Household Waste Sorting at the Source

ThesisThesis for forthe the Degree Degree of Doctor of Doctor of Philosophy of Philosophy

HOUSEHOLDHOUSEHOLD WASTE WASTE SORTING SORTING ATAT THE THE SOURCE SOURCE A Aprocedure procedure for for improvement improvement

KamranKamran Rousta Rousta

SwedishSwedish Centre Centre for forResource Resource Recovery Recovery

UNIVERSITYUNIVERSITY OF OFBORÅS BORÅS

Borås,Borås, Sweden Sweden 2018 2018 Abstract

is a challenging task that involves both social and technical aspects. The intersection of these two aspects is a key part of source separation i.e., separating the waste where it is generated.

Some Swedish stakeholders, including recycling companies and landlords, felt that

material recycling was challenging due to the lack of an analysis tool that can help improve the source separation system at the local level. The central point of this thesis focuses on creating an analysis tool that can be used to improve sorting at the source in any waste

collection system. The objectives were to answer the following research questions: 1) How can the current sorting behaviour be evaluated? 2) How can appropriate interventions for

improving this behaviour be identified? 3) How can the effect of interventions be assessed?

A participatory research based on Action Research framework (Look, Think, Act) was

designed in a pilot area in the city of Borås, Sweden. Data was collected based on a combination of quantitative and qualitative methods since recycling behaviour is a complex phenomenon. Pick analysis (waste composition study) was found to be a relevant method to Copyright © Kamran Rousta evaluate the recycling behaviour and assess the effects of interventions in a recycling scheme. Similarly, using interviews strengthened the understanding of the need of inhabitants to Swedish Centre for Resource Recovery University of Borås participate in the source separation system in order to identify effective interventions. A SE‐501 90 Borås, Sweden Recycling Behaviour Transition (RBT) procedure was developed based on the pilot study. Digital version: http://urn.kb.se/resolve?urn=urn:nbn:se:hb:diva‐12821 The RBT procedure proved to be an effective tool for improving recycling behaviour. Also, ISBN 978‐91‐88269‐68‐3 (printed) the procedure was designed for being adaptable for any waste separation system. It was ISSN 978‐91‐88269‐69‐0 (pdf) ISSN 0280‐381X, Skrifter från Högskolan i Borås, nr. 84 concluded that the factors influencing recycling behaviour are complex and sensitive to local circumstances and individual perceptions. Therefore, the RBT procedure helps identify the Cover Photo: City of Borås, by Kamran Rousta appropriate interventions based on the local context. Printed in Sweden by Responstryck AB Borås 2018 Data collected in the pilot area revealed that the convenience factors in recycling behaviour are crucial for improving participation in the recycling scheme. By placing recycling containers close to the vicinity of waste generation, which is a convenience factor,

iii ii

Abstract

Sustainable waste management systems are needed to handle the increasing amount of household waste across the globe. EU Waste Hierarchy as a strategy could manage this issue, where reducing the waste is preferred to recycling which is, in turn, preferred to generating energy from the waste. Increased material recycling in a household waste management system is a