Productive farms 2.0 0.0
Pasture and fodder 6.7 2.1
Employment 3.0 1.8
Access to credit 1.0 0.5
Diversified IGAs 3.6 2.6
Human Education 3.0 1.8
Health care for humans 3.5 3.1
Food security 5.1 2.5
Natural Natural resource management 3.7 3.1
Physical Water for humans 4.0 1.8
Access to markets 3.7 2.2
Telecommunications 2.1 1.1
Roads 2.5 1.7
Water for livestock 5.9 2.9
Shelter 2.4 1.0
Social Peace and security 6.1 4.1
Marsabit Central, Laisamis and Maikona districts 49 Figure 7. Community attainment of resilience per SLF category – all respondents
Physical 10
8
6
4 Social Human 2
0 Current year rank Crisis year rank Full score
Natural Financial
Note: Possible score ranges from 0 to 10.
through a recent period of much stronger Table 7. Aggregate resilience scores peace and security. However, these scores could quickly deteriorate if the current Current Crisis year security situation changes. The remainder year rank rank of the categories were ranked similarly in All groups 3.7 2.4 current and crisis years, suggesting that, in a normal year, community members Livelihood group rank their attainment of characteristics of Agropastoral 2.6 1.6 resilience below 4 out of 10. Pastoral 4.1 2.8 Other characteristics that were highly ranked for attainment in normal times Peri-urban 3.9 1.9 include pasture and fodder (6.7), food Intervention level security (5.1) and water for livestock (5.9). Characteristics ranked as low attainment High 5.1 3.8 include financial characteristics, namely productive farms (2.0) and access Medium 3.5 2.1 to credit (1.0), and several physical Low 3.3 2.3 characteristics – roads (2.5), shelter (2.4) and telecommunications (2.1). Note: Maximum possible score is 10
Table 7 aggregates the scores given to each as these scores represent community of the five SLF categories to provide an perceptions around attainment, and thus overall ‘resilience score’. While this clearly could overstate or understate reality. masks differences between SLF categories, it is useful to provide an overall indicator Overall, the average level of attainment of of where communities see themselves, for resilience characteristics is 3.7 out of 10 at comparison across groups. Clearly, figures present, as opposed to 2.4 out of 10 during should be viewed with some caution, the crisis period in the three districts.
50 Understanding Community Resilience Results by livelihood group show some 3.4 What a resilient household important differences: looks like Focus group participants were asked to • Pastoral and peri-urban groups describe the characteristics of households ranked their current overall resilience that are more resilient compared to others, in a normal period much more highly i.e. the households that have already than agropastoral groups – 4.1/3.9 attained many (or all) of the resilience compared with 2.6 out of 10. Laisamis characteristics prioritized. The top three district is considered more resilient characteristics of a resilient household, than the other districts, and it had cited consistently by focus groups, consist a sample that consisted mostly of of the following: pastoralists. It also benefits from greater access to services, such as • The household has a business or IGA a tarmac road, functioning livestock (41 out of 41 FGDs); markets and access to savings and • A member of the household has credit services. employment or wage labour (39 out of • Despite higher scores in the current 41 FGDs); and period, peri-urban groups appeared • The household has a large herd (30 to experience a much sharper decline out of 41 FGDs). in the crisis period. Possibly this is because relief efforts during crises Typically a large herd was described as (such as food aid and water tankering) at least 200 shoats and 50 cattle and/or often focus on rural populations, so 50 camels. The fact that groups always urban groups do not benefit from the highlighted a mix of livestock implies same support. Further analysis may be that both size and diversity of the herd required on how drought and other is important. Interestingly, while these climate hazards affect urban/peri- statements were fairly consistently cited urban livelihoods. across gender/age and livelihood groups, • When the scores are broken down high intervention areas were much more by SLF category, agropastoral groups likely to cite business/employment, with estimated their attainment of social only a few citing large herds. characteristics most highly – likely because of the recent significant A few other characteristics of resilient improvement in peace and security, households were mentioned but which makes up the bulk of this significantly less often: category. • The household has educated members When the scores are disaggregated by the (and where this was mentioned it level of services and interventions, high was often in relation to the ability intervention areas ranked their level of of the educated person to obtain attainment of resilience characteristics more employment); highly than medium and low intervention • The household uses advanced areas. This could suggest that provision of agricultural techniques/tools/inputs; basic services and interventions contribute and to resilience. Equally, it could suggest • The household has good planning, that interventions are targeted to high- management and organization. resilience areas. Longitudinal data would have to be studied to draw inferences on Focus groups were then asked whether the causality of levels of intervention on the number of resilient households is resilience. increasing, decreasing or staying the same. As Table 8 shows, out of 41 FGDs, the
Marsabit Central, Laisamis and Maikona districts 51 Table 8. Is resilience increasing, decreasing or staying the same?
Increasing Decreasing Staying the same
All respondents 60% 25% 17%
Gender/age
Women 69% 25% 6%
Men 56% 25% 19%
Youth 60% 20% 20%
Livelihood group
Agropastoral 38% 38% 25%
Pastoral 76% 19% 5%
Peri-urban 60% 0% 40%
Intervention level
High 17% 50% 33%
Medium 81% 13% 6%
Low 71% 14% 14%
majority – 60 percent – said that resilience which are typically in low intervention is increasing. These findings were similar locations but have a positive outlook due across gender/age groups. to increasing herd sizes. By contrast, most high intervention areas are urban, peri- When the data were disaggregated by urban and agropastoral, which in most livelihood group, agropastoralists had a cases (particularly for the latter) see their more pessimistic view, with equal numbers resilience decreasing. saying that resilience was increasing and decreasing. Pastoralists, on the other 3.5 Interventions that contributed hand, were much more likely to say that to household resilience resilience was increasing (75 percent). In general, the positive outlook exhibited Communities were asked to list all the by pastoralists was largely attributed to a services and interventions they had relatively good build-up of livestock herds benefited from in the last two to five due to availability of pasture resources years. A reasonably wide range of resulting from normal or above normal interventions was mentioned, including seasonal rainfall since the last major water, education, livestock restocking, drought of 2010–2011, combined with the cash transfer, health care services, mobile fact that more and more of their sons and phone coverage, agricultural inputs, roads daughters were completing schooling and and other livestock support. From this long finding gainful employment. list, each community (through FGDs) was asked to identify jointly the three current Low and medium intervention groups were or previous interventions that had been much more likely to say that resilience most beneficial in building their resilience was increasing, whereas high intervention and why. Table 9 shows that interventions groups were more heavily weighted relating to water, education and restocking towards decreasing resilience (50 percent). were prioritized most regularly. This likely represents the pastoral groups,
52 Understanding Community Resilience Table 9. Ranking of resilience-building interventions
Currently/recently Further or future Total Type of intervention provided provision score
Water Water source improvement or improved storage 27 21 48 capacity
Education Bursaries, scholarships or construction/ 25 13 38 refurbishment of school facilities including boarding facilities
Restocking Programmes restocking livestock particularly 19 9 28 with drought-resilient breeds or animals such as camels
Cash transfers Transfer of resources to vulnerable people, such as the chronically food insecure (e.g. hunger 10 8 18 safety net programmes) and children (e.g. child sponsorship programmes)
Health Improvements to health services, staffing or 5 15 20 facilities
Inputs to productive farms Irrigation, greenhouses, oxen, agricultural 7 9 16 extension, etc.
Mobile phone coverage 0 15 15
Roads 4 8 12
Access to credit or other forms of business 6 6 12 support
Other livestock support Livestock markets, health services, fodder for 5 5 10 production, etc.
Groups were also asked to list the three prioritized by all livelihood groups, further or additional interventions they particularly as a means for improving food felt would best build their resilience. security and livelihoods. Interventions that Many communities restated interventions improved water storage as well as supply mentioned in the first list, with the were all seen as important for expanding justification that the current provision or the production of crops and livestock. scale of intervention was too limited and Water interventions were also justified for should be expanded. The highest scoring time savings and health reasons. interventions are summarized in Table 9. Education was seen as a benefit in itself The table shows the repeated and clear since it would lead to improved life priority given to water and education chances, such as employment for children. interventions. These interventions reflect Scholarships and bursaries, essential for the high ranking given to these factors continuing education to secondary and as characteristics of resilience by all tertiary levels, were mentioned more focus groups. Water interventions were frequently than education facilities. This
Marsabit Central, Laisamis and Maikona districts 53 again relates directly to communities’ • Ability to cope with recent shocks and ranking of secondary and tertiary education hazards; and as a key characteristic of resilience. • Priority interventions recommended by resilient households. Restocking was also rated highly. This is unsurprising given that the majority Composition and characteristics of of focus group participants were from resilient households pastoral or agropastoral areas, where The KII record sheet records the age, livestock numbers are a critical factor in gender, education level and economic household wealth. The extent to which activity of all members of the resilient restocking results in resilient or viable herd household interviewed. The resilient sizes was not explored. households varied in size from 1 to 11 members, with an average of 6 members, Other livestock interventions were also which is typical for the area. In Marsabit rated highly in pastoral areas, as were Central and Laisamis the vast majority of agricultural inputs in agropastoral areas. households interviewed (86 percent) had All these interventions were justified members who had completed primary on the basis that they would improve education, and over half (55 percent) had productive capacity, thereby increasing members who had completed secondary income levels and/or food security. Cash or tertiary education. By contrast, in transfers were justified on the same basis. Maikona only half (50 percent) had Mobile phone coverage was seen as very completed primary education, and only important in locations with limited or no 33 percent had completed secondary or coverage, particularly by pastoral groups, above. This probably reflects Maikona’s possibly because mobile communications more remote location and lower access allow pastoralists to get data on livestock to secondary education. Government market prices. statistics indicate that about 68 percent of the population in Marsabit county has no When disaggregated by livelihood group, education. Only 26 percent of the county’s agropastoralists placed the greatest population has a primary education and emphasis on water storage, agricultural 6 percent has completed secondary or interventions and education facilities. higher education.10 From these figures, it By contrast, pastoralists placed a strong would seem that resilient households are focus on health facilities and mobile significantly more highly educated than communications. the average.
3.6 How key informants achieved All 41 KII respondents had household members engaged in one or more of the resilience following: A total of 41 key informant interviews were undertaken with members of households • Wage employment or casual labour; identified as resilient. Between one and • Business or petty trade; four interviews were undertaken in each • Raising livestock; and/or sampled location. KIIs examined the • Agricultural production. following four areas: Most households interviewed (33 out of • Composition and characteristics of the 41, or 80 percent) had a livestock and/ household; or agricultural income source; however, • Pathways to resilience; only 7 resilient households (17 percent)
10 Kenya National Bureau of Statistics and Society for International Development, op. cit.
54 Understanding Community Resilience relied exclusively on livestock. These were explained that income from either of predominantly pastoralists with large these sources had been saved and used herds, the majority located in the pastoral to start/grow businesses, grow livestock locations of Maikona and Laisaimis. Only herds or invest in agricultural production. two resilient households in the locations Typically this included the purchase and relied on pastoral and some agricultural construction of water storage and irrigation production. No resilient households systems. Several households mentioned survived on agricultural production alone. poultry production as a further source of agricultural income. The business activities Overall, the vast majority of resilient described were mostly kiosks or petty households (31 out of 41, or 75 percent) trading of food, fuel, phone credits, soap reported multiple income sources, powder or other household items. Some including members with wage labour respondents ran bars or lodges. Animal and/or a business interest. Only three trading was repeatedly mentioned by households (7 percent) survived solely pastoral households as both a source of on income from wages (1/41) or a single income and a coping strategy. business income (2/41). Most resilient households with multiple income sources In addition to diversified income sources, (26/31, or 83 percent) also maintained a nine households attributed their resilience pastoral or agricultural income source. to being organized, carefully managing One household was receiving a pension household expenditures and income, and from previous employment. Consequently refraining from wasteful expenditures such the diversification of income sources as alcohol. Twelve mentioned use of their is clearly a key strategy for resilient own savings (often as part of self-help/ households, with most retaining their credit groups) and three cited support agricultural or pastoral activities as well as from family or friends in terms of loans or wage and/or business incomes. According remittances. One household reported that to government statistics, 10 percent of the the camel provided by an NGO enabled county’s population is working for pay.11 them to become resilient.
Most wage earners were government Ability to cope with recent shocks and employees, civil servants, teachers or hazards military personnel. Other employment Almost all key informants referred to the mentioned included NGO workers, a 2010–2011 drought as the last major driver and a religious minister. Several hazard. Most indicated they were better households relied on casual or temporary placed to cope with it than others due to labour, but no details were recorded. No the additional income sources outlined private sector employment was mentioned, above. Wage earners and most businesses reflecting the dearth of any significant were not affected by the drought. private sector employers in the area. Critically, 17 (65 percent) of the households Pathways to resilience with income from livestock stated that All respondents cited their multiple they sold animals and used the proceeds sources of income as the reason for their to support the remaining livestock (to resilience. Those with multiple incomes access water, fodder, health care) and pay (31 respondents) cited the capacity to for the household’s basic needs such as rely on other sources if one was reduced food and school fees. Moving livestock to or affected by drought. Households areas of good grazing and water was also with a wage earner or business regularly mentioned.
11 Ibid.
Marsabit Central, Laisamis and Maikona districts 55 Two households with agricultural income water for livestock, crop production sources cited the use of irrigation and/ and business uses. or market gardens, which enabled them • Farm production: Interventions to continue producing and selling crops to improve farm production were during drought. Many mentioned that highly rated (10 respondents), these income sources had enabled them particularly in agropastoral locations to amass savings that were used to get in Marsabit Central. Many of the their households through the drought interventions relate to expanded use period. Several households mentioned of irrigation and market gardens. their involvement in savings and credit or The use/distribution of oxen and self-help groups. Others mentioned loans more agricultural extension services from friends and family as well as credit were also cited. Other interventions from traders as factors enabling them to mentioned included peace and cope. security, educational support and improved health facilities and services. Some households mentioned interventions as important coping factors. These This list is somewhat similar to interventions included cash/food for work schemes, cash mentioned by the focus group participants. transfers (such as the Hunger Safety Net However, it is notable that resilient Programme), training from NGOs, peace households did not mention interventions forums and water tanks. related to humanitarian response, such as restocking and cash transfer programmes. Priority interventions Compared to the focus group participants, Key informants were asked for the three they put greater emphasis on business most important interventions to improve skills and savings and credit, most likely their communities’ resilience. Interventions because of the positive role that small most frequently mentioned were justified business and IGAs have played on their on the basis that they would increase ability to cope. productive assets and business skills, and hence income. 4. Summary of feedback from local consultations • Livestock production: Interventions around improving livestock production The findings presented above emerged were cited most frequently (18 during feedback sessions with local respondents). These included stakeholders in Marsabit. Separate improving livestock market access, sessions were held with community livestock management practices, representatives (on 3 September) and with restocking with more drought-resistant technical stakeholders (on 4 September) to breeds, pasture management, fodder accommodate their differing requirements production and animal health care. due to varying levels of literacy, language • Business skills and savings and credit: skills and technical sophistication (see Expansion of business skill and savings Appendix 3 for the list of the technical and credit opportunities were also stakeholders participants). Participants in widely cited (15 respondents). These the meeting with technical stakeholders interventions included business were briefed on the discussions with training, promotion of IGAs through community representatives. No major savings and credit/self-help groups, gaps or disparities were found between and increased access to credit and the views of the two groups. The main banking services. feedback points on the data presented • Water: Water interventions (13 are included here, to provide additional respondents) were highly rated, usually understanding and context. in relation to the ability to store more
56 Understanding Community Resilience • Participants felt that the ranked livelihoods, especially in Marsabit resilience characteristics resonated Central. This may be because crop with the reality in these communities. farming has declined over the last Generally speaking, there were few years due to the lack of rain, no surprises, and both community particularly in highland areas. representatives and technical • Participants agreed with the resilience- stakeholders confirmed that the building measures cited. The only statements prioritized by the intervention participants considered communities were what they expected. critical that was not mentioned was However, the participants felt that livestock support services such as several characteristics were missing veterinary services and extension or under-prioritized. Some of the key services on animal husbandry practices. observations and comments made In addition, participants stressed the regarding the priority characteristics need for interventions to enhance and their attainment rates are as community and household capacity follows: across all sectors so as to avoid - Road networks: Road networks were dependency on external support and mentioned but not ranked highly maximize the long-term impacts and because they were not associated sustainability of these interventions. with trade. Road networks may have • Participants felt the characteristics of been mentioned partly because a resilient household – employment, of the massive infrastructure other small business enterprises and development being undertaken large herd size – were accurate. They in the county by the national felt these characteristics reflected the government. local reality in Marsabit. No particular - Natural resource management: comments were made. Forest conservation supports lives • Attainment of resilience scores is and livelihoods, particularly in the based on perception and was seen mountain areas of Marsabit Central, as subjective. Hence, the participants by maintaining reliable flows of water noted the risk that the scores could downstream and providing trees for be arbitrary and heavily influenced timber and charcoal. However, the by how the community was feeling communities interviewed failed to at the time of the interview, including make this connection, and seemed in relation to the current seasonal to have little if any awareness of conditions (i.e., wet season or dry the importance of natural resource season). Accordingly, the conclusion management. was made that the attainment scores - Livestock markets: Communities are useful indicators of community did not rank livestock markets highly perceptions but are not fully objective as a priority resilience characteristic and therefore should not be relied because they lost a lot of animals upon in isolation but should be viewed during the last drought and have together with other quantitative been focused on rebuilding stocks, data. For example, local peace and not trading at market. As a result, security fluctuates significantly over this characteristic was underscored, the short term, and the situation can though it is very important for vary from location to location. Hence resilience. the attainment scores may not fully - Productive farms: Productive farms capture the real and detailed picture did not rank highly compared of what is happening on the ground to other identified resilience over an extended period. While characteristics, despite the critical the attainment score in June 2013 importance of the sector for local demonstrated relative local safety and
Marsabit Central, Laisamis and Maikona districts 57 stability, it could drop considerably in resilience, while youth focused on other months. human characteristics. Men placed • Participants highlighted the following more weight on financial characteristics specific recommendations related to than did women and youth. building resilience: • When asked to describe the specific - Provide continued support for characteristics of resilient households pastoralists and the overall in their communities, focus group livestock sector as the principal participants consistently emphasized livelihood group in the county. income and assets. Resilient Focus services and interventions on households were described as having improving livestock production and access to employment, diversified productivity. income-generation opportunities - Map out alternative income- and/or large herds, almost to the generation opportunities in view exclusion of any other characteristics. of the specific contexts of different Critically, it was the diversity of these locations, as the next step towards three characteristics in combination livelihood diversification support. that seemed to be key, by allowing - Conduct further analysis of priority households to spread risk across resilience characteristics, for income sources. This was consistent example, what has hindered or with the findings from interviews with improved education, what areas resilient households, who cited access require greater investments (quality, to multiple incomes as the primary enrolment levels, etc.). reason for their resilience. Access to education and credit were consistently 5. Conclusions and described as the direct means to recommendations achieving higher income and greater assets and hence key driving factors to reaching a resilient status. 5.1 Conclusions • The level of attainment of resilience The evidence presented above is characteristics was consistently extensive, adding a significant amount low across all groups even at the of understanding to the community time of the assessment, which was perspective on resilience. The following considered normal or above normal conclusions are drawn from the evidence in terms of precipitation. However, presented. the drop in crisis times was not as great as might be expected. Overall, • Many of the community perspectives for all three districts in the current on the building blocks of resilience period (attributed as a normal year), converged around widely used overall scores hovered around 3.7 out sectoral indicators. The top five of 10 in terms of attainment. The only statements used to describe a resilient exception was social characteristics, community for all three districts which were ranked closer to 5 out of included peace and security (score 10, likely due to the success of recent of 792), education (608), water for initiatives on peace and security. humans (598), health care for humans In crisis times, scores for different (232) and access to markets (232). resilience characteristics typically When categorized by SLF category, contracted to an approximate score of physical characteristics of resilience 2.4 out of 10. ranked most highly, followed by human • Pastoralists scored themselves as and social characteristics. Interestingly, the most resilient relative to other women and men placed the greatest groups. However, their scores may still weight on physical characteristics of be considered low (4.1 current and
58 Understanding Community Resilience 2.8 during crisis). Most of them (76 should focus on these areas first and percent) also viewed their resilience foremost. These include peace and level as increasing. Agropastoral security, education, water, health care groups scored themselves as the least for humans and access to markets. resilient (2.6 current and 1.6 during Importantly, this set of indicators crisis), with highly mixed views on the encompasses both long-term and change in their resilience level over short-term issues and includes factors time (38 percent saw it as increasing, that are not always immediately 38 percent as decreasing and 25 considered part of DRR strategies, as percent as unchanged). well as factors that require a long-term • High intervention areas scored commitment to investment. Hence a themselves as the most resilient. broader conception of DRR is required However, this group was also most if resilience is to be built. Specifically: likely to say their resilience was - In the short term, interventions such decreasing. Causality cannot be as access to markets and credit are inferred; high intervention areas less costly and could be prioritized could be more resilient because of (bearing in mind that their success interventions, or it could be that will be handicapped if longer term interventions are targeted to groups interventions are not developed in with higher levels of resilience. tandem). • The interventions that are most highly - In the long term, interventions such rated for helping to build resilience as education and development of link directly to the most highly infrastructure such as roads and rated characteristics of resilience mobile communication facilities are – water, education and restocking. critical – ignoring them because Interestingly, resilient households cite they are expensive and instead a different set of priority interventions, investing in cheaper short-term namely measures to improve livestock programmes is likely to be a false production, business skills, savings economy. and credit, water interventions and - Maintaining peace and security measures to improve agriculture. must move higher up the agenda • The community perspective and must reflect understanding on resilience is central to our of the multi-sectoral nature of the understanding, but it does not causes and underlying factors that capture all relevant issues. Clearly, for contribute directly and indirectly to instance, the fact that communities insecurity. gave less weight to natural resource • Planning and implementation of future management does not mean it is interventions should be based on not important; quite the opposite. the extent to which they build and Feedback from the stakeholder group diversify incomes and assets, either was critical in shining light on some of directly or indirectly. Employment, these gaps in community awareness, income generation and large herds including the importance of livestock were consistently and uniformly used markets, agricultural inputs and road to describe resilient households. When networks. these households were asked about the pathways that led to resilience, access 5.2 Recommendations to services such as education and markets were the factors that helped • The high priority given to a relatively them diversify their income base. small set of indicators for building • The community perspective must be resilience indicates that disaster incorporated into the DRR/resilience risk reduction (DRR) strategies agenda. The assessment findings show
Marsabit Central, Laisamis and Maikona districts 59 that communities are highly aware of of the wider spectrum of factors that the long-term and short-term factors affect it and consider how they can re- that contribute to or undermine their focus their efforts (or those of others) resilience. Too often consultations with and budgets in these areas. communities focus on interventions • Ongoing monitoring and evaluation that are already designed by ‘technical are critical to measuring changes experts’ approved by ‘decision- in resilience. The consistency in makers’ in a top-down manner or community comments suggests that a for which there is budget available. few key indicators could be identified to Failure to respond to the local context monitor resilience more systematically. and integrate the needs and priorities Some of these indicators (such as of the beneficiaries may partially household income levels) are already account for the fact that the districts being measured as part of ongoing where the assessment was conducted data collection exercises. Others (such have faced drought crises every few as peace and security) would require years despite the level of humanitarian some level of consensus to agree an and development support provided in appropriate indicator(s) and methods the region. Stakeholders involved in of collection. building resilience need to be aware
60 Understanding Community Resilience Appendix 1. Participants in CoBRA Marsabit assessment
Phone No. Name Organization Email address number
1 Henry Halakano NDMA 0727695504 [email protected]
2 Sora Adano Wario NDMA 0710373717 [email protected]
3 Benjamin Migolo FH-Kenya 0703542949 [email protected]
4 Mary Philip Soso FH-Kenya 0721829777 [email protected]
5 Forole Paul Karimi FH-Kenya 0714306466 [email protected]
6 Anisa Lesurmat FH 0701285363 [email protected]
7 Gobu Wario Kenya Red Cross Society 0725877751 [email protected]
8 Michael Ngutu Food and Agriculture 0722844782 [email protected] Organization of the United Nations
9 Orge G. Bajji Community Initiative 0721862507 [email protected] Facilitation and Assistance
10 Mohamed Hassan Horn of Africa 0711468200 [email protected] Development Initiative
11 Barako Giro Katelo Pastoralist Community 0208009585 [email protected] Initiative Development and Assistance
12 Sharamo M. Alex District Education Office 0725746560 [email protected]
13 Meshack Omarre Building Owners and 0713642155 [email protected] Managers Association International
14 Francis Opio UNDP DDC 0727313670 [email protected]
15 Catherine Fitzbibbon UNDP Consultant 0734498324 [email protected]
16 Laban Macopiyo UNDP Consultant 0728466622 [email protected]
17 Moses Ojota UNDP Consultant 0772953186 [email protected]
Marsabit Central, Laisamis and Maikona districts 61 Appendix 2. Complete resilience statements and scores for CoBRA Marsabit
Resilience Bean Total bean Full resilience statement12 characteristic score score
Financial
Productive farms Farmers would be more productive and profitable (i.e., they 114 554 would have inputs like quality tools, oxen, fertilizers and improved knowledge of good farming practices).
Diversified IGAs Many households would be involved in other IGAs such as 102 small businesses and trading.
Livestock herds Pastoralists would have herds large enough to sustainably 84 support their families.
Pasture and fodder There would be sufficient pasture (or fodder) for livestock at 73 all times of the year.
Employment There would be many opportunities for jobs and other 68 forms of paid employment through government, factories and other businesses.
Access to credit People would have good access to affordable credit and 61 would be saving money (through banks, microfinance organizations, community savings and credit).
Health care for The community would have access to high-quality and 51 livestock affordable animal health services, including veterinary services and vaccinations, whenever they need them.
Land Everyone would have secure access to/ownership of land/ 1 property.
Human
Education All children would be able to complete primary/secondary/ 608 1,032 tertiary education.
Health care for The community would have access to quality and affordable 232 humans basic health care locally.
Food security All households would be able to feed themselves well 145 every day.
Health The community would have access to high-quality and 34 affordable basic health care locally [unspecified as to whether this is for human or animal purposes]
Skills The community would have the skills and structure to plan 13 and implement solutions to their own problems.
Natural
Natural resource Local rangelands and other natural resources would be well 83 105 management managed so they do not become degraded over time.
Wildlife conservation The natural environment and wildlife would be conserved 13 for tourism and other purposes.
Forests Trees and forest cover would be well and sustainably 7 managed to provide various services for future generations.
Alternative fuel Communities would use environmentally friendly/ 2 sustainable sources of fuel for cooking.
62 Understanding Community Resilience Resilience Bean Total bean Full resilience statement12 characteristic score score
Physical
Water for humans The whole community would have access to sufficient, 598 1,657 good-quality water at all times of the year.
Telecommunications There would be a reliable mobile phone network to all 247 communities all the time.
Access to markets The community would have easy access to markets to buy 232 goods and sell their produce.
Roads There would be good-quality roads to the community. 149
Water for livestock Livestock would have access to sufficient water at all times 146 of the year.
Shelter Everyone would live in good-quality housing. 103
Irrigation Farmers would be irrigating land to improve the production 60 of crops for consumption and sale.
Water [Human or livestock use not specified] 50
Sanitation Everyone would have good sanitation. 33
Electricity The community would have access to affordable electricity 12 supply.
Transport 9
Manufacturing 8
Recreation facilities 7
Access to processing 3
Social
Peace and security The whole community would enjoy continual peace and 792 829 security.
Good governance The community would be served by efficient and non- 18 corrupt community leaders and management structures.
Community The community would have the skills and structures to plan 9 empowerment and implement solutions to their own problems.
Religion 5
Women Women would be fully involved in local development and 5 leadership.
12 Note that blank statements indicate characteristics that were added by communities during focus group discussions, and as such did not have a generic full statement associated with them.
Marsabit Central, Laisamis and Maikona districts 63 Appendix 3. Participants in CoBRA Marsabit technical stakeholders’ feedback session
No. Name Organization Email address
1 Henry Halakano NDMA [email protected]
2 Sora Adano Wario NDMA [email protected]
3 Qanchora Huqa NDMA
4 Benjamin Mwigolo FH-Kenya [email protected]
5 Mary Philip Soso FH-Kenya [email protected]
6 Forole Paul Karimi FH-Kenya [email protected]
7 Michael Ngutu Food and Agriculture Organization of the [email protected] United Nations
8 Orge G. Bajji Community Initiative Facilitation and [email protected] Assistance
9 Mohamed Hassan Horn of Africa Development Initiative [email protected]
10 Doyi Barako Pastoralist Community Initiative [email protected] Development and Assistance
11 Meshack Omarre Building Owners and Managers Association [email protected] International
12 Letikirich S.R Ministry of Agriculture
13 Alex Sharamo Ministry of Education – Chalbi
14 Abdullah Boya Ministry of Health
15 Christopher Muia Ministry of Health [email protected]
16 Peter Gelkuku Kenya Agricultural Research Institute - Marsabit
17 Ruto Chepchumba Office of the President
18 Mamo Sora Agricultural Sector Development Support [email protected] Programme
19 Amos Kipruto Concern [email protected]
20 Guyo Yattni Tocha Kanacho Nomadic Education Foundation
21 Dub Guyo Resilience and Economic Growth in the Arid [email protected] Lands–Accelerated Growth – REGAL - IR
22 Tesso Dido Hifo Marsabit [email protected]
23 Francis Opio UNDP DDC [email protected]
24 Laban Macopiyo UNDP Consultant [email protected]
64 Understanding Community Resilience Annex 2 Community-based Resilience Analysis Assessment Report Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts Turkana county, Kenya
Commissioned by UNDP Drylands Development Centre
Under the framework of the European Commission Directorate General for Humanitarian Aid and Civil Protection’s Drought Risk Reduction Action Plan
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 65 Contents
1. Introduction...... 67 2. Approach...... 67 2.1 Characteristics of field site...... 67 2.2 Data collection ...... 69 2.3 Constraints and limitations of data collection process ...... 70 3. Findings...... 70 3.1 Main hazards or shocks ...... 71 3.2 Characteristics of a resilient community ...... 71 3.3. Extent to which the community has achieved resilience...... 76 3.4. What a resilient household looks like...... 77 3.5 Interventions that contributed to household resilience...... 79 3.6. How key informants achieved resilience...... 80 4. Summary of feedback from local consultations...... 82 5. Conclusions and recommendations...... 83 5.1 Conclusions...... 83 5.2 Recommendations...... 84
66 Understanding Community Resilience 1. Introduction the key feedback and consolidated inputs generated at review and validation The comprehensive Community-Based workshops for the draft assessment report, Resilience Analysis (CoBRA) assessment attended by technical stakeholders from was undertaken in six districts of Turkana local government and non-governmental county, Kenya, from 1–6 July 2013. The organizations working in the area on 2 exercise was carried out with financial October and by community representatives support from the European Commission on 3 October. Directorate General for Humanitarian Aid and Civil Protection under the framework A detailed explanation of the conceptual of its Drought Risk Reduction Action Plan. framework that underpins the methodology This assessment builds on the initial field is contained in the CoBRA Conceptual trial of the draft CoBRA tool undertaken in Framework and Methodology document.1 late 2012. 2. Approach The multi-agency assessment was jointly led by the UNDP Drylands Development Centre (DDC) and the Government of 2.1 Characteristics of field site Kenya’s National Drought Management The field assessment was conducted in Authority (NDMA). Logistical and practical six districts (Turkana North, Turkana West, support was provided by Oxfam. A wide Loima, Turkana Central, Turkana South, range of international and local NGOs Turkana East) of Turkana county, Kenya. operating in the area also participated Within these six sampled subdistricts, 20 in the assessment by providing local administrative locations were selected staff as facilitators to undertake CoBRA for the assessment (Figure 1), based training and fieldwork. A list of agencies on the county’s four livelihood zones: participating in the CoBRA training and/ pastoral, agropastoral, urban and fishing.2 or field data collection is provided in Pastoralism is still the predominant Appendix 1. livelihood, accounting for 55 percent of the population. Almost one quarter The CoBRA assessment has four broad (23.4 percent) of the people are in urban/ objectives: peri-urban areas, and the remainder are agropastoralists (15.6 percent) and 1. Identify the priority characteristics of fisherfolk (6 percent). resilience for a target community; 2. Assess communities’ achievement of The sampling frame listing the locations these characteristics at the time of where focus group discussions (FGDs) the assessment and during the most and key informant interviews (KIIs) were recent crisis or disaster; undertaken is shown in Table 1. 3. Identify the characteristics and strategies of resilient households; and Turkana county is situated in the arid 4. Identify the most highly rated north-western region of Kenya, sharing interventions or services in building international borders with Ethiopia, Sudan local resilience. and Uganda. The district has an estimated total population of 855,399 as per the 2009 This report outlines findings of the Turkana census, and the majority of the population CoBRA assessment. It also incorporates is ethnic Turkana. The county covers an
1 www.youblisher.com/p/784724-Community-Based-Resilience-Analysis-CoBRA-Conceptual-Framework-and-Methodology. 2 The Food Economy Group, Livelihood Profiles from Six Livelihood Zones in Turkana County, Kenya: Assessed Using the Household Economy Approach (Nairobi: Oxfam and Save the Children).
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 67 Figure 1. Surveyed locations in Turkana county
Surveyed locations
area of 77,000km.2 It has recently been county is classified as part of the Arid and divided into 6 districts comprised of 17 Semi Arid Lands (ASALs). administrative divisions. Turkana consists mostly of low-lying plains Turkana county is the poorest in Kenya; an with isolated mountains and hill ranges. estimated 88 percent of the population It receives unreliable and erratic rainfall lives below the absolute poverty line.3 of less than 100 mm annually, coming in Poverty in Turkana is exacerbated by the two seasons: the long rains occur between harsh environment, poor infrastructure and April and July and the short rains between low access to basic services, in addition October and November. The county has a to other underlying causes of poverty that warm and hot climate with temperatures are experienced elsewhere in Kenya. The ranging between 33°C and 40°C. As
3 Kenya National Bureau of Statistics and Society for International Development, Exploring Kenya’s Inequality: Pulling Apart or Pooling Together? (Nairobi, Ascent Limited, 2013).
68 Understanding Community Resilience Table 1. FGDs and KIIs undertaken for Turkana CoBRA assessment
No. of No. of District Location Livelihood FGD composition FGDs KIIs
Kaeris Pastoral 2 2 Turkana North Loruth Pastoral 2 2
Kakuma Urban 2 4 men 2 4 women Pelekech Pastoral 2 4 youth 2 Turkana West Lokichoggio Pastoral 2 2
Mogila Pastoral 2 2
Lorengippi Pastoral 2 2
Kotaruk Agropastoral 2 2 Loima Lomeyan Pastoral 3 3 3 men Nadapal Agropastoral 3 trials* 4 women 3 trials 3 youth Lorengelup Pastoral 3 trials 3 trials 1 businessperson Kanamkemer Urban 2 trials 1 urban dropout 2 trials 8 trials Turkana Central Lodwar Township Urban 2 2
Kalokol Fisherfolk 2 2
Namukuse Fisherfolk 1 1
Katilu Agropastoral 2 2
Turkana South Kalapata Pastoral 3 4 men 3 4 women Lokichar Urban 2 2 4 youth Katilia Agropastoral 3 3 Turkana East Lokori Urban 2 2 36 + 8 36 + 8 Total trial sites trial sites
* Prior to undertaking full FGDs and KIIs in the field, the facilitators and supervisors were trained on the CoBRA methodology. Trial FGDs and KIIs were organized during the training as part of the field-based exercises, and those results, which were confirmed by the supervisors as of adequate quality, were included in the final survey results (42 FGDs and 42 KIIs in total). an ASAL county, Turkana experiences in Kenya and in adjacent countries. This frequent, successive and prolonged insecurity regularly leads to heavy loss of droughts. The last severe drought was life and livestock and undermines both experienced between 2010 and 2012, population movements and development when 58 percent of the population was efforts. assessed as needing food aid.4 2.2 Data collection Drought puts pressure on natural resources and escalates the area’s chronic insecurity. Full details of the methodology used Cattle rustling is common between the to undertake a CoBRA assessment are Turkana and neighbouring pastoral tribes included in the main CoBRA assessment
4 Kenya Food Security Steering Group, Short Rains Assessment Report, 2010.
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 69 report and implementation guidelines. the extent to which the community had Data were collected using qualitative attained the top-ranked characteristics methods, i.e., focus group discussions of resilience. The trained facilitators (FGDs) and key informant interviews (KIIs).5 asked communities to select the level of attainment on a scale of 0–10, with 10 Field work was undertaken by a team of being the perfect or ideal situation and 12 facilitators and 3 supervisors. The team 0 an absolute lack of attainment. This was divided into three groups of five approach had been adopted to enable people each, consisting of a supervisor scores to incorporate non-household and two pairs of facilitators. Each group related issues, such as peace and security, covered two of the six districts surveyed. and those that require a more qualitative Each FGD was facilitated by two staff from assessment, such as health care services. It the participating government departments was difficult for communities to objectively and NGOs. Table 1 summarizes the number define the scale or levels of attainment, and locations of FGDs and KIIs undertaken which presented a challenge in most cases for this assessment. The facilitators and both to facilitators and to focus group supervisors, who were familiar with the participants. local context, jointly selected the locations and compositions of the focus groups, Long distances between communities using statistical data and criteria such as Turkana county is very large, and it was not livelihood zones and level of interventions. quite clear from the assessment whether Whenever possible, separate FGDs were the sample size was adequate to cover organized for men, women and youth the diversity of communities that live there (aged 15–30, both sexes) to ensure that (but far apart) and their livelihoods, and views on resilience from different gender whether the aggregation of sample strata and age groups would be adequately actually represented shared community captured in the discussions. Each group values and opinions. From the feedback had around 15 participants. workshop, however, it emerged that the manner in which sampling area and In each focus group location, one to four locations were selected – with a focus on KIIs were undertaken with households major geographic areas (North, Central perceived to be resilient, as identified and South in combination with livelihood by focus group participants during their zones) – allowed for a significant amount discussions. In some cases, the households of variation and good representation. were also identified as resilient through conversations with local chiefs or other 3. Findings senior residents. A total of 36 FGDs and KIIs took place, in addition to those at the This section reports on the summarized 8 trial sites. findings from the CoBRA field work conducted during June 2013 in Turkana. 2.3 Constraints and limitations of Specifically, the findings sought answers to the following questions: data collection process Scoring the attainment of resilience • What are the main hazards or shocks characteristics facing the communities? • What are the characteristics of a After jointly defining resilience, focus resilient community? group participants were asked to score • To what extent has the community achieved resilience characteristic?
5 UNDP-DDC, Understanding Community Resilience: Findings from Community-Based Resilience Analysis Assessments – Marsabit, Turkana and Kajiado counties, Kenya, and Karamoja sub-region, Uganda (Nairobi, UNDP-DDC, 2014 [in press]).
70 Understanding Community Resilience • What does a resilient household look is followed by an analysis by category of like? respondent – gender/age, livelihood group • What interventions contributed to and level of intervention in the community household resilience? are used to disaggregate findings and • What additional intervention would identify differences across groups. best build resilience? • How did key informants achieve and Analysis – all respondents maintain resilience? Focus group participants were asked to identify and rank statements used 3.1 Main hazards or shocks to describe a resilient community. Each participant was given six beans to rank the The main hazard reported in all the three most significant statements (three FGDs was drought. Communities viewed beans for the most significant statement, drought as the most significant contributor two for the second and one for the third) to livestock losses and the single most in terms of priority for building resilience. important factor limiting their resilience The bean scores were then totalled for capacity. Communities also reported that, each statement. For ease of comparison, while past droughts have been longer, the the statements were grouped into the five most recent drought, of 2010–2011, was sustainable livelihood framework (SLF) the most severe. The current period was categories.6 perceived as principally ”normal”, with a few of the communities stating that it was Table 2 lists the most highly ranked a “good” year. statements used to describe a resilient community within each of the five SLF Other hazards/crises reported include categories. (Note that many more conflict (largely cattle rustling with statements/characteristics were included bordering communities of Pokot and also in the ranking, but were given low scores cross-border raids with communities from and hence are not reported here.) The full Ethiopia, Sudan and Uganda). Competition list with scores is provided in Appendix over scarce grazing fields, water resources 2, which also lists the total scores for all and pasture escalate inter-tribal animosity the resilience characteristic statements especially during drought, often resulting identified by the focus group participants in armed conflicts, which are predatory in in each category. nature and very destructive. To a limited extent, the communities also reported Figure 2 shows the highest scored livestock diseases and human diseases as statements used to describe a resilient other hazards that plague them. community by all focus group participants in the three districts in the order of scores. 3.2 Characteristics of a resilient Figure 3 presents the total score under each of the five SLF categories. community Focus group participants were asked The results show that the surveyed to describe what they viewed as the communities perceived education as the characteristics of a resilient community. The most important characteristic of resilience. data are first presented for the entire set It should be noted that in almost all focus of respondents to give an overall picture groups, this meant access to secondary of the most highly rated statements. This and/or tertiary levels of education. Water
6 The sustainable livelihood framework presents the main factors that affect people’s livelihoods and typical relationships between them. It identifies five core asset categories or types of capital upon which livelihoods are built: financial, human, natural, physical and social. For further details, please refer to: UK Department for International Development (DFID), Sustainable Livelihoods Guidance Sheets (London, DFID, 1999).
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 71 Table 2. Community ranking of resilience characteristics by SLF category
Total bean SLF category Top-ranking resilience characteristics (‘bean scores’) score* Financial Diversified income generating activities (IGAs) (375) 1,162 Access to credit (287) Employment (135) Livestock herds (125)
Human Education (763) 1,170 Health care for humans (231) Food security (173)
Natural Natural resource management (NRM) (118) 118
Physical Water for humans (396) 1,295 Access to markets (265) Irrigation (260) Roads (109)
Social Peace and security (367) 437
*Total bean score does not match the sum of the bean scores in the middle column because the middle column reflects only the top-ranking characteristics while the total in the right column reflects all the characteristics discussed in each category.
Figure 2. Top-ranking resilience characteristics – all respondents 900 800
700
600
500
400 300
200
100 0
NRM Roads
Irrigation Education
Employment Food security Livestock herds Diversified IGAs Access to credit Water for humans Access to markets Peace and security
Health care for humans
for humans, diversified IGAs and peace capacity received the lowest overall and security followed, with similar scores. ranking. The statements covering physical capital rank most highly, largely due to the very Analysis by gender and age high ranking given to access to adequate The priority resilience statements were water for humans. The human and financial analysed by gender and age group, categories are also ranked highly, while through groups comprised of women, men statements relating to social and natural and youth (aged 15–30, both sexes). The
72 Understanding Community Resilience Figure 3. Top-ranking resilience characteristics by SLF category – all respondents
1400 Others Peace and security 1200 Roads Irrigation 1000 Access to markets Water for humans 800 NRM Food security 600 Health care for humans Education 400 Livestock herds Employment 200 Access to credit Diversified IGAs 0 Financial Human Natural Physical Social
Figure 4. Top-ranking resilience categories by gender/age group
100%
80%
60% Social 40% Natural Financial 20% Human Physical 0% Women Men Youth bean score allocated to each of the five Table 3. Top-ranking resilience characteristics SLF categories was taken as a percentage by gender/age group of the total bean score, and the results are represented in Figure 4 to demonstrate Gender/ Top 3 resilience Total age characteristics score* the differing priorities that men, women and youth place on resilience statements. Women Education (284) 1,500 Diversified IGAs (209) Further to this, the most highly ranked Water for humans (143) resilience statements by gender/age Men Education (253) 1,485 group are presented in Table 3. Peace and security (202) Water for humans (162) The data suggest the following points: Youth Education (227) 1,197 Access to markets (119) • All three groups ranked education as Access to credit (113) the most important characteristic of resilience. Access to clean water for * Total score does not match the sum of the scores in the middle column because the middle column reflects only the top-ranking humans was also listed by all three characteristics while the total in the right column reflects all the groups. characteristics discussed in each category.
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 73 • Access to credit and diversified IGAs The data suggest the following points: (which are closely linked) also feature very highly, particularly among women • Education was the top-ranked and youth. characteristic of resilience across all groups, with the exception of Analysis by livelihood group agropastoral communities, where it The analysis included a range of livelihood is trumped by peace and security. groups: agropastoral, pastoral, urban Agropastoral groups are soft targets and fisherfolk. Figure 5 illustrates the for rustlers because they are sedentary differences in bean score allocation and unarmed, and therefore peace according to the five SLF categories and security are high on their agenda. among the four livelihood groups.
Figure 5. Top-ranking resilience categories by livelihood group 100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Agropastoral Pastoral Urban Fishing
Further to this, the most highly ranked • Diversified IGAs and water for humans resilience statements by livelihood groups also feature highly among the majority are shown in Table 4. of groups.
Table 4. Top-ranking resilience characteristics Analysis by intervention level by livelihood group The assessment was conducted in 20 administrative locations in the six districts. Livelihood Top 3 resilience Total group characteristics score* Consultation with field teams, NDMA and other government line departments was Agropastoral Peace and security (190) 1,085 used to map accessibility to and presence Education (156) Food security (86) of basic services and interventions in each location as follows: Fisherfolk Education (66) 300 Water for humans (59) • Number and level of health care Diversified IGAs (49) facilities; Pastoral Education (272) 1,697 • Tarmac road; Irrigation (166) • Other main road; Water for humans (153) • Mobile phone coverage; Urban Education (268) 1,100 • Well-functioning livestock market; Diversified IGAs (143) • Savings and credit programmes; Water for humans (124) • Number and level of education * Total score does not match the sum of the scores in the middle facilities; column because the middle column reflects only the top-ranking characteristics while the total in the right column reflects all the characteristics discussed in each category.
74 Understanding Community Resilience • Banks; and Table 5. Top-ranking resilience characteristics • Bank agents/Mpesa.7 by intervention level
The mapping exercise excluded Intervention Top 3 resilience Total level characteristics score* interventions that are universally provided in all locations, such as food aid and water High Education (442) 2,185 interventions, or provided only at very low Diversified IGAs (255) scale, such as less than 500 beneficiaries. Water for humans (191) The locations were then divided into three Medium Education (66) 800 even groups: Water for humans (59) Diversified IGAs (49) • The bottom third, representing low Low Education (136) 1,197 intervention areas, which have up Peace and security (95) to 5 categories of interventions Health care for humans (54) implemented in their area; * Total score does not match the sum of the scores in the middle • The middle third, representing column because the middle column reflects only the top-ranking medium intervention areas, which have characteristics while the total in the right column reflects all the 6 to 11 categories of interventions characteristics discussed in each category. implemented in their area; and • The top third, representing high The data suggest the following points: intervention areas, which have 12 to 34 categories of interventions • Education was the top-ranking implemented in their area. characteristic, with diversified IGAs and water also playing a key role. Figure 6 shows the differences in bean • High intervention areas seem to score allocation according to the five SLF have a greater focus on education, categories among the three location groups diversified IGAs and credit, whereas with different service/intervention levels. low intervention areas placed a greater emphasis on peace and Further to this, the most highly ranked security. resilience characteristics by group are presented in Table 5.
Figure 6. Top-ranking resilience categories by level of intervention
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% High Medium Low
7 M-Pesa is a mobile phone-based money transfer and microfinancing service offered by Safaricom and Vodacom in Kenya and Tanzania. (M stands for mobile and pesa means money in Kiswahili.)
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 75 3.3. Extent to which the community As with the previous section, the findings has achieved resilience are presented first for all respondents and Focus group participants were asked then disaggregated by specific livelihood to score the extent to which they had groups. In this section, the analysis by achieved their priority characteristics of gender is not included, because focus resilience. They were asked to score each groups were asked to rank the attainment statement twice, first for the current period of the resilience characteristic statements (agreed to be a normal period) and second for the entire community, and therefore for the last significant crisis period (almost any differences between men and women universally identified as the drought period in the same community would be based of 2010–2011). on perceptions.
Table 6 presents the scores by SLF category Figure 7 shows a spider diagram with for the top-ranked characteristics of several rings. The outer ring represents a resilience. Social and natural characteristics perfect or ideal score for all statements in of resilience have the highest scores: in that SLF category. The blue band shows other words, communities ranked social communities’ average attainment of those and natural characteristics as the area characteristics in the current period, and where they have the greatest degree of the red line represents the last crisis period. attainment of resilience. This category was The scores are ranked on a scale from 0 dominated by responses relating to peace to 10, with 10 being perfect attainment of and security and NRM. The remainder of that characteristic (for example, the entire the categories were scored similarly in community has access to sufficient, good current and crisis years, suggesting that, quality water at all times in the period) in a normal year, community members and 0 being no attainment (no one in the score their attainment of characteristics of community has access to sufficient, good resilience at approximately 2 out of 10. quality water at all times in the period).
Table 6. Community attainment of resilience by SLF category – top-ranked characteristics
Current period Crisis period SLF category Top-ranked characteristics score score
Overall average 2.5 1.3
Financial Diversified IGAs 2.3 1.2
Access to credit 1.4 0.8
Employment 2.5 1.5
Livestock herds 3.2 2.0
Human Education 2.1 1.6
Health care for humans 3.0 2.6
Food security 2.5 1.6
Natural NRM 3.1 0.7
Physical Water for humans 2.0 1.3
Access to markets 2.7 2.3
Irrigation 0.9 0.6
Roads 1.8 1.1
Social Peace and security 3.3 1.8
76 Understanding Community Resilience Figure 7. Community attainment of resilience per SLF category – all respondents
Physical 10.0
8.0
6.0
4.0 Social Human 2.0
Current year rank 0.0 Crisis year rank Full score
Natural Financial
Note: Possible score ranges from 0 to 10.
Table 7 aggregates the scores given to each Table 7. Aggregate resilience scores of the five SLF categories to provide an overall ‘resilience score’. While this clearly Current Crisis year year rank masks differences between SLF categories, it is useful to provide an overall indicator All groups 2.5 1.3 of where communities see themselves, for comparison across groups. Clearly, figures Livelihood group should be viewed with some caution, Agropastoral 2.5 1.4 as these scores represent community perceptions around attainment and could Pastoral 2.5 1.4 therefore overstate or understate reality. Urban 2.4 1.2
All groups rank their attainment of Fisherfolk 1.3 1.1 resilience characteristics at 2.5 out of 10, Intervention level on average. This drops to 1.3 out of 10 in a crisis year. High 2.4 1.5 Medium 3.1 1.2 There is very little difference between groups, with the exception that fishing Low 2.4 1.2 communities rank their attainment of Note: Maximum possible score of 10 resilience characteristics very low, at 1.4 in the current year. It emerged during 3.4. What a resilient household consultation that most of the families looks like and locals there do not significantly Focus group participants were asked to benefit from fishing; the main winners are describe the characteristics of households traders, middlemen and people higher up that are more resilient compared to others, the value chain who offer the fisherfolk that is, the households that have already minimal prices, transport the fish in cooled attained many (or all) of the resilience trucks to major urban areas and sell them characteristics prioritized. The top three at higher prices. characteristics of a resilient household,
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 77 cited consistently by focus groups, were as Focus groups were asked whether the follows: number of resilient households is increasing, decreasing or staying the same (Table 8). • Households that have a business or IGA (34 out of 42 focus groups); Out of 42 focus groups, the majority (46 • Households with large herds (27 out of percent) said that resilience was increasing. 42 focus groups); and These findings were similar across gender/ • Households in which a member has age groups, with the exception of youth, employment/wage labour (26 out of who were more likely to say that resilience 42 focus groups). was increasing (58 percent compared with 40 percent for the women and 43 percent There was little difference between for the men). gender/age, livelihood and intervention groups. When disaggregated by livelihood group, agropastoralists had a more pessimistic A few other household characteristics were view; only 9 percent said that resilience mentioned but significantly less often: was increasing. Urban groups, on the other hand, were much more likely to say that • Household has educated members resilience was increasing (73 percent). (where this was mentioned it was often in relation to the ability of the When disaggregated by intervention educated person to find employment); group, high intervention groups were and much more likely to say that resilience was • Household has land. increasing (59 percent).
Table 8. Is resilience increasing, decreasing or staying the same?
Increasing Decreasing Staying the same Mixed response
All respondents 46% 27% 5% 22%
Gender/age
Women 40% 33% 0% 27%
Men 43% 21% 7% 29%
Youth 58% 25% 8% 8%
Livelihood group
Agropastoral 9% 45% 9% 36%
Pastoral 44% 25% 6% 25%
Urban 73% 18% 0% 9%
Fisherfolk* 100% 0% 0% 0%
Intervention level
High 59% 23% 0% 18%
Medium 36% 36% 18% 9%
Low 25% 25% 13% 38%
*The fishing response is based on only one observation.
78 Understanding Community Resilience 3.5 Interventions that contributed water and education interventions. These to household resilience reflect the high ranking given to these Communities were asked to list all the factors as characteristics of resilience by services and interventions they had all focus groups. Water interventions were benefited from in the last two to five years. prioritized by all livelihood groups for From this long list, each community (through obvious reasons, particularly for improving focus groups) was asked to identify jointly food security and livelihoods. the three current or previous interventions that had been most beneficial in building Education was seen as a benefit in itself their resilience and why. Table 9 shows that and also as leading to improved life interventions relating to education, water, chances, such as employment for children. cash transfers and health were prioritized Education facilities were ranked more most regularly. frequently than scholarships and bursaries.
Groups were also asked to list the three When disaggregated by livelihood further or additional interventions they group, agropastoralists placed the felt would best build their resilience. greatest emphasis on cash transfers Many communities restated interventions and education, while pastoralists also mentioned in the first list, with the ranked health and credit highly. High and justification that the current provision or medium intervention areas tended to rank scale of intervention was too limited and education and water services most highly, should be expanded. However, it is also whereas low intervention areas focused on interesting to note that education and cash transfers. This may reflect the higher cash transfers were mentioned much less prevalence of cash transfers as a coping frequently, while access to credit was mechanism in low intervention areas where given a much higher priority as a future resilience is likely to be lower. intervention. Interestingly, irrigation did not feature as The highest scoring interventions are highly as the other interventions, despite summarized in Table 9, which shows high levels of investment in irrigation the repeated and clear priority given to schemes in this area. Feedback from
Table 9. Ranking of resilience-building interventions
Currently/ Further Type of intervention recently or future Total provided provision
Education Bursaries, scholarships or construction/refurbishment of school 30 10 40 facilities including boarding facilities
Water 17 18 35 Water source improvement or improved storage capacity
Health 14 12 26 Improvements to health care services, staffing or facilities
Cash transfers Transfer of resources to vulnerable people, such as the 15 5 20 chronically food insecure (e.g., Hunger Safety Net Programme) and children (e.g., child sponsorship programmes)
Access to credit or other forms of business support 4 13 17
Irrigation 0 12 12
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 79 local consultations suggested that this is households in Turkana are significantly because many of the irrigation projects better educated than the average.9 have failed due to poor involvement of the community and inappropriate All 42 KII respondents had household technologies; limited land ownership, members engaged in one or more of the leading to less farmer control over following activities: irrigation schemes; and frequent flooding in 2013 within most schemes, resulting in • Business or petty trade (35 respondents near total crop failure in the county. out of 42); • Raising livestock (30); 3.6. How key informants achieved • Waged employment or casual labour (19); resilience • Fishing (3); and A total of 42 key informant interviews were • Agricultural production (9). undertaken with members of households identified as ‘resilient’. One to three KIIs Overall, the vast majority of resilient were undertaken in each sampled location. households (38, or 90 percent) reported KIIs examined the following four areas: multiple income sources, including members with wage labour and/or a • Composition and characteristics of the business interest. Most households household; interviewed (33, or 78 percent) had a • Pathways to resilience; source of income from livestock and/or • Ability to cope with recent shocks and agriculture or fishing. Only four households hazards; and (9.5 percent) survived solely on one income • Priority interventions recommended source: fishing (two) or a single business by resilient households. income (two). A business interest or activity was the most common income source for Composition and characteristics of most resilient households interviewed resilient households (33) combined with either a pastoral or The KII record sheet records the age, agricultural income and/or a wage labour gender, education level and economic source. The diversification of income activity of all members of the household. sources is clearly a key strategy for resilient The resilient households varied in size from households. According to government 1 to 10 members with an average of 5.7 statistics, only 6 percent of the county’s members, which is typical for the area. population is working for pay.10 The majority of households interviewed (69.1 percent) had members who had Most wage earners were government completed at least primary education, employees, civil servants, chiefs and and 18 households (42.9 percent) had workers in schools or hospitals. Other members who had completed secondary workers mentioned included a guard, a or tertiary education. Government statistics mason and NGO staff. Four households indicate that in Turkana county, 82 percent cited reliance on casual labour, but no of the population has no education. Only details were recorded. 15 percent of the county’s population has a primary education and 3 percent has a Pathways to resilience secondary or higher education.8 From Virtually all respondents cited their these figures, it would seem that resilient multiple sources of income as the reason
8 Kenya National Bureau of Statistics and Society for International Development, op. cit. 9 Kenya Commission on Revenue Allocation, Kenya County Fact Sheets, December 2011. 10 Ibid.
80 Understanding Community Resilience for their resilience. The presence of a wage while others mentioned use of savings, earner or business income was mentioned loans from family and reliance on business 9 and 13 times respectively. Fifteen or wage incomes. Two respondents said households mentioned that having a large they cut back on expenditures and food herd and/or the efficient trading and sale intake and two others mentioned the value of livestock was an important factor or of school feeding programmes for school- strategy in making them resilient. At least age children. eight respondents described themselves as having good business skills, being Few households mentioned any specific ‘business oriented’ or having hard-working external interventions as valuable in or well-managed households. coping with the drought period. The only programmes mentioned were school Households with a wage earner or feeding (two households) and an NGO business regularly explained how income voucher programme (one household). It from either of these sources had been was interesting that there was no explicit saved and used to start/grow businesses, mention of food aid, cash transfers or water grow livestock herds and/or invest in interventions, all of which are widespread agricultural production. The business in Turkana. activities described were mostly shops and small kiosks or charcoal production. Priority interventions Unfortunately limited detail was recorded Key informants were asked for the three on the types of goods or services provided most important interventions to improve by shop/kiosk owners. Trading in animals their communities’ resilience. Those most was repeatedly mentioned by households highly mentioned were justified on the basis in all livelihood zones as both a source of that they would increase productive assets income and a coping strategy. and business skills, and hence income.
In addition to diversified income sources, • Expansion of savings and credit nine households stated that the use of opportunities (21 respondents): their own savings (often as part of saving These were cited most widely. The and credit groups) had contributed to their importance of accessing affordable resilience. Two cited inheritance or gifts credit was seen as very important, with from family. One household reported that a few also mentioning business training a voucher scheme by an NGO had enabled and the promotion of diversified IGAs. them to become resilient. • Irrigation (18 respondents): This was also highly rated particularly, but Ability to cope with recent shocks and not exclusively, by agropastoralists. hazards The presence of significant irrigation Almost all key informants referred to the projects in Turkana seems to have 2010–2011 drought as the last major influenced the resilient households, hazard. Most indicated that they were and there is much support for better placed to cope with this than others expansion of irrigation, even for due to the additional income sources projects supporting small market outlined above. Wage earners and most gardens. businesses were not affected by the • Education (15 respondents): drought. Most respondents referred to the construction of tertiary facilities or Responses on this issue mostly reiterated bursaries for secondary or tertiary the strategies and actions mentioned when education. describing how the household became • Water interventions (11 respondents): resilient. Again, over 15 households These were commonly cited, mentioned the timely sale of livestock, particularly by pastoral groups.
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 81 Other interventions mentioned by more adequately in the county. They than five respondents included roads highlighted access to credit and water (eight) and livestock-related interventions as sectors that are critical to resilience including markets (eight) and health (six). but are largely neglected. They also felt that education was critical as the The interventions prioritized by resilient county had only a handful of schools households are very similar to those (largely supported by the Catholic mentioned by the focus group participants, Diocese of Lodwar) beyond primary but with different prioritization. schools and until recently only one or Interventions to expand savings and credit two tertiary institutions. and business support, etc., receive much • Participants agreed with the resilience- higher priority. This is not surprising given building measures cited. The only that respondents are much more involved intervention they considered critical in business and commercial activities. that was not mentioned was livestock support services, namely health 4. Summary of feedback from services and water. They mentioned local consultations that this was likely due to the fact that livestock were relatively healthy at the The findings were presented at feedback time of the assessment and therefore sessions with local stakeholders in not a high-priority issue. Turkana. Separate sessions were held with • Participants felt that the technical stakeholders (on 2 October) and characteristics of a resilient household with community representatives (on 3 – employment, diversified IGAs and October) to accommodate their differing large herds – were accurate. They requirements due to varying levels of made no particular comments. literacy, language skills and technical • Attainment of resilience scores is sophistication (see Appendix 3 for the list based on perception as well as timing of the technical stakeholders participants). of the assessment. There is a risk that Participants in the meeting with community the scores can be taken as precise and representatives were briefed on the not placed within the local context. discussions with technical stakeholders. Further, they are heavily influenced No major gaps or disparities were found by seasonality and current conditions. between the views of the two groups. The As a result, an assessment undertaken major feedback on the data presented is in a few years might not provide a included here, to help add understanding sense of whether or how resilience and context to the findings. has improved, as it could be affected equally by the set of unique local • Participants felt that the ranked conditions of that period, rather than resilience characteristics resonated truly reflecting how conditions have with the reality in these communities. changed over time. Generally speaking there were • The meeting concluded that the no surprises, and the statements assessment results could be usefully prioritized by the communities were adopted by identifying ‘keystone’ confirmed by both the community indicators for monitoring. CoBRA representatives and technical highlighted a few resilience statements stakeholders as what they expected. that the communities in Turkana They also felt that the list was county largely view as common adequately exhaustive. Meeting priorities, which can be monitored participants agreed that many of the independently as a small set of key highly ranked statements reflected markers of resilience. These indicators issues that were not addressed could also be compared to standard
82 Understanding Community Resilience data sets, such as poverty indicators the means to achieve higher income and Household Economy Analysis and acquire more assets, and hence data, to identify the gaps in long-term were key driving factors to reach a monitoring frameworks on building resilient status. resilience. • The level of attainment of resilience • Participants felt the CoBRA results characteristics was consistently would enable the county government low across all groups even at the to better understand community time of the assessment, which was perspectives and needs. They considered normal or above normal suggested that the time was ripe to in terms of precipitation. Overall, for explore methods to incorporate the all livelihood zones, resilience scores findings into the county strategic in the current period (attributed as a planning process. normal year) hovered around 2.5 out of 10 in terms of attainment. During a 5. Conclusions and crisis period the scores of all resilience recommendations characteristics typically fell to half this, 1.3 out of 10 on average. The woefully low scores in all characteristic 5.1 Conclusions areas may reflect the high levels of The evidence presented above is underlying vulnerability of all groups extensive and adds a significant amount in Turkana. The scores do not vary of understanding of the community much in areas with different levels of perspective on resilience. The following intervention. Fishing communities conclusions are drawn from the evidence score themselves much lower than the presented. other livelihood groups. • There was a general level of optimism • Education emerged as the most widely that the number of resilient households cited characteristic of resilience. With is increasing. This optimism is most a score of 763, education remained marked among youth in urban areas among the top three characteristics and groups already living in areas across all age, gender and livelihood with a high level of services, where 58 groups. The other top characteristics to 73 percent felt that resilience was were water for humans (396), diversified increasing. IGAs (375), peace and security (367), • There was significant overlap and access to credit (287). between interventions that were • When asked to describe the specific most highly rated for helping to characteristics of resilient households build resilience and the most highly in their community, focus group rated characteristics of resilience: participants consistently emphasized education, water and credit/business income and assets. Resilient support. Resilient households cited households were described as having a similar set of priority interventions, multiple income sources, most typically but they ranked access to credit and a diversified IGA combined with a business skills and irrigation much primary livelihood activity such as crop higher. It would seem that communities agriculture, pastoralism and/or fishing. in Turkana are keen to expand and This was consistent with findings from diversify their livelihoods and move the KIIs with resilient households, beyond sole reliance on pastoral and who cited access to multiple income rain-fed agricultural income sources. sources as the primary reason for their Education, irrigation and credit are resilience. Access to education and seen as central requirements for this credit were strongly linked with having transformation.
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 83 5.2 Recommendations directly or indirectly. Diversified IGAs, Despite the significant variations in employment opportunities and large livelihoods and topography in Turkana, herds were consistently and uniformly communities repeatedly prioritized used to describe resilient households. a relatively small set of resilience When these households were asked characteristics. These included education, about the pathways that led to water, diversified IGAs, peace and security, resilience, they mentioned access to and access to credit. Importantly, these education, credit, markets, etc., as indicators encompass both long-term and the factors that helped them reach a short-term issues and include factors that diversified income base. are not always immediately considered a • The community perspective must part of disaster risk reduction strategies, be incorporated into the disaster as well as factors that require a long-term risk reduction/resilience agenda. commitment to investment. Thus: The assessment findings show that communities are highly aware of the • A broader conception of disaster risk long-term and short-term factors reduction is required if resilience is to that contribute to or undermine their be built. Specifically: resilience. Too often consultations with - Investment in expanding access communities focus on interventions to higher level education must that are already designed by ‘technical be a priority. Increasing access to experts’ approved by ‘decision- education will result in significant makers’ in a top-down manner or capital and revenue costs, but it is for which budget is available. Failure a prerequisite if communities are to to respond to local contexts and expand and diversify their income integrate the needs and priorities of sources, which appears to be a high the beneficiaries may partially account priority. for the fact that the districts where - Water requires continual the assessment was conducted have investment, particularly to ensure been facing drought crises repeatedly that communities can access water every few years despite the level more reliably throughout the year. of humanitarian and development - Parts of the county need more focus support provided in the region. on peace and security, as conflict Stakeholders involved in building fundamentally undermines progress resilience need to be aware of the and development. Communities wider spectrum of factors that affect seem to lack understanding about resilience and consider how they can how to tackle insecurity or who re-focus their (or others’) efforts and should do it. Clearly efforts to date budgets in these areas. appear to have had limited impact. • Continuous monitoring and evaluation - In the short term, less costly is critical to measuring changes interventions, such as access to in resilience. The consistency in credit and business skills, could be statements by communities suggests prioritized. However it should be that a few key indicators could be borne in mind that the success such identified to monitor resilience interventions could be handicapped more systematically. Some of these if longer term interventions, as indicators are already being measured mentioned above, continually fail as part of data collection exercises, to be developed in tandem. such as household income sources and • Planning and implementation of levels. Agreeing on an appropriate future interventions should be based indicator and method of collection, on the extent to which they build and such as for peace and security, would diversify incomes and assets either require some level of consensus.
84 Understanding Community Resilience Appendix 1. Participants in CoBRA Turkana assessment
No. Name Organization Phone no. Email address
1 Emmanuel Kisangau NDMA 0724 005 104 [email protected]
2 Paul K. Njuguna Ministry of Agriculture 0710 911 644 [email protected]
3 Ladan Muvea Ministry of Education 0721 703 714
4 John Emal Ekaran Ministry of Health 0721 698 316 [email protected]
5 James Atana Ministry of Livestock 0727 466646 [email protected]
6 Ambrose Lore Oxfam 0721 886 648 [email protected]
7 Moses Nawoton Oxfam 0713891308 [email protected]
8 Josephat Toto Veterinarian Sans Frontiers 0726 610 578 [email protected] (VSF) -Germany
9 Mercyline Shena VSF- Belgium 0728 458399 [email protected]
10 Francis Anno VSF- Belgium 0728 853 070 [email protected]
11 Alice Emathe Turkana Pastoralist 0723 373 197 [email protected] Development Organization
12 Gabriel Ekaale Elato Child Fund 0724 630 888 [email protected]
13 Serah Pedo Friends of Lake Turkana 0729 888611
14 Collins Hawala Kenya Red Cross Society 0717 480 191
15 Elizabeth Lokolio Community representative 0726815707 [email protected]
16 Brenda Mana Social Worker 0714 727029 [email protected]
17 Anthony Njengi International Rescue 0727 965 013 [email protected] Committee
18 Johnson Ngikol United Nation Volunteer 0718 066336 [email protected]
19 Erenius Nakadio Lutheran World Federation 0720 931191 [email protected]
20 Gregory Ikakoro Lokando 0714 953962 [email protected]
21 Solomon Koech DanChurchAid 0722 438 244 [email protected]
22 Laban MacOpiyo UNDP DDC 0728 466 622 [email protected]
23 Catherin Fitzgibbon UNDP DDC 0734498324 [email protected]
24 Francis Opiyo UNDP DDC 0727 35 3670 [email protected]
25 Yuko Kurauchi UNDP DDC 020 7624509 [email protected]
26 Agnes Ndegwa UNDP DDC 020 762 4640 [email protected]
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 85 Appendix 2. Complete resilience statements and scores for CoBRA Turkana
Resilience Bean Total bean Full resilience statement11 characteristics score score
Financial
Diversified IGAs Many households would be involved in other IGAs, 375 1,162 such as small businesses and trading.
Access to credit People would have good access to affordable 287 credit and would be saving money (through banks, microfinance institutions, community savings and credit).
Employment There would be many opportunities for jobs and other 135 forms of paid employment through government, factories and other businesses.
Livestock herds Pastoralists would have herds large enough to 125 sustainably support their families.
Health care for The community would have access to high-quality and 95 livestock affordable animal health services, including veterinary services and vaccinations, whenever they need them.
Land Everyone would have secure access to/ownership of 66 land/property.
Productive farms Farmers would be more productive and profitable 41 (i.e., they would have inputs like quality tools, oxen, fertilizers and improved knowledge of good farming practices).
Cash transfers Households would receive cash transfers. 17
Fishing inputs 16
Transport 3
Remittances 2
Human
Education All children would be able to complete primary/ 763 1,170 secondary/tertiary education.
Health care for The community would have access to high-quality and 231 humans affordable basic health care locally.
Food security All households would be able to feed themselves well 173 every day.
Skills The community would have the skills and structure to 3 plan and implement solutions to their own problems.
Natural
NRM Local rangelands and other natural resources would 118 118 be well managed so they do not become degraded over time; trees and forest cover would be well and sustainably managed to provide for future generations.12
11 Note that blanks indicate statements that were added by communities during focus group discussions, and as such did not have a generic full statement associated with them. 12 Note that several statements relating to NRM were merged for natural capital.
86 Understanding Community Resilience Resilience Bean Total bean Full resilience statement11 characteristics score score
Physical
Water for humans The whole community would have access to sufficient, 396 1,295 good-quality water at all times of the year.
Access to markets The community would have easy access to markets to 265 buy goods and sell their produce.
Irrigation Farmers would be irrigating land to improve the 260 production of crops for consumption and sale.
Roads There would be good-quality roads to the community. 109
Water for livestock Livestock would have access to sufficient water at all 80 times of the year.
Electricity The community would have access to affordable 58 electricity supply.
Telecommunications There would be a reliable mobile phone network to all 54 communities all the time.
Sanitation Everyone would have good sanitation. 33
Shelter Everyone would live in good-quality housing. 27
Water [Human or livestock use not specified] 13
Social
Peace and security The whole community would enjoy continual peace 367 437 and security.
Community The community would have the skills and structures to 23 empowerment plan and implement solutions to their own problems.
Women Women would be fully involved in local development 28 and leadership.
Good governance The community would be served by efficient and 18 non-corrupt community leaders and management structures.
Wives Men would have many wives (dowry). 1
Turkana North, Turkana West, Loima, Turkana Central, Turkana South and Turkana East districts 87 Appendix 3. Participants in CoBRA Turkana technical stakeholders feedback session
No. Name Organization Email address
1 Julius Taigong NDMA [email protected]
2 Emmanuel Kisangau NDMA [email protected]
3 Brenda Mana NDMA [email protected]
4 Pius Eiton NDMA [email protected]
5 Lorungole Edna NDMA
6 John Elim NDMA
7 Paul K. Njuguna Ministry of Agriculture [email protected]
8 John Emal Ekaran Ministry of Health [email protected]
9 James Atana Ministry of Livestock [email protected]
10 Mercyline Shena VSF- Belgium [email protected]
11 Josephat Toto VSF-Germany [email protected]
12 Alice Emathe Turkana Pastoralist Development [email protected] Organization
13 Gabriel Ekaale Elato Child Fund [email protected]
14 Serah Pedo Friends of Lake Turkana [email protected]
15 Collins Hawala Kenya Red Cross Society [email protected]
16 Elizabeth Lokolio Community representative [email protected]
17 Anthony Njengi International Rescue Committee [email protected]
18 Johnson Ngikol UN Volunteer [email protected]
19 Victor Lekaram Ministry of Planning [email protected]
20 Simon Juma Ministry of Planning [email protected]
21 Lubanga Makanji Japan International Cooperation Agency [email protected]
22 Rukia Abubakkar Kenya Red cross Society [email protected]
23 Nancy Lomear Agency for Pastoralist Development [email protected]
24 Isaac Aremon Turkana Pastoralist Development [email protected] Organization
25 Fredrick Merie World Food Programme [email protected]
26 Rose Muhindi Trade [email protected]
27 Laban MacOpiyo UNDP DDC [email protected]
28 Francis Opiyo UNDP DDC [email protected]
88 Understanding Community Resilience Annex 3 Community-based Resilience Analysis Assessment Report Kotido and Kaabong Districts Karamoja Sub-region, Uganda
Commissioned by UNDP Drylands Development Centre
Under the framework of the European Commission Directorate General for Humanitarian Aid and Civil Protection’s Drought Risk Reduction Action Plan
Kotido and Kaabong districts 89 Contents
1. Introduction...... 91 2. Approach...... 91 2.1 Characteristics of field site...... 91 2.2 Data collection ...... 92 2.3 Constraints and limitations of data collection process ...... 93 3. Findings...... 95 3.1 Main hazards or shocks ...... 95 3.2 Characteristics of a resilient community...... 95 3.3 Extent to which the community has achieved resilience...... 101 3.4 What a resilient household looks like...... 105 3.5 Interventions that contributed to household resilience...... 106 3.6 How key informants achieved resilience ...... 107 4. Summary of feedback from local and national consultations...... 109 5. Conclusions and recommendations...... 112 5.1 Conclusions...... 112 5.2 Recommendations...... 113
90 Understanding Community Resilience 1. Introduction This report outlines findings of the Kotido/Kaabong CoBRA assessment. The comprehensive Community-Based It also incorporates key feedback and Resilience Analysis (CoBRA) assessment consolidated inputs generated at was undertaken in the districts of Kotido review and validation workshops of the and Kaabong, Karamoja sub-region, draft assessment report by community Uganda, from 15–26 July 2013. The representatives in Kotido on 29 October exercise was carried out with financial 2013 and local government and non- support from the European Commission governmental technical stakeholders Directorate General for Humanitarian Aid working in the area in Moroto on 31 October and Civil Protection under the framework 2013. The report also summarizes the of its Drought Risk Reduction Action Plan. comments and recommendations made This assessment builds on an initial field at the ‘National Workshop on Enhancing trial of the draft CoBRA tool undertaken in Community Resilience: Learning from the late 2012. CoBRA’, which was convened jointly by OPM and UNDP on 28 November 2013 in The multi-agency assessment was jointly Kampala. led by the United Nations Development Programme (UNDP) Drylands Development A detailed explanation of the conceptual Centre and the Office of the Prime Minister framework that underpins the methodology (OPM). Logistical and practical support was is contained in the CoBRA Conceptual provided by the Drylands Development Framework and Methodology document.1 Centre. A wide range of international and local NGOs operating in the area 2. Approach also participated in the assessment by providing local staff as facilitators to undertake CoBRA training and fieldwork. 2.1 Characteristics of field site A list of agencies that participated in the The field assessment took place in Kotido CoBRA training or the field data collection and Kaabong districts of Uganda, in is provided in Appendix 1. the country’s northeast near the border with Kenya. The population of these The CoBRA assessment has four broad two districts represents 49 percent of objectives: the Karamoja sub-region and all the livelihood zones in the sub-region. This 1. Identify the priority characteristics of enabled sampling of parishes in pastoral,2 resilience for a target community; agropastoral, agricultural and urban/ 2. Assess the communities’ achievement peri-urban livelihood zones. A total of 18 of these characteristics at the time of parishes were sampled for the assessment the assessment and during the last (Figure 1). crisis or disaster; 3. Identify the characteristics and Kotido and Kaabong districts cover more strategies of existing resilient than 20,100 square kilometres and face households; and a multitude of human development and 4. Identify the most highly rated environmental challenges. They have interventions or services in building the lowest development indicators in the local resilience. country, with more than 80 percent of
1 www.youblisher.com/p/784724-Community-Based-Resilience-Analysis-CoBRA-Conceptual-Framework-and-Methodology. 2 Pastoral livelihood groups include those households whose income sources are mainly (more than 90 percent) livestock based.
Kotido and Kaabong districts 91 Figure 1. Surveyed parishes in Kotido and Kaabong districts
the population living below the poverty experiences livestock disease outbreaks; line.3 Average population density in the limited access to agricultural and livestock two districts is just under 50 people per inputs and markets; persistent cattle raiding; square kilometre. Resident tribes include and inter-communal conflicts. The districts the Dodoth, Jie, Minig and Nyagea, who face widespread gaps in access to basic practise agropastoral livelihoods, and services of health, water and education, the Napore and Ik, who are engaged in along with poor road infrastructure and agriculture. limited trade opportunities.
Kotido and Kaabong occupy a semi-arid 2.2 Data collection zone characterized by recurrent long dry spells and erratic rainfall. The area Full details of the methodology used is chronically food insecure and often to undertake a CoBRA assessment are
3 United Nations Office for the Coordination of Humanitarian Affairs, Uganda Humanitarian Profile, 2011.
92 Understanding Community Resilience included in the main CoBRA assessment of the study coincided with inflated food report4 and implementation guidelines. prices and elevated food insecurity due Data were collected using qualitative to poor crop performance.5 Accordingly, methods, i.e., focus group discussions some focus group participants feared (FGDs) and key informant interviews (KIIs). to be associated with wealth and were reluctant to be identified as resilient since Field work was undertaken by a team of 15 such households are at times excluded facilitators and 3 supervisors. The group from food assistance. Nevertheless, the was divided into three teams of five people facilitators managed to clear most doubts each, comprising a team supervisor and by organizing prior community mobilization two pairs of facilitators. The facilitators and exercises and explaining the purposes of supervisors, who were familiar with the the assessment to the participants. local context, jointly selected the locations and compositions of focus groups, Changing livelihoods using statistical data and criteria such as Livestock holdings in Kotido and Kaabong livelihood zones and level of interventions. districts have generally fallen over the Whenever possible, separate focus groups years, mainly because of cattle raids, were organized for men, women and youth recurrent drought and livestock diseases. (aged 15–30, both sexes) to ensure that The introduction of ox ploughs has views on resilience from different gender increased the acreage devoted to crop and age groups would be adequately farming as well as yields. Thus, some captured in the discussions. Each group communities that used to be pastoral and had around 15 participants. have lost their livestock appear to have transitioned to crop farming, despite the In each focus group location, one to four climatic challenges. They have acquired KIIs were undertaken with households ploughs and started large crop gardens, perceived to be resilient, as identified which made it difficult for facilitators to by focus group participants during their identify the appropriate livelihood in some discussions. In some cases, the households parishes. Consequently limited weight were also identified as resilient through can be given to analysis of findings by conversations with local chiefs or other livelihood zone as many ‘pastoralists’ are senior residents. A total of 36 FGDs in reality now agropastoralists. took place, along with 40 KIIs. Table 1 summarizes the number and locations Rain and poor road infrastructure of FGDs and KIIs undertaken for this Rain coupled with poor road infrastructure assessment. often restricted the mobility of the assessment team, especially in Kaabong 2.3 Constraints and limitations of district. Consequently, some FGDs and KIIs had to be rescheduled. data collection process Reluctance about being identified as Gender representation resilient There was poor representation of women By its very nature, knowledge about in the assessment team as most of the wealth systems is sensitive, and people are local partners provided male participants. secretive about their wealth. The timing Of 15 facilitators, there was only 1 woman.
4 UNDP Drylands Development Centre, Understanding Community Resilience: Findings from Community-Based Resilience Analysis Assessments – Marsabit, Turkana and Kajiado counties, Kenya, and Karamoja subregion, Uganda (Nairobi, UNDP Drylands Development Centre, 2014). 5 World Food Programme and Food and Agriculture Organization of the United Nations, Karamoja Rapid Crop and Food Security Assessment, August 2013.
Kotido and Kaabong districts 93 Table 1. FGDs and KIIs undertaken for Kotido/Kaabong CoBRA assessment
No. of Focus group No. of District Parish Population Livelihood FGDs composition KIIs
Male elders, Kalapata 6,900 Pastoral 2 2 male youth
Females (all ages); Kathile 15,900 Agropastoral 2 2 Mixed youth
Females (all ages); Lolelia 10,000 Agropastoral 2 2 Mixed youth
Females (all ages); Loteteleit 9,800 Agropastoral 2 2 Mixed youth
Females (all ages); Loyoro Napore 7,700 Agriculture 2 2 Female youth
Females (all ages); Narengepak 8,200 Agropastoral 2 2 Mixed youth Kaabong Male elders; Lomeris 13,200 Pastoral 2 2 Male youth
Male elders; Lobongia 12,700 Pastoral 2 2 Male youth
Female elders, Kamion 5,800 Agriculture 2 2 Female youth
Male elders; Kathimeri 6,100 Pastoral 2 2 Male youth
Females (all ages); Kaabong Central 3,000 Urban 2 2 Mixed youth
Male elders; Kangigetei * Pastoral 2 2 Male youth
Males (all ages); Narikapet 3,700 Peri-urban 2 2 Females (all ages)
Female elders, Lokitelakebu 9,600 Peri-urban 2 2 Female youth
Mixed elders; Lochedimeo * Agropastoral 2 2 Mixed youth
Mixed (all ages); Kotyiang 6,000 Agropastoral 2 2 Female elders Kotido Female elders, Kamoru 20,300 Agropastoral 2 2 Female youth
Female elders, Loposa 21,400 Agropastoral 2 2 Female youth
Kanawat 17,300 Agropastoral - - 1
Kotido Central 2,800 Peri-urban - - 1
Lokadeli 6,300 Agropastoral - - 2 Total 160,300 36 40
* Census figures are not available for these parishes. Note: Empty cells indicate locations where no FGDs took place.
94 Understanding Community Resilience 3. Findings decade in several parishes. Flood was also reported; it affects fewer people but This section reports on the summarized scores high in severity. To a limited extent, findings from the CoBRA field work the communities reported the hazards of conducted during July 2013 in Kotido and livestock/crop diseases, human diseases, Kaabong districts in Karamoja sub-region. animal-human conflict and wildfire. Specifically, the findings are presented according to the following categories: To a large extent and compared to these crisis periods, the communities viewed • What are the main hazards or shocks current conditions as “normal”. However, facing the communities? some communities saw the situation as • What are the characteristics of a still “bad” due to the dry spell, which was resilient community? resulting in reduced crop yields and food • To what extent has the community shortage as well as some crop diseases. achieved resilience characteristics? • What does a resilient household look 3.2 Characteristics of a resilient like? community • What interventions contributed to household resilience? Focus group participants were asked • What additional interventions would to describe what they viewed as the best build resilience? characteristics of a resilient community. In • How did key informants achieve and the following subsections, the data are first maintain resilience? presented for the entire set of respondents to give an overall picture of the most highly 3.1 Main hazards or shocks rated statements. The results are analysed further by category of respondent – Overall, drought and associated food gender/age, livelihood group and level of shortages, malnutrition and famine were intervention in the community – which are perceived by the focus groups as the most used to disaggregate findings and identify common hazards facing the districts and differences across groups. affecting most if not all the households in the communities. Participants cited the Analysis – all respondents drought of 2010–2011 most frequently Focus group participants were asked and agreed it was the main crisis period to to identify and rank statements used be considered in the CoBRA assessment. to describe a resilient community. Each While nearly 80 percent of the population member was given six beans to rank the in the country experienced some degree three most significant statements (three of food insecurity in 2011, Kotido and beans for the most significant statement, Kaabong districts were classified as two for the second most significant and emergency in terms of the Integrated Food one for the third) in terms of priority for Security Phase Classification.6 Another building resilience. The bean scores were hazard reported was conflict (largely then totalled for each statement. For clan-based livestock raiding), which has ease of comparison, the statements were decreased in impact, but nonetheless grouped into the five sustainable livelihood was experienced constantly in the past framework (SLF) categories.7
6 Uganda, Uganda Humanitarian Profile 2012. 7 The sustainable livelihood framework presents the main factors that affect people’s livelihoods and typical relationships between them. It identifies five core asset categories or types of capital upon which livelihoods are built: financial, human, natural, physical and social. For further details on SLF, please refer to: UK Department for International Development (DFID), Sustainable Livelihoods Guidance Sheets (London, DFID, 1999).
Kotido and Kaabong districts 95 Table 2. Community ranking of resilience characteristics by SLF category
SLF category Top-ranking resilience characteristics (‘bean scores’) Total bean score* Financial Productive farms (762) 1,868 Access to credit (341) Livestock herds (320) Employment (147) Health care for livestock (127) Diversified income-generating activities (IGAs) (120)
Human Education (586) 1,006 Health care for humans (315) Food security (92)
Natural None – very low scoring 17
Physical Water for humans (284) 891 Sanitation (127) Access to markets (120) Water for livestock (119)
Social Peace and security (411) 425
* Total bean score does not match the sum of the bean scores in the middle column because the middle column reflects only the top-ranking characteristics while the total in the right column reflects all the characteristics discussed in each category.
Table 2 lists the 14 most highly ranked list of the characteristics identified by the statements used to describe a resilient communities with scores and expanded community within each of the five SLF statements is provided in Appendix 2. categories. (Note that many more statements/characteristics were included Figure 2 shows the highest scored in the ranking, but were given low scores characteristics used to describe a resilient and hence are not reported here.) The full community by all focus group participants
Figure 2. Top-ranking resilience characteristics – all respondents
900 800
700
600
500
400 300
200
100 0
Education Sanitation
Employment Food security Livestock herds Access to credit Diversified IGAs Productive farms Water for humans Peace and security Accexx to markets Water for livestock
Health care for humans Health care for livestock
96 Understanding Community Resilience Figure 3. Top-ranking resilience characteristics by SLF category – all respondents
2000
1800 Others Water for livestock 1600 Access to markets Sanitation 1400 Water for humans Natural resources management 1200 Health care for humans Education 1000 Diversified IGAs Health care for livestock 800 Employment Livestock herds 600 Access to credit Productive farms 400 Peace and security
200
0 Social Financial Human Physical Natural in the two districts in the order of bean districts are transitioning away from scores. Figure 3 presents the total bean livestock rearing as the main source of score under each of the five SLF categories. livelihoods due to challenges such as the outbreak of transboundary animal In terms of the characteristics, productive diseases,8 which has led to the loss of farms, education and peace and security thousands of livestock. received the highest bean scores. In terms of the SLF categories, characteristics Internal and external conflicts have led addressing financial capital and capacity the communities in Kotido and Kaabong ranked most highly, followed by human districts to gradually shift to crop production and physical capital. Social and natural as their main source of livelihoods. Peace categories received lower scores. and security have recently been realized in most parts of Karamoja, so communities The stakeholder review and validation have also expanded their settlements to meeting in October 2013 deliberated previously less cultivated areas, including on these findings and confirmed that the Lobonya, Karwakol, Lolelia, Moruitit, high ranking of productive farms as the Nyarkidi, Kottidany and Adual, for the priority resilience characteristic represents purpose of expanding sedentary farming. the local reality. It reflects the ongoing diversification of livelihood options in the Favourable climate patterns, with more region, in particular the transition from reliable and frequent rains in the last pure pastoralism to agropastoralism or four years, have also encouraged more crop farming for increased food security. crop production in the greenbelt areas Communities in Kotido and Kaabong of Karamoja sub-region, which include
8 Diseases mentioned include peste des petits ruminants, contagious bovine pleuropneumonia, contagious caprine pleuropneumonia, East Coast fever and Brucellosis.
Kotido and Kaabong districts 97 Kaabong and Kotido districts. These characteristics, participants in the review circumstances account for growing and validation workshop shared the insight demand for agricultural inputs such as that food security is indirectly represented improved seeds, drought-resilient plant in the other characteristics. Communities species, management technology and feel that attaining the other key factors – fertilizers. having productive farms, education, peace and security, health, natural resources, The high ranking of education was also roads, markets, etc. – contributes to overall confirmed as extremely relevant. The food security in the long run. Nonetheless, communities understand that education is a a recommendation was made to analyse means to improve livelihood diversification further the food-related community and access to basic services in the long statements to better understand local run. This is particularly applicable in view food security priorities in terms of access, of the pastoralists’ shift to other economic availability, quality, etc. activities. Analysis by gender and age With regard to the low ranking of The priority resilience statements were also food security in the list of resilience analysed by gender and age group, through
Figure 4. Top-ranking resilience categories by gender/age group
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Women Men Youth
Table 3. Top-ranking resilience characteristics groups comprised of women, men and by gender/age group* youth (aged 15–30, both sexes). The score allocated to each of the five SLF categories Gender/ Top 3 resilience Total age group characteristics score** was taken as a percentage of the total bean score, and the results are represented in the Women Productive farms (254) 1,310 bar chart in Figure 4 to demonstrate the Education (249) differing priorities that men, women and Livestock herds (112) youth place on resilience statements. Men Productive farms (142) 899 Peace and security (124) The most highly ranked resilience Education (123) characteristics and their scores by gender/ Youth Education (227) 1,798 age group are presented in Table 3, which Access to markets (119) shows how highly both women and men Access to credit (113) rate productive farms. Men also rank peace *Results from the mixed focus groups that do not fit the categories and security in their top three, possibly of male, female and youth were excluded from the analysis. reflecting the fact that men are primarily **Total score does not match the sum of the scores in the middle column because the middle column reflects only the top-ranking involved in herding and therefore are more characteristics while the total in the right column reflects all the likely to be affected by cattle rustling and characteristics discussed in each category.
98 Understanding Community Resilience tribal clashes. Youth rank education and children. In this context, men are financial factors highly. This may reflect better positioned to appreciate the aspirations to move away from pastoral current level of peace and security and and agricultural livelihoods. to stress the importance of maintaining social stability as a driver for attainment The review and validation workshop of all other characteristics of resilience. provided further insight on the results as follows: Youth
Women • Education: Young people envisage a brighter future with higher education • Productive farms: Women’s high in that it should provide them with priority on productive farms reflects more opportunities and livelihood the division of labour, as women are options. Educated youth in the districts more involved in farm work than men. are seen as role models and agents of Women also have greater access to change. Awareness on the importance and control of agricultural produce as of education is also increasing with they are responsible for feeding the support from the Government and family. other partners. Additionally, culture and • Education: Women are widely seen as tradition are gradually giving way to prioritizing education as a long-term modern life in Karamoja. For example, component of resilience, as it leads electricity finally arrived in Moroto in to better employment opportunities May 2013 and has had a great impact that can supplement household on the lives of many Karamojong. livelihoods. • Access to markets: Youth are more • Livestock: Having large numbers of proactive in commercialization, so are livestock is still seen as an important naturally more inclined to see market coping strategy in the districts during access as a component of resilience. disasters. • Access to credit: Credit is a benchmark for increased commercialization. Men Analysis by livelihood group • Productive farms: Sole dependence Resilience statements were also analysed on livestock is risky, and the for three livelihood groups – agropastoral, need to transform and diversify pastoral and peri-urban. Figure 5 illustrates livelihoods has been strongly felt the differences in bean score allocation by in the districts. Agricultural crop the five SLF categories among the three production is increasingly perceived livelihood groups. as a direct source of livelihoods in terms of household consumption Further to this, the most highly ranked and commercialization, as a result resilience characteristics by livelihood of awareness-raising campaigns groups are presented in Table 4. undertaken by the Government and NGOs, particularly in recognition of The data suggest the following: recent favourable rainfall patterns. • Peace and security: Men are • Pastoral and agricultural groups traditionally involved in raiding and place far greater weight on financial fighting with other ethnic groups and characteristics, whereas peri-urban thus are direct victims of insecurity, and agropastoral groups give more while women tend to remain home weight to the physical characteristics to take care of the household and of resilience.
Kotido and Kaabong districts 99 Figure 5. Top-ranking resilience categories by livelihood group
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Agropastoral Pastoral Peri-urban Agricultural
Table 4. Top-ranking resilience characteristics following basic services and interventions, by livelihood group which were mapped out in all the surveyed parishes in Kotido and Kaabong districts: Livelihood Top 3 resilience Total group characteristics score* • Number and level of education Agropastoral Productive farms (452) 2,411 facilities; Education (346) • Police, prisons, army barracks; Health care for humans (207) • Number and level of health facilities; Pastoral Productive farms (200) 996 • Tarmac road; Livestock herds (156) • Other main road; Peace and security (120) • Well-functioning livestock market; Agricultural Productive farms (67) 400 • Water supply; Access to credit (61) • Savings and credit programmes; Education (49) • Cash transfers; • Mobile money services; Peri-urban Peace and security (78) 400 • Banks/banking services; and Education (76) Access to credit (49) • Mobile phone coverage.
* Total score does not match the sum of the scores in column 2 because column 2 reflects only the top-ranking characteristics while Excluded from the mapping exercise were the total reflects all the characteristics discussed in each category. interventions that are universally provided in all the parishes, such as food aid, or • Agropastoral, pastoral and agricultural provided only at very low scale, such as to groups all rank productive farms as less than 500 beneficiaries. The parishes the most important characteristic were then divided into three even groups: of resilience, whereas peri-urban households focus on peace and • The bottom third represents low security. It is interesting to note that intervention areas, where up to agricultural and peri-urban households four categories of interventions are rank access to credit as the second implemented. most important characteristic of • The middle third represents resilience, while this does not come up medium intervention areas, where with the other groups. five categories of interventions are implemented. Analysis by intervention level • The top third represents high Consultations with the local CoBRA intervention areas, where six or assessment team and government more categories of interventions are line departments were used to map implemented. accessibility to and presence of the
100 Understanding Community Resilience Figure 6. Top-ranking resilience categories by level of intervention
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% High Medium Low
Figure 6 shows the differences in scores Table 5. Top-ranking resilience characteristics according to the five SLF categories by intervention level among the three location groups with different services/intervention levels. Intervention Top 3 resilience Total level characteristics score*
Further to this, the most highly ranked High Productive farms (433) 2,406 resilience characteristics by group are Education (332) presented in Table 5. Health care for humans (291)
Medium Productive farms (123) 1,202 The data suggest the following: Education (77) Livestock herds (77) • All three groups consistently place the greatest emphasis on financial Low Productive farms (206) 599 characteristics of resilience. Education (176) Health care for humans (121) • The weightings are more or less consistent across intervention groups, * Total score does not match the sum of the scores in the middle column because the middle column reflects only the top-ranking though low intervention areas tend characteristics while the total in the right column reflects all the to put more emphasis on human characteristics discussed in each category. characteristics of resilience than do the medium and high intervention groups. universally identified as the drought period • While all three groups rank productive of 2010–2011). farms and education as their top characteristics of resilience, high The following subsections present the and low intervention groups ranked findings first for all respondents and then human health in the top three, while disaggregated by specific livelihood medium intervention groups ranked groups. The analysis by gender is not livestock. included in this section, because focus groups were asked to rank the attainment 3.3 Extent to which the community of the resilience characteristic statements for the entire community, and therefore has achieved resilience any differences between men and women Focus group participants were asked in the same community would be based to score the extent to which they had on perceptions. achieved their priority characteristics of resilience. They were asked to score each Figure 7 shows a spider diagram with statement twice, first for the current period several rings. The outer ring represents a (agreed to be a normal period) and second perfect or ideal score for all statements in for the last significant crisis period (almost that SLF category. The blue band shows
Kotido and Kaabong districts 101 Figure 7. Community attainment of resilience per SLF category – all respondents
Physical 10
8
6
4 Social Human 2
Current year rank 0 Crisis year rank Full score
Natural Financial
Note: Possible score ranges from 0 to 10.
communities’ average attainment of those category was dominated by responses characteristics in the current period, and relating to peace and security, and the the red line represents the last crisis period. high ranking reflects the benefits these The scores are ranked on a scale from 0 to communities have gained from the recent 10, with 10 reflecting perfect attainment of period of much higher social stability that characteristic (for example, the entire and safety. However, these scores could community has access to sufficient, good- quickly deteriorate if the security situation quality water at all times during a calendar deteriorates. year), and 0 reflecting no attainment (no one in the community has access to On the whole, there are distinct differences sufficient, good-quality water at all times in the resilience characteristic attainment of the calendar year). scores between the current period (perceived to be normal) and the crisis Table 6 presents the scores by SLF period (the 2010–2011 drought), indicating category for the top-ranked characteristics the communities are highly vulnerable to of resilience. Overall in a normal period, shocks. In particular, the dynamic nature community members rank their attainment of the 2010–2011 drought, which was of characteristics of resilience on average characterized not only by severe rainfall 4.8 out of 10, as opposed to 1.9 during deficits but also by occasional flash floods the crisis period. Clearly, figures should be and sporadic ethnic clashes, may partially viewed with some caution, as these scores account for the large gaps. The scores represent community perceptions and may also reflect the need for more timely have no statistical significance, and they support in response to disasters in the could overstate or understate reality. districts at the onset of climatic hazards, before the impacts of shocks undermine Social characteristics of resilience have resilience characteristics. the highest score (7.1). In other words, communities ranked social characteristics After peace and security, the characteristics as the area where they have the greatest most affected by shocks are water for degree of attainment of resilience. This livestock, sanitation, education and
102 Understanding Community Resilience Table 6. Community attainment of resilience by SLF category – top-ranked characteristics
Current period Crisis period score SLF Top-ranked characteristics score (2010–2011 category (July 2013) drought)
Overall average 4.8 1.9
Financial Productive farms 4.6 2.3
Access to credit 3.7 1.3
Livestock herds 3.6 3.8
Employment 4.0 1.3
Diversified income-generating activities (IGAs) 4.9 2.4
Human Education 5.6 2.2
Health care for humans 4.7 2.0
Food security 4.2 2.4
Physical Water for humans 4.6 1.4
Water for livestock 5.2 1.6
Sanitation 5.1 1.6
Access to markets 5.2 2.3
Social Peace and security 7.1 2.2 markets. Drought clearly affects the The review and validation meeting in availability of water for livestock; however October 2013 confirmed the accuracy of sanitation and markets are often more the current high attainment rate of peace highly affected by floods. Education and security characteristics (score of 7.1 (scores reflect school attendance) can overall). The participants noted that the be affected by all shocks, including price score can possibly be higher in some inflation, which reduces household ability areas. The police presence in almost every to pay education fees and other costs. subcounty and the highly visible Uganda Peoples’ Defence Forces patrolling The characteristics least affected by shocks roads and borders have contributed are livestock herds, food security, diversified tremendously to improving the security IGAs, employment and health. While in Karamoja, justifying the score. Effective some of these characteristics may be less community mobilization for conflict dependent on weather, the chronically low resolution and peace building, led by attainment rates in both normal and crisis politicians, the local authority and partners, periods may imply the need for long-term was also highlighted as an important driver development support to improve fulfilment for peace. The International Day of Peace of these characteristics in the districts in the was commemorated on 21 September future. For instance, livestock herds are now 2013 in Moroto. Former warriors are now significantly smaller than before; a herd engaged in IGAs, such as brick-making, of more than 10 cattle is now considered food for work, cash for work and quarry, large in the area. This leaves little room for apiary and charcoal businesses. mass livestock loss in a drought. Similarly, food insecurity reflects chronically low As for the attainment of characteristics agricultural production and incomes. related to productive farms, participants
Kotido and Kaabong districts 103 raised concerns about the score attained. IGAs, such as agriculture. Such a shift has They stressed the need to analyse the reduced the number of livestock overall, scores carefully, noting the possibility of hence diminishing supply and demand in bias or dependency syndrome among the livestock sector. the respondents, particularly in view of the overwhelming amount of agricultural Table 7 aggregates the scores by livelihood support provided to these communities by groups and intervention levels. While it the Government and non-governmental clearly masks differences between SLF actors such as OPM, National Agricultural categories, it is useful to provide an overall Advisory Services and Northern Uganda indicator of where different community Social Action Fund. groups see themselves as per their livelihood activities and locations. Views on attainment of the education characteristics were more varied: Some Results by livelihood group show some noted the increase in schools all over the important differences. Peri-urban groups districts, while others pointed out low ranked their resilience characteristics most education quality, such as inadequate highly (5.7). Agropastoral groups also teacher-student ratios. Further analysis gave their resilience a relatively high score may be required to determine what in normal periods (5.3), whereas purely indicator should be measured – rates of pastoral or agricultural groups ranked their enrolment or completion. resilience much lower even during the “normal” period, at 3.9, falling slightly to As for the unique rating of the livestock 3.2 in the crisis period. herd characteristics (showing higher resilience in the crisis period than in the When the scores are disaggregated by normal period) , participants commented level of services and interventions, high that the increased access to markets due intervention areas rank their level of to improved peace and security in the attainment of resilience characteristics recent past (i.e., the normal period) might more highly than medium and low have encouraged the pastoralists to shift intervention areas, though only by a slim from livestock-based livelihoods to other margin.
Table 7. Aggregate resilience scores The joint review and validation meeting deliberated on the gaps in resilience scores Current Crisis year between agropastoral/peri-urban (higher) year rank rank and pastoral/agricultural (lower) livelihood All groups 4.8 1.9 groups. The participants’ insights are as follows: Livelihood group
Agropastoral 5.3 2.1 Agropastoral
Pastoral 3.9 1.6 • The present favourable climate in Peri-urban 5.7 2.1 Karamoja has resulted in good food production. Food crises have been a Agricultural 3.2 0.9 huge challenge in the sub-region over Intervention level the years because of drought and insecurity. Food security in Karamoja High 5.0 1.5 links directly to building resilience. Medium 4.7 2.2 • Government policy puts a strategic focus on crop production. Under Low 4.6 2.9 this framework, agropastoralists are Note: Maximum possible score of 10 the main recipients of increasing
104 Understanding Community Resilience support from the government and that are more resilient compared to others, other partners in Kotido and Kaabong that is, the households that have already districts, as part of the transition from attained many (or all) of the resilience the former purely pastoral practices qualities prioritized. Five characteristics of a to more diversified livelihoods. Many resilient household were cited consistently of them also benefit from free and by focus groups: subsidized farm inputs. • Improved animal health through • Having a large herd (mentioned by 26 vaccination and disease control out of 36 focus groups); has been instrumental in improving • Earning a certain level of income (19); resilience in the Karamoja sub-region. • Having a large farm (18); In addition, restocking interventions • Having a business or IGA (16); and have been led by OPM through the • Having access to employment/wage Ministry of Karamoja Affairs. Initiatives labour (10). by other partners – such as the Peace Recovery and Development Focus groups were asked whether the Plan, Second Northern Uganda number of resilient households was Social Action Fund Project, Karamoja increasing, decreasing or staying the same. Integrated Development Programme Table 8 presents findings for the group as and Karamoja Livelihoods Programme a whole as well as findings disaggregated – have also improved livelihoods in by gender/age, livelihood groups and Kotido and Kaabong districts. intervention level. The disaggregation is based on a small sample size and therefore Peri-urban should be viewed cautiously.
• Exploration of alternative livelihoods Overall, 52 percent of the respondents said and diversification of income sources their resilience was increasing. Women in peri-uban areas was suggested as were more likely to say it was increasing largely responsible for the high scoring (62 percent) than men and youth. by this livelihood group. • Better access to basic services such Peri-urban groups, who ranked their as health centres, schools, transport resilience scores the highest, were also and markets might account for the the most likely to say that resilience was positive outlook, along with access increasing (75 percent). This is probably to information on related livelihood because urban dwellers are more likely opportunities and various supporting to benefit from access to employment, interventions such as credit services markets and diversified IGAs, all of which (e.g., village saving and loan are considered key characteristics of a associations). resilient household.
Pastoral Agropastoral households, who also scored their resilience relatively highly, • Limited opportunity for livelihoods were also very likely to say that resilience expansion through pastoralism in is increasing (67 percent). Opinions in the districts was suggested as an agricultural households were divided, and explanation for the low rating. pastoral groups were much more likely to say that resilience was decreasing. 3.4 What a resilient household Pastoralists’ pessimism is possibly due to the significant decline in herd size in the looks like last decade, which has undermined the Focus group participants were asked to social and financial status that pastoralists describe the characteristics of households once had in Kotido and Kaabong districts.
Kotido and Kaabong districts 105 Table 8. Is resilience increasing, decreasing or staying the same?
Increasing Decreasing Staying the same
All respondents 52% 31% 17%
Gender/age
Women 62% 23% 15%
Men 40% 40% 20%
Youth 47% 35% 18%
Livelihood group
Agricultural 50% 50% 0%
Agropastoral 67% 13% 21%
Pastoral 10% 70% 20%
Peri-urban 75% 25% 0%
Intervention level
High 54% 33% 13%
Medium 50% 33% 17%
Low 50% 25% 25%
When disaggregated by intervention Groups were also asked to list the group, all groups reported similar findings, three additional interventions they felt with more or less half reporting increasing would best build their resilience. Many resilience. communities restated interventions similar to those mentioned in the first list, with the 3.5 Interventions that contributed justification that the current provision or scale of intervention was too limited and to household resilience should be expanded. Communities were asked to list all the services and interventions they had The repeated and clear priority given to benefited from in the last two to five years. water and education interventions reflects A reasonably wide range of interventions the high ranking assigned to these factors was mentioned, falling into the following as characteristics of resilience by all focus categories: water, education, livestock groups. Water interventions were prioritized inputs, agricultural inputs, IGAs, peace by all livelihood groups, not surprising and security measures, roads and given their direct impacts, particularly on environmental improvement. From this improving food security and livelihoods. long list, each community (through focus Education was seen as a benefit in itself groups) was asked to identify jointly the and one that would improve life chances, three current or previous interventions such as future employment. Scholarships that had been most beneficial in building and bursaries were seen as essential to help their resilience, and to explain why. Table 9 poorer children access education facilities. shows that, among existing interventions, those relating to water, education, health, Access to credit, specifically village productive farms and access to credit were savings and loans schemes, was also prioritized most regularly. frequently cited. It enables households
106 Understanding Community Resilience Table 9. Ranking of resilience-building interventions
Currently Further Type of intervention or recently or future Total score provided provision Water 22 11 33 Water source improvement or improved storage capacity Education Bursaries, scholarships or construction or refurbishment 19 14 33 of school facilities including boarding facilities Health 19 10 29 Improvements to health services, staffing or facilities Inputs to productive farms Irrigation, greenhouses, oxen, agricultural extension 15 10 25 services, etc. Access to credit or other forms of business support 4 14 18 Particularly village saving and loan schemes to access money that can be used to start interviewed. The households varied in size businesses, and also to buy necessities in from 1 to 13 members with an average times of hardship. of 6.2 members, which is typical for the area.9 The number of children and adults There were no distinct differences in was virtually equal, with an average of 3.2 ranking when disaggregated by age/ adults and 3.0 children. Households with gender, livelihood group or intervention a higher proportion of adults have more level. productive labour available, which may be a significant factor in their increased 3.6 How key informants achieved prosperity. resilience Two thirds of households interviewed (65.8 A total of 40 KIIs were undertaken with percent) had members who had completed members of households identified as primary education, and one third (34.3 resilient. One to three interviews were percent) had members who had completed conducted in each sampled parish, secondary or tertiary education. According examining these topics: to government statistics, the literacy rate in Karamoja/Kotido is only 21 percent,10 • Composition and characteristics of the so it seems that resilient households are household; significantly more educated than average. • Pathways to resilience; • Ability to cope with recent shocks and All 40 KII respondents had household hazards; and members engaged in one or more of the • Priority interventions recommended. following activities:
Composition and characteristics of • Business or petty trade (32); resilient households • Agriculture (31); The KII record sheet listed the age, gender, • Wage employment or casual labour education level and economic activity of (21); and/or all members of the resilient household • Livestock raising (16).
9 Uganda OPM, Karamoja Integrated Disarmament and Development Programme Report, 2011. 10 Uganda Bureau of Statistics, 2012.
Kotido and Kaabong districts 107 Most households interviewed (38 out of incomes and agricultural or pastoral 40, or 95 percent) mentioned income from production is the most common model. livestock11 and/or agriculture, but only one resilient household relied exclusively Pathways to resilience on agricultural income. It had a large herd Most respondents stated that they had and a large land holding. No resilient always been resilient, and they cited the household interviewed survived solely on income generated from the multiple agricultural or pastoral production. sources as the key factor in their resilience. Trading in cereals and livestock, by buying All other resilient households (39/40, or when prices are low and selling when they 98 percent) reported multiple income rise, seems a very common route to raising sources. Business activities such as grain the capital required for another business and livestock trading were mentioned by activity, such as brewing or petty trading. 32 households (80 percent). The majority of Four households mentioned using their these households (26/32) mentioned more own savings (often as part of self-help/ than one business activity. A wide range of credit groups) and two cited inheritance business activities were mentioned: or dowry payments as enabling them to expand herds or invest in business activities. • Brewing (20); • Grain trading (8); Ability to cope with recent shocks and • Shop or retail business (8); hazards • Livestock trading (7); The types and nature of hazards affecting • Charcoal or firewood production and the key informant households varied and sale (6); included drought and floods. The most • Rental of property or land (5); commonly stated coping strategy was • Rental of ox or other farm equipment reliance on business or wage income (5); not affected by natural hazards. Some • Brick production (4); and indicated temporary measures such as • Other (9). charcoal production and employment in gold mines (out of the area) as common None of the 21 households reporting coping strategies used only at certain wage income relied solely on this income. times of the year or when agricultural Two households also cited remittances production was poor. from employed family members living elsewhere. Most wage earners were Three households reported stockpiling either government employees (e.g., local food and/or reducing consumption or councilors, teachers or community workers) expenditures. One household mentioned or involved in casual labour. Gold mining a cash transfer from an NGO. No other was the most frequently mentioned casual external interventions were mentioned. employment. Most of these wages would be low and insufficient to support families Priority interventions alone, hence the continued reliance on Key informants were asked for the three business and agricultural income streams. most important interventions to improve their communities’ resilience. Interestingly, It is clear that diversification of income the same five intervention areas emerged sources is a key strategy for resilient for the key informants as for the focus households. Multiple small business groups.
11 Some households mentioned trading in animals but it was not totally clear from the data if that meant they also produced or held livestock.
108 Understanding Community Resilience Interventions relating to agricultural inputs 29 October) and with local government (24 responses) rated most highly. These and non-governmental technical included a range of specific actions such as stakeholders (in Moroto on 31 October) to distribution of oxen and drought-resistant accommodate their differing requirements seeds, grain stores and agricultural tools, due to varying levels of literacy, language and capacity building. The importance skills and technical sophistication (see of expanding agricultural production and Appendix 3 for the list of the technical hence improving both food security and stakeholders participants). Participants in income was clearly a high priority for this the meeting with technical stakeholders group. were briefed on the discussions with community representatives. No major Education interventions (23 responses) gaps or disparities were found between were equally highly ranked. Focus groups the views of the two groups. The also ranked education second. Bursaries findings were also discussed at a national and scholarships for secondary or tertiary workshop, ‘Enhancing Community education accounted for nearly half of the Resilience: Learning from the CoBRA’, responses. Two responses specified nursery convened jointly by OPM and UNDP on 28 education and one mentioned adult November in Kampala (see Appendix 4 for education. Education was always justified the list of participants). as an investment in the future and a way to increase employment opportunities. • Participants felt that the ranked resilience characteristics resonated Expansion of savings and credit groups with the reality in the communities and business training opportunities (11 where the data were collected. respondents) was also widely cited. This Generally speaking, there were was seen as essential not only to raise no surprises. Both community capital for business activities that would representatives and technical increase incomes but also to pay for stakeholders confirmed that the household expenditures such as school statements prioritized by the fees. The higher priority given to accessing communities were as they expected. credit among this group is unsurprising The increasing number of government given that so many are involved in business and development partners in the and trading activities. Water interventions districts had contributed to improving (11 respondents) were also frequently local conditions, as measured by cited for a range of health, food security reduced human and animal mortality and irrigation reasons. rates and improved health/nutrition status. Participants also noted that Other interventions mentioned by at least service delivery had improved in the five respondents included health (8) and Karamoja sub-region generally: the security (5). police had become a permanent presence in almost every subcounty, 4. Summary of feedback and health and education facilities from local and national and veterinary services had become consistently available. Security officers consultations even participated in the validation workshops in Kotido and Moroto. The findings were presented as part of feedback sessions with local stakeholders In particular, the assessment elaborated some months after the assessment. the local contexts very clearly, including Separate sessions were held with the ongoing livelihood transition from community representatives (in Kotido on livestock raising to crop farming. The
Kotido and Kaabong districts 109 participants stressed the importance - Road access, such as in Kacheri of maintaining the delivery of basic subcounty of Kotido District. Roads services in both normal and crisis are needed to facilitate access periods and establishing an enabling to farmland and markets and to environment for resilience building transport goods. through provision of technical, - Land-related issues, such as land technological and mechanical/ ownership, including both formal infrastructure support. tenure and customary systems, • Participants felt that several particularly in light of the trend characteristics were missing or given towards opening up land for insufficient priority. These included: agriculture. More attention may also - Natural resource management, be needed on other infrastructure which is critical to protecting the such as electricity and dams. assets that are the foundation - Employment, particularly for youth. of local livelihoods. Natural After acquiring an education, young resources are often misused for people need gainful employment short-term economic gain, such to meet their livelihood needs and as through charcoal production, be resilient. particularly during crisis periods, • Participants agreed with the without due consideration of the resilience-building measures cited. long-term consequences. Natural They highlighted the importance resource management was not of incorporating the results into captured and prioritized clearly the district development plan for by the communities during the Karamoja, as the results represented assessment because of the difficulty the views of the communities. in tracing the link to resilience. • Participants largely agreed that the Participants in the workshops also rate of attainment of community commented that the community resilience characteristics (4.8) might have incorporated the issue reflects the improved local situation, of natural resources (land, water, understanding that the scores are soil, forest/trees pasture, etc.) in based on perceptions that have the economic, human, physical and no statistical significance. Crop social categories rather than the production has increased in recent environmental category, making years because of improved peace the community’s consideration of it and security in Kaabong and Kotido less apparent. and continuous support from the - Productive pastoralism, with Government and development an emphasis on the need for partners, coupled with adequate and support for fodder production regular rainfall in the region. and livestock health. Kotido and • Participants felt that the characteristics Kaabong districts still have many of a resilient household identified pastoralists, and a nomadic lifestyle by the focus group participants – prevails in the Karamoja sub- employment, diversified IGAs and region. Support must therefore be large herd size – were accurate. They balanced between the livestock made no particular comments on this and agricultural sectors to improve topic. land use and food production. For • Participants highlighted the following example, vaccination services in the specific recommendations related to villages have been inadequate, so building resilience: more attention is needed on animal - Coordination: Many agencies are treatment. supporting disaster risk reduction
110 Understanding Community Resilience (DRR) and resilience-building in humanitarian agencies. Careful Karamoja, but their interventions analysis of the assessment could are not being coordinated. This also help address some of the has caused duplication of services remaining research questions on and inefficient use of funds. The the CoBRA methodology and Government needs to revamp provide inputs on how the tool can coordination from top to bottom be refined further. as part of the planning and - Make use of CoBRA programming process. recommendations: The Government - Natural resource management: It and development/humanitarian was felt that management of natural partners should consider resources should be included in incorporating the CoBRA findings the ranking of resilience-building and recommendations into planning interventions. Communities’ and programming processes at both limited focus on this issue may be local and national levels, including because they feel it is represented Vision 2040, Uganda’s long-term indirectly through other resilience development framework, which will characteristics or because they inform other short- and medium- have limited awareness of its term planning frameworks. The role in attaining other resilience process of mainstreaming the characteristics. Daily life in CoBRA findings should be guided Karamoja involves activities linked by the National Development to natural resource management, Authority, the main government including gold mining and use planning unit. The National Platform of pasture and trees for charcoal for Disaster Risk Reduction could production, so appropriate natural support the process through resource management is crucial to technical backstopping. community resilience. - Complement other DRR works - Continuous assessment: The need and resilience measurement tools: for continuous assessment and CoBRA findings are expected monitoring of priority resilience to inform various actions related characteristics/indicators was to DRR and climate change recommended in the validation adaptation in the country. Today workshops in Kotido and many agencies are undertaking Kaabong districts. Participants important related interventions, yet also recommended verifying the these are often not coordinated, resilience characteristics and their resulting in duplication of services. attainment rates, for example, vis- The multi-agency and multi- à-vis other quantitative data sets. sector nature of the CoBRA helps • Participants recommended a number bridge this gap, enhancing local of next steps for CoBRA: coordination among DRR/climate - Conduct further analysis on CoBRA change adaptation interventions methodology and findings: A under a common vision of detailed analysis of the CoBRA resilience. Further efforts should findings should be undertaken for be made to link the CoBRA tool both of the surveyed districts to with other initiatives related to DRR produce more contextualized policy and climate change adaptation. An and practice recommendations. example is the Regional Analysis This exercise could be led by the Unit, jointly implemented by the Government with partners from Food and Agriculture Organization academia and development/ of the United Nations (FAO), World
Kotido and Kaabong districts 111 Food Programme and the United to credit (341) and livestock herds Nations Children’s Fund in support (320). The extremely high ranking of the Drought Disaster Resilience of productive farms may reflect the and Sustainability Initiative of the recent shift away from pastoralism Intergovernmental Authority on to more settled agricultural and Development. agropastoral lifestyles. The community - Develop a roll-out strategy: clearly senses a lack of knowledge, Participants expressed concern skills, tools and equipment. about whether COBRA would • Productive farms are highly ranked be a sustainable tool. UNDP, by women, possibly reflecting their the Government and other UN/ primary role as farmers. Men, on the NGO partners should explore other hand, rank peace and security opportunities to roll out the CoBRA higher, not surprising given their methodology in other drought- primary activity of herding, which affected districts in Karamoja and makes them more vulnerable to cattle beyond. The assessments should rustling and tribal clashes. Youth rank be implemented together with the education and financial factors highly, ongoing livelihood zone updating possibly reflecting aspirations to move exercise in the sub-region, being beyond pastoral and agricultural led by OPM and FAO. livelihoods. - Incorporate missing resilience • Focus group participants consistently characteristics: Natural resource described income and assets as the management was repeatedly most specific characteristics of resilient mentioned as a missing resilience households. They were described characteristic that needs to be as having one or multiple sources of addressed. Deforestation, mineral income through business activities, mining and bush burning are access to employment or a large herd destructive and unsustainable and/or farm. The critical factor was practices commonly observed in the diversity of these assets, allowing Karamoja, undermining the health households to spread risk against a of ecosystems. Participants also range of crises and shocks. This was recommended assessing various consistent with comments made by key community assets derived from informants from resilient households, pastoralism and crop farming to who cited access to multiple incomes determine how they contribute to as the primary reason for their resilience building in Kotido and resilience. Access to education and Kaabong districts. credit were consistently described during both FGDs and KIIs as key to 5. Conclusions and achieving higher incomes and larger recommendations assets, and hence was a critical factor in reaching a resilient status. • Overall, the focus groups felt they 5.1 Conclusions were halfway to attaining most of the Some of the key findings from the focus highly rated resilience characteristics, groups: with an overall resilience score of 4.8 out 10. However, resilience scores • Participants in Kotido and Kaabong were perceived to drop significantly repeatedly described the top five during crisis periods, to an average resilience characteristics as productive of 1.9 out of 10, showing these farms (bean score of 762), education communities’ extreme vulnerability to (586), peace and security (411), access drought shocks. Access to credit and
112 Understanding Community Resilience sustainable livestock herds were the by the CoBRA assessment, and characteristics reaching the lowest households are adapting to this levels of attainment in normal periods, changing environment by diversifying while water for both humans and their livelihood sources. In planning livestock was scored lowest during and prioritizing investments in these crisis periods. districts, therefore, priority attention • Urban and agropastoral groups scored should be given to the following: themselves as the most resilient - Support the sustainable expansion relative to other groups (5.7 and 5.3 of agricultural production: respectively in the current period). The Comprehensive agricultural majority of these groups (75 percent extension services are required to and 67 percent respectively) also enhance the capacity of farming believed that the number of resilient households to sustainably expand households was increasing. Focus agricultural production. Rainfed groups were generally optimistic; 52 agriculture in semi-arid areas may percent agreed that the number of generate only marginal incomes. resilient households was growing. Expansion of agricultural activities in Women were more optimistic than these areas should be strategically men or youth. planned with due consideration • There is significant overlap between to the type and strains of crops to the interventions most highly rated be grown and the kinds of farming in helping to build resilience and technology (such as irrigation) the most highly rated characteristics and water and soil conservation of resilience. Education, productive techniques to be utilized. farms and access to credit were highly - Expand accessibility and quality of rated interventions. Water and health education: Communities recognize were also highly rated as interventions that higher education levels are but not so highly rated as resilience strongly linked with employment characteristics. Interestingly, peace and other opportunities to expand and security interventions were not income. Access to secondary and rated as a tool for building resilience, tertiary level education in Kotido despite the fact that peace emerged as and Kaamong districts is still very a key resilience characteristic. Resilient low. Authorities need to consider households gave similar ratings to how access can be expanded interventions to build resilience, with so enrolment and completion higher priority to productive farms and rates at least meet the Ugandan savings and credit. average. Investment in education is expensive but essential to the 5.2 Recommendations districts’ long-term resilience and development prospects. • The high priority given to a relatively - Improve access to credit: Limited small set of resilience characteristics access to credit has emerged as a should be used to inform programming key barrier preventing households in this area. from investing in their agricultural (and pastoral) production and other Households are adapting to a more IGAs. Improving access to credit sedentary lifestyle by expanding may be a relatively inexpensive their participation in agriculture and intervention that can be undertaken agropastoral livelihoods. The ability to by multiple actors. It should be survive purely on nomadic pastoralism coupled with business support has declined in the districts covered and advice to enhance the success
Kotido and Kaabong districts 113 rate of micro and small business often, consultations with communities ventures. Implementers must bear focus on interventions that are already in mind that widespread success designed by ‘technical experts’ and of such interventions in building approved by ‘decision-makers’ in a resilience is dependent on carrying top-down manner or for which there out other longer term interventions is budget available. The critical factors in tandem. identified by communities in this study - Place a strong focus on maintaining must be incorporated into future peace and security: Insecurity and planning and decision-making on social instability have long hampered programmes and projects to ensure development in Karamoja, that these plans, programmes and exacerbating the impacts of climatic projects have a real impact on DRR and shocks such as drought and flood. other sectors. Stakeholders involved in It is clear to the communities building resilience need to be aware that peace and security needs to of the wider spectrum of factors that remain a high priority. The long- affect resilience among communities term success of all interventions and households and consider how will depend on establishing and they can re-focus efforts and budgets maintaining a peaceful, stable in these areas. and secure environment in which • Ongoing monitoring and evaluation livelihoods can flourish. will be critical to measure changes • A key criterion in prioritizing future in resilience. The priority issues interventions should be the extent to highlighted by communities in the which they build and diversify incomes CoBRA assessment can be used by and assets either directly or indirectly. governments and other stakeholders Financial security, established through as part of the indicators to track trends having business income, large in local resilience. Undoubtedly, some farms and/or large herds and wage of these indicators are already being employment, was uniformly described monitored as part of data collection, as the reason for household resilience. such as on household income and/ A resilient community will ultimately or households with members who be one in which the vast majority of complete secondary school. Other households have achieved financial important indicators may not figure security. Eventually, this should be the in monitoring, such as the number benchmark of a successful resilience of household income sources or the strategy. number of households accessing • The community perspective must be a savings and credit mechanism. incorporated into efforts to plan and Discussions on an appropriate set of implement resilience. The assessment indicators and monitoring methods findings show that communities are will be important to reach consensus highly aware of the long-term and among stakeholders working to build short-term factors contributing to resilience in the districts. or undermining their resilience. Too
114 Understanding Community Resilience Appendix 1. Participants in CoBRA Kotido/Kaabong assessment
No. Name Organization Phone number Email address 1 Francis Opio UNDP +254-733-770-496 [email protected] 2 Laban MacOpiyo UNDP +254-728-466-622 [email protected] 3 Courtenay Cabot UNDP [email protected] Venton 4 Moses Ojota UNDP +256-772-953-186 [email protected] 5 Martin Odong Office of the Prime +256-783-458-054 [email protected] Minister 6 Moses Aisu Okurut Arbeiter-Samariter-Bund +256-772-837-861 [email protected] 7 Tom Olinga Otukol Adventist Development +256-772-595-896 [email protected] and Relief Agency 8 Agnes Atyang Individual participant +256-772-608-696 [email protected] 9 Walter Kilama Okot Uganda Red Cross +256-775-986-196 [email protected] 10 Moses Einau Samaritan’s Purse +256-774-142-202 [email protected] 11 Godfrey Lotukei Samaritan’s Purse +256-779-768-109 [email protected] 12 Louis Akena MAP International +256-773-144-759 [email protected] 13 Jonathan Logwero Kotido Peace Initiative +256-777-193-716 [email protected] 14 Joseph Emong Warrior Squad +256-771-811-966 [email protected] 15 Francis Lele Dodoth Agro Pastoral +256-755-029-941 [email protected] Development Organization 16 Godfrey Okello Multi Community-Based +256-782-186-813 [email protected] Development Initiative 17 James Ramachiotti CARITAS +256-789-421-943 [email protected] Anyang 18 Silvester Okot ACTED +256-782-255-856 [email protected] 19 Mark Longole Guti Restless Development +256-772-822-394 [email protected] 20 Simon Peter Ojoto CARITAS-Moroto +256-782-389-697 [email protected] 21 Joseph Asibo AZOA [email protected] 22 Christine Lokiru Kotido district local government 23 Lochap Paul CARITAS [email protected] 24 Oscar Okech K Pastoralism & Poverty [email protected] Frontiers 25 Gilbert Buzu World Food Programme [email protected] 26 Bernard Obin Rural Livelihood [email protected] Improvement Concern- Uganda 27 Levi Abura Kotido district local [email protected] government 28 Paskali Panvuga Kotido district local [email protected] government 29 Robert Kennedy Kotido district local Okuda government 30 Lokwi Calistas Kotido Local Council V Adome Chair-Person
Kotido and Kaabong districts 115 Appendix 2. Complete resilience statements and scores for CoBRA Kotido/Kaabong
Resilience Bean Total Full resilience statement characteristic score score
Financial
Productive farms Farmers would be more productive and profitable 762 1,868 (i.e., they would have inputs like quality tools, oxen, fertilizers and improved knowledge of good farming practices).
Access to credit People would have good access to affordable 341 credit and would be saving money (through banks, microfinance organizations, community savings and credit).
Livestock herds Pastoralists would have herds large enough to 320 sustainably support their families.
Employment There would be many opportunities for jobs and other 147 forms of/ paid employment through government, factories and other businesses.
Health care for livestock The community would have access to high-quality and 127 affordable animal health services, including veterinary services and vaccinations, whenever they need them.
Diversified IGAs Many households would be involved in other IGAs 120 such as small businesses and trading.
Vehicles It would be common to own a motorbike or other 44 motor vehicle.
Land Everyone would have secure access ownership of land/ 7 property.
Human
Education All children would be able to complete primary/ 586 1,006 secondary/tertiary education.
Health care for humans The community would have access to high-quality and 315 affordable basic health care locally.
Food security All households would be able to feed themselves well 92 every day.
Skills The community would have the skills and structure to 13 plan and implement solutions to their own problems.
Natural
Natural resource Local rangelands and other natural resources would 11 17 management be well managed so that they do not become degraded over time; trees and forest cover would be well and sustainably managed to provide for future generations.12
Alternative fuel Communities would use environmentally friendly / 6 sustainable sources of fuel for cooking.
12 Note that several statements relating to natural resource management were merged for natural capital.
116 Understanding Community Resilience Resilience Bean Total Full resilience statement characteristic score score
Physical
Water for humans The whole community would have access to sufficient, 284 891 good quality water at all times of the year.
Sanitation Everyone would have good sanitation. 127
Access to markets The community would have easy access to markets to 120 buy goods and sell their produce.
Water for livestock Livestock would have access to sufficient water at all 119 times of the year.
Roads There would be quality roads to the community. 77
Shelter Everyone would live in good quality housing. 72
Telecommunications There would be a reliable mobile phone network to all 53 communities all the time.
Irrigation Farmers would be irrigating land to improve the 21 production of crops for consumption and sale.
Electricity The community would have access to affordable 18 electricity supply.
Social
Peace and security The whole community would enjoy continual peace 411 425 and security.
Good Governance Communities would be served by efficient and 14 non-corrupt community leaders and management structures.
Kotido and Kaabong districts 117 Appendix 3. Participants in CoBRA Kotido/Kaabong technical stakeholders’ feedback session
No. Name Organization Email address
1 Solomon Elungat Office of the Prime Minister [email protected]
2 Moses Aisu Okurut Arbeiter-Samariter-Bund [email protected]
3 Tom Olinga Otukol Adventist Development and Relief [email protected] Agency
4 Walter Kilama Okot Uganda Red Cross [email protected]
5 Moses Einau Samaritan’s Purse [email protected]
6 Godfrey Lotukei Samaritan’s Purse [email protected]
7 Godfrey Okello Multi Community-Based Development [email protected] Initiative
8 Silvester Okot Agency for Technical cooperation and [email protected] Development
9 Mark Longole Guti Restless Development [email protected]
10 Simon Peter Ojoto CARITAS [email protected]
11 Sp George Obia District Police Commissioner [email protected]
12 Loput Simon Peter Deutsche Gesellschaft für Internationale [email protected], Zusammenarbeit [email protected]
13 Okecho Yoronimo Action Against Hunger [email protected]
14 Omara Emmy Adventist Development and Relief Agency
15 Emmanuel Isingoma Cooperation and Development Coopdeve.isingoma@gmail
16 Martha Iriama Dan Church Aid Mir.uganda@dca-dk
17 Nangole Joseph Lobot Zionist Organization of America [email protected]
18 Angello Zachary [email protected]
19 Mike Kidon O Moroto
20 Panvuga Kotido [email protected]
21 Francis Opio UNDP [email protected]
22 Laban MacOpiyo UNDP [email protected]
23 Laz Ocira UNDP [email protected]
118 Understanding Community Resilience Appendix 4. Participants in the ‘National Workshop on Enhancing Community Resilience: Learning from the CoBRA’
No. Name Title Organization
1 Hon. Eng. Hillary Onek Minister of Disaster Preparedness, Ministry of Disaster Preparedness, Relief and Refugees Relief and Refugees
2 Hon. Akello Rose Lilly Member of Parliament Parliament
3 Hon. Aleper Margaret Member of Parliament Parliament Achilla
4 Richard Kakaire P/A Hon. Alaso Alice Asianut Parliament
5 Menhya Gerald Assistant Commissioner for OPM Disaster Preparedness
6 Kibungo Jonas Moses SAS for Under Secretary P&D OPM
7 Owaro Johnson Agriculture Officer OPM
8 Vincent Woboya Principal Disaster Management OPM Officer
9 Lugaizi Isa Geologist Ministry of Energy and Mineral Development
10 Olive Nalugo Civil/ Structural Engineer Ministry of Lands, Housing and Urban Development
11 SP Kihanda Hassan Ag. Deputy Director, Fire and Uganda Police Rescue
12 Bogdan Stefanescu Head of Section, Rural European Union Delegation to Development Uganda
13 Nathalie Ann Denjon Reporting Officer Acted
14 Thore Karlsson Country Director Adventist Development and Relief Agency (ADRA)
15 Booker Ajuoga Public Relations ADRA
16 Jasper Okello Regional DRR Program Assistant DanChurchAid (DCA)
17 John Musinguzi Programme Officer DCA
18 Isaac Bwire Programme Manager IIRR
19 Edyegu Stephen WASH Project Manager International Organization for Migration (IOM)
20 Alexander Billings Project Officer IOM
21 Christopher i Orach Deputy Dean, School of Public Makerere University Health
22 Jeff Mungu Programme Officer World Food Programme
23 Gerard Omodoi Chief of Party ZOA
24 Adam Vinaman Yao Deputy Representative FAO
25 James Okoth National Programme Manager FAO
Kotido and Kaabong districts 119 No. Name Title Organization
26 Geoffrey Muhindo DRR CCAO Oxfam
27 David Othieno DRR Coordinator Plan Uganda
28 Pascal Onegiu Okello DRR Advisor UN International Strategy for Disaster Reduction
29 Mawanda Shaban Senior Programme Manager, DRR Uganda Red Cross Society
30 Safiou Esso Ouro-Doni Deputy Country Director UNDP
31 Jose Manzano DRR Advisor UNDP/OPM
32 Gilbert Arguyo DRR Analyst UNDP
33 Yuko Kurauchi Program Specialist UNDP
34 Francis Opiyo Project Coordinator UNDP
35 Catherine Fitzgibbon Consultant UNDP
36 Lazarus Ocira Consultant UNDP
120 Understanding Community Resilience Annex 4 Community-based Resilience Analysis Assessment Report Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts Kajiado county, Kenya
Commissioned by UNDP Drylands Development Centre
Under the framework of the European Commission Directorate General for Humanitarian Aid and Civil Protection’s Drought Risk Reduction Action Plan
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 121 Contents
1. Introduction...... 123 2. Approach...... 123 2.1 Characteristics of field site...... 123 2.2. Data collection ...... 124 2.3 Constraints and limitations of data collection process ...... 125 3. Findings...... 125 3.1 Main hazards or shocks ...... 127 3.2 Characteristics of a resilient community...... 127 3.3 Extent to which the community has achieved resilience...... 131 3.4 What a resilient household looks like...... 134 3.5 Interventions that contributed to household resilience...... 135 3.6 How key informants achieved resilience ...... 136 4. Summary of feedback from local consultations ...... 138 5. Conclusions and recommendations...... 140 5.1 Conclusions...... 140 5.2. Recommendations...... 141
122 Understanding Community Resilience 1. Introduction workshops to review and validate the draft assessment report, one held for community The comprehensive Community-Based representatives on 6 November and the Resilience Analysis (CoBRA) assessment other for local government and non- was undertaken in three districts of Kajiado governmental technical stakeholders county, Kenya, between 19 and 30 August working in the area on 7 November. 2013. The exercise was carried out with financial support from the European A detailed explanation of the conceptual Commission Directorate General for framework that underpins the methodology Humanitarian Aid and Civil Protection is contained in the CoBRA Conceptual under the framework of its Drought Risk Framework and Methodology document.1 Reduction Action Plan (DRRAP). This assessment builds on the initial field trial 2. Approach of the draft CoBRA tool undertaken in late 2012. 2.1 Characteristics of field site The multi-agency assessment was jointly The field assessment was conducted in led by the United Nations Development five districts (Kajiado Central, Mashuru, Programme (UNDP) Drylands Development Loitoktok, Kajiado North and Isinya) of Centre (DDC) and the Government of Kajiado county, Kenya, which is in the Kenya’s National Drought Management country’s southeast, bordering United Authority (NDMA). Logistical and practical Republic of Tanzania. A total of 18 support was provided by UNDP-DDC. A administrative locations were selected for wide range of international and local NGOs the assessment (Figure 1), representing operating in the area also participated pastoral, agricultural, agropastoral and in the assessment by providing local urban/peri-urban livelihood zones. The staff as facilitators to undertake CoBRA county covers 21,902 sq km and is part training and fieldwork. A list of agencies of the arid and semi-arid lands (ASALS) participating in the CoBRA training and in Kenya. Livestock rearing is the main field data collection can be found in livelihood activity although rainfed and Appendix 1. irrigated agriculture is practised in some parts of the county. • The CoBRA assessment has four broad objectives: Kajiado was chosen because it is one • Identify the priority characteristics of of the semi-arid counties prone to resilience for a target community; drought in Kenya. In the last decade • Assess the communities’ achievement the county has experienced periods of of these characteristics at the time of prolonged drought that have hampered the assessment and during the last the livelihoods of both pastoralists and crisis or disaster; agropastoralists. According to the district • Identify the characteristics and livestock production officer, drought had strategies of resilient households; and severely affected pastoralists, reducing the • Identify the most highly rated county’s livestock population by almost interventions or services in building three quarters, from 350,000 to 95,000 just local resilience. in 2009. Around 89 percent of households engage in livestock farming, which is the This report outlines findings of the Kajiado main source of income. Crop farming is CoBRA assessment. It also incorporates practised by 46.4 percent of households feedback and inputs generated at two during the rainy season, but in the dry
1 www.youblisher.com/p/784724-Community-Based-Resilience-Analysis-CoBRA-Conceptual-Framework-and-Methodology/
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 123 Figure 1. Surveyed locations in Kajiado county
Surveyed locations
season this figure falls to only 4.2 percent.2 relief food for survival; currently, 13.2 percent of the population is receiving The increasing frequency and intensity of relief food through the Protracted Relief drought cycles has meant that communities and Recovery Operation by the World are unable to completely recover from the Food Programme, being implemented previous cycle before the next one begins. by the Neighbours Initiative Alliance. To In addition, the amount of land suitable for tackle these challenges, communities in grazing is diminishing due to subdivision the county have adopted harmful coping of communally owned land and increasing mechanisms, such as cutting trees for urbanization, which also impede the charcoal production and sand harvesting. traditional drought coping mechanism of migration for water and pasture. As a 2.2. Data collection result, growing numbers of pastoralists are falling out of pastoralism every year, Complete details of the methodology used becoming more dependent on emergency to undertake a CoBRA assessment are
2 Welthungerhilfe, Household Baseline Survey for Improved Drought Cycle Management in Vulnerable Poor Pastoral and Agro- Pastoral Communities in Kajiado District, 2010.
124 Understanding Community Resilience included in the main CoBRA assessment was for very short periods, or they report3 and implementation guidelines. asked for compensation, which was Data were collected using qualitative not provided, consistent with the methods, in the form of focus group project guidelines. In cases where it discussions (FGDs) and key informant was difficult to get communities to interviews (KIIs). participate in the assessment, the teams found alternative groups to Field work was undertaken by a team of 18 participate using their local contacts. members. The team was divided into three • In areas such as Loitoktok, where the groups, each with five facilitators and a key livelihood activities are irrigated team supervisor. Each group covered two crop agriculture and livestock districts, except that one group covered production, communities also had only the expansive Loitoktok district. The little time available to participate. facilitators and supervisors, who were • For the team, traversing this very familiar with the local context, jointly large county sometimes proved time selected the locations and compositions consuming, leaving limited time for of focus groups, using statistical data interviews. and criteria such as livelihood zones and • In most cases, the scoring of level of interventions. Whenever possible, attainment of the identified resilience separate focus groups were organized for characteristics was skewed towards men, women and youth (aged 15–30, both lower values. The facilitators had to sexes) to ensure that views on resilience probe to bring the values to seemingly from different gender and age groups realistic levels. would be adequately captured in the discussions. FGDs were facilitated by two 3. Findings staff from the participating government departments and NGOs, and each group This section reports on the summarized had around 15 participants. findings from the CoBRA field work conducted during the month of June In each focus group location, one to four 2013 in Kajiado. Specifically, the findings KIIs were undertaken with households are presented according to the following perceived to be resilient, as identified categories: by focus group participants during their discussions. In some cases, the households • What the main hazards or shocks were also identified as resilient through facing the communities? conversations with local chiefs or other • What are the characteristics of a senior residents. A total of 36 FGDs and resilient community? KIIs took place, in addition to those at the • To what extent has the community 8 trial sites. A total of 36 FGDs and 36 KIIs achieved resilience characteristics? took place throughout the county (Table 1). • What does a resilient household look like? 2.3 Constraints and limitations of • What interventions contributed to household resilience? data collection process • What additional intervention would • It proved difficult to work in the best build resilience? major urban areas of Kajiado • How did key informants achieve and because community members were maintain resilience? busy. When they gave their time, it
3 UNDP-DDC, Understanding Community Resilience: Findings from Community-Based Resilience Analysis Assessments – Marsabit, Turkana and Kajiado counties, Kenya, and Karamoja sub-region, Uganda (Nairobi, UNDP-DDC, 2014 [in press]).
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 125 Table 1. FGDs and KIIs undertaken for Kajiado CoBRA assessment
Population Livelihood District Location FGDs KIIs FGD composition (2009) zone
Male elders; Mailwa 4,364 2 2 Pastoral females (all ages)
Kajiado Male elders; Elangata Wuas 4,648 2 2 Pastoral Central female elders
Male elders; Namanga N.A. 2 2 Peri-urban females (all ages)
Male elders; Imaroro 5,815 2 2 Agropastoral female elders
Mixed elders; Mashuru Merueshi 3,335 2 2 Pastoral Mixed youth
Male elders; Poka 5,567 2 2 Pastoral female elders
Males (all ages); Kuku 8,144 2 2 Pastoral females (all ages)
Male elders; Olchoro 7,584 2 2 Agricultural female elders
Mixed elders; Inkariak-Ronkena 4,649 2 2 Agricultural Mixed youth Loitoktok Males (all ages); Entarara 17,685 2 2 Agricultural females (all ages)
Males (all ages); Njukini 9,022 2 2 Pastoral females (all ages)
Mixed elders; Rombo 10,897 2 2 Agricultural Mixed youth
Males (all ages); Magadi 5,525 2 2 Pastoral females (all ages)
Mixed elders; Olkiramatian 7,947 2 2 Pastoral Kajiado Mixed youth North Mixed elders; Kiserian 26,761 2 2 Urban Mixed youth
Males (all ages); Oloolua 33,754 2 2 Peri-urban females (all ages)
Males (all ages); Kitengela 58,208 2 2 Urban female elders Isinya Male elders; Ilpolosat N.A. 2 2 Agropastoral female elders
Total 36 36
N.A. means no data available.
126 Understanding Community Resilience 3.1 Main hazards or shocks each statement. For ease of comparison, The main hazard reported in all the the statements were grouped into the five FGDs was drought. Communities viewed sustainable livelihood framework (SLF) it as the most significant contributor categories.4 to livestock losses and the single most important factor limiting their resilience. Table 2 lists the 12 highest ranked Communities reported the most recent characteristics/statements used to drought, of 2009, as the main “crisis” describe a resilient community, within the period. Communities further indicated that five SLF categories. (Note that many more the current period when the assessment statements/characteristics were included was undertaken was principally a “normal” in the ranking, but because they received period, though a few communities stated low scores they are not reported here. that it was a “good” period (Determining The full list of statements and scores is this involved comparing a current period, provided in Appendix 2). here represented by the preceding 12 months, to the long-term climatic norm for Table 2. Community ranking of resilience the specific location under assessment). characteristics by SLF category To a limited extent, the communities also Top 3 resilience Total reported livestock diseases and human SLF characteristics bean category diseases as other hazards in Kajiado. (‘bean scores’) score*
Financial Access to credit (139) 649 3.2 Characteristics of a resilient Productive farms (126) community Employment (106)
Focus group participants were asked Human Education (608) 1,013 to describe what they viewed as the Health care for humans (329) characteristics of a resilient community. Natural N.A. 94 The data are first presented for all the respondents, to give an overall picture Physical Roads (434) 1,812 of the most highly rated statements. Water for humans (403) This is followed by an analysis of each Water for livestock (208) Access to markets (185) category of respondent by gender/age, Sanitation (139) livelihood group and level of intervention Irrigation (139) in the community. This analysis is used to disaggregate findings and identify Social Peace and security (119) 139 differences across groups. *Total bean score does not match the sum of the bean scores in the middle column because the middle column reflects only the top-ranking characteristics while the right column reflects all the Analysis – all respondents characteristics discussed in each category. Focus group participants were asked Note: N.A. means this category ranked too low for inclusion in to identify and rank statements used the table. to describe a resilient community. Each participant was given six beans to rank the Figure 2 shows the ranking of characteristics three most significant statements in terms of a resilient community among all focus of priority for building resilience (three group participants in the three districts. beans for the most significant statement, Figure 3 presents the total score under two for the second and one for the third). each of the five SLF categories. The bean scores were then totalled for
4 The sustainable livelihood framework presents the main factors that affect people’s livelihoods and typical relationships between them. It identifies five core asset categories or types of capital upon which livelihoods are built: financial, human, natural, physical and social. For further details, please refer to: UK Department for International Development (DFID), Sustainable Livelihoods Guidance Sheets (London, DFID, 1999).
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 127 Figure 2. Top-ranking resilience characteristics – all respondents 700 600
500
400 300
200
100 0
Roads
Irrigation Education Sanitation Electricity
Employment
Access to credit Productive farms Water for humans Access to markets Water for livestock Peace and security
Health care for humans
Figure 3. Top-ranking resilience characteristics by SLF category – all respondents
2000
1800
1600 Others Peace and security 1400 Electricity Sanitation 1200 Irrigation Access to markets 1000 Water for livestock Water for humans 800 Roads Wildlife 600 Health care for humans Education 400 Employment Productive farms 200 Access to credit
0 Financial Human Natural Physical Social
In terms of characteristics supporting Analysis by gender and age resilience, education, roads, water and The priority resilience characteristics were health ranked most highly. In terms of the also analysed by gender and age group, SLF categories, characteristics concerning through groups comprised of women, physical capital ranked most highly, men and youth (aged 15–30, both sexes followed by human and financial scores. together) (Figure 4). The bean score Natural and social capacity received the allocated to each of the five SLF categories lowest overall scores. was taken as a percentage of the total bean
128 Understanding Community Resilience Figure 4. Top-ranking resilience categories by gender/age group 100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Women Men Youth score, and the results demonstrate the Participants in the review and validation differing priorities that men, women and meeting deliberated on these findings youth place on resilience characteristics. and shared the insight that the youth likely ranked roads highly given their Further to this, the most highly ranked aspiration for the ‘modern lifestyle’, with resilience characteristics by gender/age access to alternative modes of transport group are presented in Table 3. such as motorcycles and better access to education, information and off-farm The data suggest the following points: employment/business opportunities in central locations. • Women and men similarly placed high value on education, water and health. Analysis by livelihood group • Youth, by contrast, ranked roads Top-ranking resilience categories were also most highly, well above any other analysed for a range of livelihood groups characteristic. Education, water and – agropastoral, pastoral, agricultural, employment all came in second place, urban and peri-urban. Figure 5 illustrates yet employment was not ranked in the the differences in bean score allocation top three by men or women. according to the five SLF categories among the five livelihood groups. Table 3. Top-ranking resilience characteristics by gender/age group Further to this, the most highly ranked resilience characteristics by livelihood Gender/ Top 3 resilience Total group are presented in Table 4. age group characteristics score* Women Education (270) 1,305 The data suggest the following points: Water for humans(137) Health care for humans (128) • Education is prioritized by all five Men Education (225) 1,380 groups, though there are distinct Water for humans (180) differences between them in terms of Health care for humans (158) the level of priority. Youth Roads (149) 507 • Agropastoral and pastoral groups are Education (47) similarly focused on education, water Water for humans (47) and health. Employment (47) • Agricultural groups include education *Total score does not match the sum of the scores in the middle in their top three statements, but column because the middle column reflects only the top-ranking characteristics while the right column reflects all the characteristics irrigation and access to markets also discussed in each category. feature.
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 129 Figure 5. Top-ranking resilience categories by livelihood group
100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% Agropastoral Pastoral Agricultural Urban Peri-urban
Table 4. Top-ranking resilience characteristics deliver products to market or trading of by livelihood group farm products, for example) and their need for better access to goods and services, Livelihood Top 3 resilience Total government offices, etc. group characteristics score*
Agropastoral Water for humans (91) 516 Analysis by intervention level Education (83) The three districts assessed comprise 6 to Health care for humans (68) 11 administrative locations. Consultation Pastoral Education (271) 1,665 with field teams, NDMA and other Health care for humans (218) government line departments was used Water for humans (208) to map accessibility to/presence of the following basic services and interventions: Agricultural Education (104) 718 Irrigation (68) Access to markets (66) • Number and level of health facilities; • Tarmac road; Peri-urban Education (74) 200 • Other main road; Roads (42) Productive farms (20) • Mobile phone coverage; • Well-functioning livestock market; Urban Roads (148) 608 • Savings and credit programmes; Access to markets (82) • Number and level of education Education (75) facilities; *Total score does not match the sum of the scores in the middle • Banks; and column because the middle column reflects only the top-ranking 5 characteristics while the right column reflects all the characteristics • Bank agents/Mpesa. discussed in each category. Interventions that are universally provided • Peri-urban and urban communities in all locations (such as food aid and water place a greater emphasis on roads. interventions) or provided only at very low scale (to fewer than 500 beneficiaries) were The joint review and validation workshop excluded from this mapping exercise. The in November 2013 provided the insight locations were then divided into three that the higher priority given to roads by groups: urban and peri-urban populations reflects their desire for roads related to market • The bottom third represented low needs (for use by motorcycle taxies that intervention areas, which have up
5 M-Pesa is a mobile phone-based money transfer and microfinancing service offered by Safaricom and Vodacom in Kenya and Tanzania. (M stands for mobile and pesa means money in Kiswahili.)
130 Understanding Community Resilience to 8 categories of interventions The data suggest the following points: implemented. • The middle third represented medium • High intervention groups placed intervention areas, which have 9 the greatest emphasis on physical to 12 categories of interventions characteristics of resilience, notably implemented. roads and water. Education was • The top third represented high also listed as one of the top three intervention areas, which have 13 characteristics. Medium intervention to 85 categories of interventions groups followed a similar profile. The implemented. high ranking of roads reflects the generally poor quality of roads in the Figure 6 shows the difference in scores for area compared with other infrastructure the five SLF categories among the three and services, such as mobile phone intervention groups. coverage, schools and health facilities. • Low intervention groups gave more Further to this, the most highly ranked emphasis to human characteristics of resilience characteristics by intervention resilience, notably health interventions level are presented in Table 5. for humans.
Figure 6. Top-ranking resilience categories by level of intervention 100%
80%
60% Social Natural 40% Financial Human 20% Physical
0% High Medium Low
Table 5. Top-ranking resilience characteristics 3.3 Extent to which the by intervention level community has achieved resilience Intervention Top 3 resilience Total level characteristics score* Focus group participants were asked High Roads (244) 1,405 to score the extent to which they had Education (234) achieved their priority characteristics of Water for humans (177) resilience. They were asked to score each Medium Water for humans (129) 1,019 statement twice, first for the current period Education (127) (agreed to be a normal period) and second Roads (122) for the last significant crisis period (almost Low Education (248) 1,283 universally identified as the drought period Health care for humans of 2009). (142) Water for livestock (107) Figure 7 shows a spider diagram with
*Total score does not match the sum of the scores in the middle several rings. The outer ring represents column because the middle column reflects only the top-ranking a perfect or ideal score for all statements characteristics while the total in the right column reflects all the characteristics discussed in each category. in that SLF category. The blue band
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 131 Figure 7. Community attainment of resilience per SLF category – all respondents
Physical 10.0
8.0
6.0
4.0 Social Human 2.0 Current year rank 0.0 Crisis year rank Full score
Natural Financial
Note: Possible score ranges from 0 to 10.
shows the average attainment of those as the area in which they have the greatest characteristics by communities in the degree of attainment of resilience. This current period, and the red line represents category was dominated by responses the last crisis period. The scores are ranked related to peace and security, not on a scale from 0 to 10, with 10 being surprising given that this part of Kenya perfect attainment of that characteristic remains relatively stable. The remainder (that is, the entire community has access to of the categories were ranked similarly in sufficient, good quality water at all times in current and crisis years, suggesting that the current/crisis period), and 0 being no in a normal year community members attainment (no one in the community has rank their attainment of characteristics of access to sufficient, good quality water at resilience at 3.5 out of 10. all times in the current/crisis period). Table 6 presents the scores by SLF category for Other characteristics highly ranked for the top-ranked characteristics of resilience. attainment in normal times (relatively speaking) include access to credit (4.4), As with the previous section, the findings water for livestock (4.4), education (4.3) are presented first for all respondents and access to markets (4.3). Characteristics and are then disaggregated by livelihood that were deemed to be poorly attained groups. This section does not include the include employment (2.1), roads (2.1) and analysis by gender. This is because focus irrigation (2.5). groups were asked to rank the attainment of the resilience characteristic statements The joint review and validation meeting for the entire community, and therefore in November 2013 reviewed the any differences between men and women scores closely and made the following in the same community would be based observations: on perceptions. • Peace and security: High attainment Social characteristics of resilience have for peace and security was attributed the highest score (4.1), meaning that to the strength of traditional cultural communities ranked social characteristics institutions in combination with
132 Understanding Community Resilience Table 6. Community attainment of resilience by SLF category – top-ranked characteristics
SLF category Top-ranked characteristics Current period score Crisis period score
Overall average 3.5 2.3
Financial Access to credit 4.4 3.2
Productive farms 3.1 2.0
Employment 2.1 1.3
Human Education 4.3 2.5
Health care for humans 3.2 1.7
Natural N.A. - -
Physical Roads 2.1 1.4
Water for humans 3.2 1.5
Water for livestock 4.4 1.4
Access to markets 4.3 2.8
Sanitation 3.5 2.0
Irrigation 2.5 1.6
Social Peace and security 4.5 4.2
N.A.: This category ranked too low for inclusion in the table.
traditional leadership such as chiefs undertaken very slowly and in limited and elders. Strong formal governance parts of the county. structures (such as police posts) and informal structures (such as community Table 7 aggregates the scores given to policing) collaborate and complement each of the five SLF categories to give an each other within the county. The overall ‘resilience score’. While this clearly favourable climate conditions in masks differences between SLF categories, recent years also increased on-farm it is useful in providing an overall indicator employment opportunities and of where communities see themselves, for reduced the number of idle youth, comparison across groups. The figures which also accounts for improved should be viewed with some caution, security and stability. as these scores represent community • Access to credit: The high score on perceptions around attainment, and may access to credit was attributed to overstate or understate reality. the increased number of financial institutions such as microcredit In the three districts, the average level schemes and credit facilities, especially of attainment of resilience characteristics within Kajiado Central district. was 3.5 out of 10 in the current ‘normal’ • Water for livestock: Concerted efforts period, dropping to 2.3 out of 10 during by the Government, NGOs and the the crisis period. private sector over the past few years have made water more accessible to Results by livelihood group show that most community members. urban groups viewed themselves as the • Roads: Roads are poorly maintained most resilient while peri-urban groups and construction of new roads is perceived themselves as the least.
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 133 Table 7. Aggregate resilience scores rural lands to private owners and thus have less access to resources, while tending to Current year Crisis year be overlooked or neglected by external rank rank development/humanitarian interventions, All groups 3.5 2.3 which are typiclaly concentrated in either very rural areas or urban centres. Livelihood group Agropastoral 2.7 2.0 On the other hand, agricultural groups Pastoral 3.3 2.4 are increasingly optimistic, as they have been experiencing improving access to Agricultural 4.0 2.8 permanent irrigation structures, such Urban 4.5 2.3 as in Loitoktok, which directly leads to higher farm production. They also see Peri-urban 2.5 0.5 more agricultural business opportunities Intervention level (production of tomatoes, onions, horticultural products, etc.) and better High 3.9 2.4 links to markets due to the construction of Medium 2.0 2.3 a major road.
Low 3.2 2.2 3.4 What a resilient household Note: Maximum possible score of 10 looks like When the scores are disaggregated by Focus group participants were asked to the level of services and interventions, describe the characteristics of households high-intervention areas ranked their level that are more resilient compared to others of attainment of resilience characteristics – that is, the households that have already more highly than medium- and low- attained many (or all) of the resilience intervention areas. This could suggest that characteristics prioritized. The top three interventions are contributing to resilience. characteristics of a resilient household, Equally, it could suggest that interventions cited consistently by focus groups, were: are targeted to high-resilience areas. Longitudinal data would have to be studied • Having a business or income- to draw inferences on the causality of high generating activity (32 of the 36 focus interventions on resilience. Proximity to groups); Nairobi could also be an important factor • Having a member with employment or – areas that are closest to Nairobi are more wage labour (27); and likely to have access to markets, business • Having a large herd (19). and job opportunities, and this access might be a more important element of A few other household characteristics were resilience scores. mentioned, but significantly less often:
The joint review and validation meeting • Having educated members (and where attributed the lower resilience scores by this was mentioned it was often in the peri-urban group to the fact that this relation to the ability of the educated group is in a livelihood transition, moving person to get employment). from pastoral/agropastoral activities to • Having land ownership or access other income-generating activities. Some rights. of them are so-called ‘pastoral dropouts’. Newly settled in the vicinity, which is Focus groups were asked whether neither urban nor rural, they are largely the number of resilient households is wage earners with limited assets and low increasing, decreasing or staying the same. purchasing power. They often sell their As Table 8 shows, 63 percent of 36 focus
134 Understanding Community Resilience Table 8. Is resilience increasing, decreasing or staying the same?
Increasing Decreasing Staying the same
All respondents 63% 34% 3%
Gender/age
Women 73% 27% 0%
Men 46% 46% 8%
Youth 67% 33% 0%
Livelihood groups
Agropastoral 60% 40% 0%
Pastoral 73% 20% 7%
Agricultural 86% 14% 0%
Peri-urban 0% 100% 0%
Urban 33% 67% 0%
Intervention level
High 50% 50% 0%
Medium 67% 33% 0%
Low 75% 17% 8% groups said that resilience is increasing. 3.5 Interventions that contributed These findings were similar across gender to household resilience and age. Communities were asked to list all the services and interventions they had When disaggregated, women and youth benefited from in the last two to five tended to be more optimistic than men, years. A reasonably wide range of and agricultural and pastoral households interventions was mentioned, including were more optimistic than peri-urban and water, education, health, credit, diversified urban groups. Areas with low levels of income-generating opportunities, mobile intervention were more likely to state that phone coverage, productive farms, resilience was increasing (75 percent) than roads, women’s empowerment and high intervention areas (50 percent). All livestock support. From this long list, each of these figures are based on a relatively community (through FGDs) was asked small sample size and therefore should be to identify jointly the three current or interpreted with some caution. previous interventions that had been most beneficial in building their resilience and The joint review and validation meeting why. Groups were also asked to list the attributed the optimistic views of women three further or additional interventions to the fact that they have recently had more they felt would best build their resilience. access to credit facilities and increasingly Many communities restated interventions engaged in business activities to diversify similar to those mentioned in the first income options. In the past, women list, with the justification that the current rarely owned assets, but the situation has provision or scale of intervention was too been changing and they now have more limited and should be expanded. The opportunities to own assets. results are summarized in Table 9.
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 135 Table 9. Ranking of resilience-building interventions
Currently/ Further Type of intervention recently or future Total score provided provision Education Bursaries, scholarships or construction/refurbishment of 35 11 46 school facilities including boarding facilities Water 17 16 33 Water source improvement or improved storage capacity Health 12 20 32 Improvements to health services, staffing or facilities Productive farms Irrigation, greenhouses, oxen, agricultural extension 3 12 19 services etc. Roads 5 10 15 Access to quality, tarmac roads Markets Access to markets for buying and selling goods and 2 7 9 livestock
The table shows the repeated and clear • Pathways to resilience; priority given to education, water and health • Ability to cope with recent shocks and interventions, reflecting the high ranking hazards; and given to these factors as characteristics of • Priority interventions recommended resilience by all focus groups. Education by resilient households. was seen as a benefit in itself and one that would lead to improved life chances, Composition and characteristics of such as employment for children. Water resilient households interventions were prioritized by all The KII record sheet records the age, livelihood groups, particularly for improving gender, education level and economic food security and livelihoods. activity of all members of the resilient household interviewed. The resilient Productive farms, roads and markets households varied in size from 3 to 14 were all justified on the basis that they members, with an average of 6.4 members, would contribute to household income which is relatively high. The vast majority by improving yields and facilitating sale at of households interviewed (88.8 percent) market. had members who had completed at least primary education and nearly half of the 3.6 How key informants achieved households (44.4 percent) had members who had completed tertiary education. resilience Government statistics indicate that about A total of 36 KIIs were undertaken (two in 31 percent of the population in Kajiado each location) with members of households county has no education, 42 percent has identified as resilient. KIIs examined the a primary education and 28 percent has following areas: a secondary or higher education.6 From these figures, it would seem that ‘resilient • Composition and characteristics of the households’ are significantly better household; educated than the average.
6 Kenya National Bureau of Statistics and Society for International Development, Exploring Kenya’s Inequality: Pulling Apart of Pooling Together? (Nairobi, Ascent Limited, 2013).
136 Understanding Community Resilience All 36 KII respondents had household earner. Employment (where stated) was members engaged in two or more of the a mix of jobs in the public sector (such following: as nurses and teachers) and the private sector. Many of the private sector jobs were • Livestock raising (33 respondents); professional or managerial; accountant, • Business (31); laboratory technician, IT specialist, • Waged employment or casual labour journalist and sales and marketing were (21); and/or all mentioned. Other jobs cited included • Agricultural production (18). game rangers, NGO staff, drivers and casual labour (not specified). According to the findings, no household stated that it relied on one of these income Pathways to resilience sources exclusively. Three households said All respondents cited one or more of their they were livestock farmers but also cited multiple sources of income (specified the same as a business activity (for example above) as the reason for their resilience. milk trading or steer fattening). This reflects Apart from income sources the most the commercial orientation of the livestock frequently mentioned factors were access farming households interviewed. Even to credit and loans (16 households), if these three households are excluded, education (12) and other training received nearly 92 percent of resilient households from various sources such as NGOs or interviewed had multiple income sources. previous employment. Consequently the diversification of income sources is clearly a key strategy and Good livestock management, including criterion of resilience. timely sale and restocking, was mentioned by at least 13 households as a factor. Many After livestock rearing, business activities noted that livestock had been sold to invest were most frequently cited as a source of in other business ventures or pay school income. Varied businesses were mentioned, fees. Six households mentioned having with at least 17 households reporting their own water supply as a factor leading involvement in several business activities. to resilience, and four of these had used According to government statistics, 33 the water to expand into irrigated crop percent of the county’s population is production, mostly fruits and vegetables. working for pay.7 The most commonly cited business activities are listed below with the Ability to cope with recent shocks and number of citations in brackets: hazards Almost all key informants referred • Livestock trading or fattening, etc. (12 to drought as their primary hazard. households); Interviewees mentioned different droughts • Milk trading (9); depending on which had personally • Rental of land or property (9); affected them most, from periods as • Shop or kiosk (6); long ago as 2003, although the 2009 • Bead work (5); drought was most frequently mentioned. • Running hotel or guest house (5); Virtually all households noted two coping • Taxi or transport business (5); strategies: having sources of income not • Grain and vegetable trading (5); and affected by drought and good livestock • Other (7). management, including timely destocking. Wage earners and most businesses did not Of the 21 households reporting wage seem affected by droughts. income, half had more than one wage
7 Ibid.
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 137 Almost all livestock owners (33 of 36 Education interventions (12 respondents) households) described selling animals were also highly ranked. Most respondents early on in drought periods and using the referred to the importance of constructing income to buy fodder or move remaining tertiary facilities or bursaries for secondary animals elsewhere. This was described or tertiary education. more as a standard good practice than an emergency ‘coping’ strategy. Three Support to expand agricultural production households reported losing significant (9 respondents) and irrigation (5 numbers of animals during droughts, but respondents) were highly rated, particularly noted they had restocked after the drought agricultural extension services and using income from other sources or savings. provision of quality seeds and fertilizers.
Four households cited their own water Other interventions mentioned by more source or use of irrigated production than five respondents included employment as factors in coping with drought. Only creation (8), roads (6) and water (6). three respondents mentioned external support, in the form of government food With the exception of education and distribution. agricultural inputs, the interventions prioritized by key informants were quite Priority interventions different from those mentioned by focus Key informants were asked for the three group participants. Key informants ranked most important interventions to improve credit and business skills as a priority, their communities’ resilience. Interestingly, reflecting their engagement in business many respondents referred to actions activities. The inclusion of good livestock that should be taken by households and agricultural practices in the list of themselves rather than externally delivered critical interventions highlights how these interventions. This reflects a much greater households take a more commercial and level of self-reliance in this area, which has self-reliant approach than others. not had the same history of humanitarian and development programmes by NGOs 4. Summary of feedback from and the Government. Consequently many local consultations idiosyncratic actions or interventions were mentioned. The findings discussed above were presented as part of a review and The expansion of savings and credit feedback session with local stakeholders opportunities and business training in Kajiado. Separate sessions were held were cited most widely (22 times). The with community representatives (on 6 importance of accessing affordable credit November 2013) and with technical was seen as very important, and a few stakeholders (on 7 November) to key informants also mentioned business accommodate their differing requirements training and promotion of diverse income- due to varying levels of literacy, language generating activities. skills and technical sophistication (see Appendix 3 for the list of the technical A wide range of livestock practices and stakeholders participants). Participants in interventions were mentioned (18 times), the meeting with technical stakeholders including expanded fodder production were briefed on the discussions with through paddocking8 or storing acacia seeds. community representatives. No major Raising awareness on timely sale and trading gaps or disparities were found between of livestock was also regularly mentioned. the views of the two groups. The main
8 This involves fencing off land to grow pasture for use and/or harvest during dry seasons.
138 Understanding Community Resilience feedback is included here to add of local natural resources, leading understanding and context to the findings. communities to feel they accrue limited profit out of them; and (3) • Participants felt that the ranked negative feelings with regard to resilience characteristics resonated wildlife conservation, as people with the reality in these communities. tend to associate wild animals with They said the statements prioritized killing of people and destruction of by the communities were what they crops, with limited compensation expected, although they felt that a few coming from the Kenya Wildlife characteristics were over- or under- Service. prioritized: - Extension services: Though - Road networks: The community communities did not rank extension felt that roads (ranked second by services highly, particularly in the the focus groups) should not have livestock sector, the participants received such a high ranking in in the feedback session felt that comparison to other statements. extension services are critical They felt the value was probably and need to be emphasized in skewed by the strong influence of this county, with its dependence the youth, and that this score did on livestock rearing. Participants not necessarily represent broad felt that the main reason for community views. They said that the low priority was community youth generally aspire to be more knowledge about animal health mobile, especially with alternative and the accessibility of private/ transport methods such as NGO services and facilities. As a motorbikes, which can also provide result, there is limited use of formal employment opportunities. government extension services, - Employment: The feedback which are largely restricted to workshop felt that employment vaccination. should have ranked higher, given • Participants agreed with the resilience- the many unemployed youth in the building measures cited. They county. acknowledged communities’ reduced - Natural resource management: focus on livestock interventions due Natural resource management to the shift in livelihoods from purely is critical to Kajiado county as it pastoral to a mix of income-generating supports many livelihoods and activities. Nevertheless, livestock the county’s economy, through support services were highlighted both wildlife tourism and pasture- as the critical measure. They also pastoralism in national parks recognized that the ultimate outcome (e.g., Amboseli) and community of resilience building will result from conservancies. Yet the assessment a combination of linked drivers rather failed to prioritize this characteristic. than independent processes or Feedback workshop participants felt interventions. that it should have been among the • Participants felt that the characteristics top five resilience characteristics. of a resilient household – employment, Some of the reasons suggested for diversified income generating the low priority of natural resource activities and large herd size – were management were (1) lack of accurate. They had no additions. recognition of natural resources, • Participants highlighted the following which are taken for granted based specific recommendations related to on the assumption that they have the CoBRA methodology: always been there and will always - It is important to recognize that the be; (2) limited community ownership findings are informed by seasons
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 139 and could vary if the focus groups 5. Conclusions and were held in a different season. recommendations - The composition of the focus groups should be as balanced as possible 5.1 Conclusions in terms of specific age, gender and livelihood groups, in order to avoid Some of the key findings from the FGDs: any bias that could result from an imbalance in group representation. • The top five statements used to • Participants highlighted the following describe a resilient community by specific recommendations related to focus group participants in Kajido the CoBRA findings: were education (score 608), roads - Conduct a follow-up study to better (434), water for humans (403), health understand resilience within peri- care (329) and water for livestock urban areas in Kajiado county. (208). The extremely high ranking of - Incorporate the findings of the education reflects participants’ linkage CoBRA analysis into the county of educational attainment and income drought contingency plan levels and diversity. Roads are also revision process in preparation highly rated given that they provide for operationalizing the national access to Nairobi – the largest market drought contingency fund. in the country. - Use the CoBRA report to inform • Roads are particularly highly ranked integrated development planning by youth, who also rated employment processes in the county. equally with education and water. This - Tailor upcoming project concepts reflects their focus on markets and and interventions to address the jobs, which are highly dependent on priority interventions identified by good access to urban areas. Reducing the communities, some of which journey times to Nairobi is likely to require long-term commitments improve economic development in such as education. Kajiado. - As an analysis and planning tool, • When asked to describe the specific repackage the CoBRA analysis characteristics of resilient households report in a tailored form (short in their community, participants briefs based on the needs and consistently focused on income and interests of the audience), share financial assets. Resilient households them with NDMA for review, were described as having higher disseminate them to the County incomes because they benefited from Development Office and other a combination of income-generating/ stakeholders, and use them in business activities, access to training county government officials employment and/or a large herd and/ to better understand community or farm. The diversity of these income perspectives and needs. The report and asset sources is the critical factor, should also serve as a blueprint as it enables households to spread for county strategic planning their risk, protecting against shocks. processes. This should be done in Drought was seen as a manageable collaboration with NGOs and other shock when livestock is managed partners to support a partnership well and agricultural production is approach to implementation. irrigated. This finding was consistent - Develop a tailored project that with the comments of key informants targets some of the most optimistic from resilient households, whose groups, including women and access to multiple incomes and a solid agriculture livelihood groups, to asset base (in terms of livestock herds push for rapid impacts. or farm size) was cited as the primary
140 Understanding Community Resilience reason for their resilience. Access to remain a critical component of the Kajiado education and credit was consistently economy. Lack of higher level education described as the direct means to and poor transport links have been acquire more income and larger assets identified as critical barriers to realizing and hence was the key driving factor this potential, hence the high priority to achieve resilience. given to both as characteristics of • Overall, focus groups scored their resilience. attainment of the highly rated resilience characteristics as low, with 5.2. Recommendations an overall resilience score of 3.5 out 10. However, the drop in scores during • The high priority given to a relatively crisis periods, to an average of 2.3 small set of issues as key characteristics out of 10, was not seen as terribly of resilience should be used to inform significant. strategic planning in this area. In • Urban and agricultural groups scored planning and prioritizing investment themselves as the most resilient in Kajiado, priority attention should be relative to other groups (4.5 and 4.0 given to the following: respectively in the current period). - Expand access to quality Overall, groups were optimistic, with education: Communities recognize 63 percent agreeing that the number the strong link between education of resilient households was growing. levels and employment and Pastoral and agricultural groups were business opportunities. Authorities the most likely to state that resilience should examine the quantity and was increasing (73 percent and 86 quality of secondary and tertiary percent respectively). education facilities available in the • There is significant overlap between county. Secondary education is not the interventions that are most highly free in Kenya; consideration should rated in helping to build resilience and be given to efforts that support the the most highly rated characteristics poorest and least educated groups of resilience. Education, water, health to successfully complete higher and roads were four of the top five level education. highly rated interventions. Resilient - Improve the road network: Access households also cited educational to markets and employment, interventions as important but particularly in and around Nairobi, generally focused on a different set is highly constrained by the poor of priority interventions. Resilient quality of roads. Recent construction households give much higher priority of new roads, such as the Nairobi- to savings and credit, business skills Namanga road, benefits parts of the and agricultural and livestock inputs. county but also serves to highlight This reflects a more commercial focus the state of most other roads. Roads on livelihoods by these households. are not cheap to construct, but their economic and social benefits are The county’s relative proximity to Nairobi well documented. is a distinct advantage that has supported - Improve consistent access to the commercialization of both pastoral water: Households in urban areas and agricultural production, often spoke frequently about improved through small-scale irrigation. Access to access to piped water, while those Nairobi provides a wide range of other in rural pastoral areas stressed employment and market opportunities. the importance of boreholes, and It is clear that communities recognize the those in agricultural regions noted importance of livelihood diversification, that installation of irrigation water is although pastoral production systems key. At present, 66 percent of the
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 141 county’s population has access to consistently and uniformly used to improved water sources.9 describe resilient households. In a - Increase agricultural productivity: resilient community, the vast majority As agriculture expands in the of households will have achieved this district, there are large tracts of level of financial security. Ultimately horticultural and flower farms, this should be the impact indicator of individual farm greenhouses and a successful resilience strategy. irrigated farms, which dominate • The community perspective must in the Loitoktok area. It has grown be incorporated into efforts to plan, tremendously due to construction implement and monitor resilience and of an access road that created development in Kajiado. The assessment opportunities to market and trade findings show that Kajiado communities produce with other parts of Kajiado are increasingly changing their lifestyles and Nairobi. It also supports a good and livelihoods to improve their long- number of smallholder farmers. This term and short-term resilience. The area offers tremendous opportunity area does not benefit from large- to build resilience among a large scale humanitarian or development population. The community should programmes, so most resilient be supported in terms of receiving households have achieved this status by help to improved irrigation and their own efforts. Their strategies should water management practices; be more closely monitored and copied to agricultural inputs such as certified expand opportunities to more marginal seeds, fertilizers and pest control households. Stakeholders involved in measures; and packaging and building resilience need to be aware of marketing. the wider spectrum of factors that affect - Enhance access to credit: Lack resilience and consider how they can re- of credit has emerged as a key focus efforts and budgets in these areas. barrier that prevents households • Ongoing monitoring and evaluation from investing in their agricultural will be critical to measuring changes (and pastoral) production and in resilience. The priority issues other income-generating activities. highlighted by communities in the Providing credit is a relatively cheap CoBRA assessment can be used by the intervention that can be undertaken Government and other stakeholders to by multiple actors. It should be identify key indicators that can be used coupled with business support and to track trends in resilience. Undoubtedly advice to enhance the success rate some of these indicators are already of micro and small business ventures. being measured as part of ongoing data Implementers must bear in mind collection, such as household income that widespread success of such levels, households with members who interventions in building resilience have completed secondary education depends on parallel provision of and kilometres of tarmac road. Others other, longer term interventions. may need to be incorporated into data • A key criterion in identifying and collection, such as number of household prioritizing future interventions income sources or number of households should be the extent to which they accessing a savings and credit build and diversify incomes and mechanism. Other issues would involve assets either directly or indirectly. some level of consensus to agree to an Financial security through one or more appropriate indicator and method of business incomes, wage employment collection, such as consistent access to and large farms and/or herds was quality water supply.
9 Kenya National Bureau of Statistics and Society for International Development, op. cit.
142 Understanding Community Resilience Appendix 1. Participants in CoBRA Kajiado assessment
Phone No. Name Organization Email address number
1 Laban MacOpiyo UNDP DDC 0728 466 622 [email protected]
2 Francis Opiyo UNDP DDC 0727 35 3670 [email protected]
3 Francis Koma NDMA 0721 279 580 [email protected]
4 Marion Mutio NDMA 0720 794 042 [email protected] Joseph
5 Alfred Mwangi NDMA 0727 258 464 [email protected] Kathare
6 Agnes Oningo Ministry of Agriculture 0725 313 980 [email protected] (Crops)
7 Florence Kasimu Ministry of Agriculture 0724 982 653 [email protected] or (Livestock and Fisheries) [email protected]
8 James Peli Ministry of Health 0713 292 670 [email protected]
9 Gedion Mbinda Ministry of Education 0714 970 609
10 Shadrack Mutiso Welthungerhilfe (GAA) 0722 669 673 [email protected] or [email protected]
11 George Meibuko Welthungerhilfe (GAA) 0716 754 207 [email protected]
12 Kitipa Naikuni World Vision 0721 727 526
13 Alais Kisota Kenya Red Cross 0722 118 183 [email protected]
14 Susan Siammy Kenya Red Cross 0726 594 760 [email protected]
15 Joyce Saiko Neighbours Initiative 0721 324 194 [email protected] Alliance
16 Charles Masingira Diocese of Kajiado 0722 594 434
17 Patrick Mainka African Medical and 0723 994 395 [email protected] or Research Foundation [email protected] (AMREF)
18 Winnie Wemo AMREF 0717 822 832 [email protected]
19 Sylvia Sumare Center for Sustainable 0722 817 177 [email protected] Drylands Ecosystems and Societies
20 Thomas Tulito Child Fund 0725 953 628 [email protected] Turere
21 Anne Ayuma Dioceses of Ngong 0722 936 099 [email protected]
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 143 Appendix 2. Resilience statements – full list of scores
Resilience Bean Total bean Full resilience statement10 characteristic score score
Financial
Access to credit People would have good access to affordable credit and 139 649 would be saving money (through banks, microfinance organizations, community savings and credit).
Productive farms Farmers would be more productive and profitable (i.e., 126 they would have inputs like quality tools, oxen, fertilizers and improved knowledge of good farming practices).
Employment There would be many opportunities for jobs and other 106 forms of paid employment through government, factories and other businesses.
Diversified income- Many households would be involved in other IGAs such 82 generating activities as small businesses and trading.
Health care for The community would have access to high-quality and 81 livestock affordable animal health services, including veterinary services and vaccinations, whenever they need them.
Livestock herds Pastoralists would have herds large enough to 51 sustainably support their families.
Assets 27
Land Everyone would have secure access to/ownership of 21 land/property.
Pasture and fodder There would be sufficient pasture (or fodder) for 16 livestock at all times of the year.
Human
Education All children would be able to complete primary/ 608 1,013 secondary/tertiary education.
Health care for The community would have access to quality and 329 humans affordable basic health care locally.
Food security All households would be able to feed themselves well 56 every day.
Disaster risk 11 management
Special needs 9
Natural
Wildlife conservation The natural environment and wildlife would be 53 94 conserved for tourism and other purposes.
Forests Trees and forest cover would be well and sustainably 41 managed to provide various services for future generations.
144 Understanding Community Resilience Resilience Bean Total bean Full resilience statement10 characteristic score score
Physical
Roads There would be good-quality roads to the community. 434 1,812
Water for humans The whole community would have access to sufficient, 403 good-quality water at all times of the year.
Water for livestock Livestock would have access to sufficient water at all 208 times of the year.
Access to markets The community would have easy access to markets to 185 buy goods and sell their produce.
Irrigation Farmers would be irrigating land to improve the 139 production of crops for consumption and sale.
Sanitation Everyone would have good sanitation. 139
Electricity The community would have access to affordable 105 electricity supply.
Shelter Everyone would live in good-quality housing. 89
Telecommunications There would be a reliable mobile phone network to all 58 communities all the time.
Manufacturing 52
Social
Peace and security The whole community would enjoy continual peace and 119 139 security.
Good governance The community would be served by efficient and 13 non-corrupt community leaders and management structures.
Charity 6
Women Women would be fully involved in local development 1 and leadership.
10 Note that blank statements indicate statements that were added by communities during FGD discussions, and as such did not have a generic full statement associated with them.
Kajiado Central, Mashuru, Loitoktok, Kajiado North and Isinya districts 145 Appendix 3. Participants in CoBRA Kajiado technical stakeholders’ feedback session
No. Name Organization Email address
1 Francis Koma NDMA [email protected]
2 Alfred Mwangi Kathare NDMA [email protected]
3 Andrew K. Mwaswasi NDMA [email protected]
4 Agnes Oningo Ministry of Agriculture (Crops) [email protected]
5 Kamoni Njoroge Ministry of Agriculture [email protected]
6 Florence Kasimu Ministry of Agriculture (Livestock [email protected] and Fisheries)
7 Lawrence Mgambo Ministry of Health
8 Shadrack Mutiso Welthungerhilfe (GAA) [email protected]
9 Kitipa Naikuni World Vision
10 Alais Kisota Kenya Red Cross [email protected]
11 Joyce Saiko Neighbours Initiative Alliance [email protected]
12 Sylvia Sumare Center for Sustainable Drylands [email protected] Ecosystems and Societies
13 Stephen Moiko Center for Sustainable Drylands [email protected] Ecosystems and Societies
14 Thomas Tulito Turere Child Fund [email protected]
15 Anne Ayuma Dioceses of Ngong [email protected]
16 Dorcas Njoroge Ministry of Water [email protected]
17 Peter Kasanne [email protected]
18 Stephen Maenji [email protected]
19 Faith Mpoke Action Aid [email protected]
20 Daniel Taama Anglican Church of Kenya [email protected]
21 TJ Somoire Child Fund [email protected]
22 Hosea Kandagor Nasim Women group [email protected]
23 Clammy Susan [email protected]
24 Evelyn Parsnip [email protected]
25 Priscilla Amiri European Commission Directorate [email protected] General for Humanitarian Aid and Civil Protection (ECHO)
26 Judith Munyao ECHO [email protected]
27 Laban MacOpiyo UNDP DDC [email protected]
28 Francis Opiyo UNDP DDC [email protected]
29 Yuko Kurauchi UNDP DDC [email protected]
30 Ruth Mwathi UNDP DDC [email protected]
146 Understanding Community Resilience United Nations Development Programme Drylands Development Centre United Nations Avenue, Gigiri P.O. Box 30552 Nairobi, 00100, Kenya www.undp.org/drylandscentre Email: [email protected]
Understanding Community Resilience: Findings from Community-Based Resilience Analysis (CoBRA) Assessments
Marsabit, Turkana and Kajiado counties, Kenya and Karamoja sub-region, Uganda Understanding Community Resilience: Findings from Community-Based Resilience Analysis (CoBRA) Assessments Analysis (CoBRA) from Community-Based Resilience Findings Understanding Community Resilience: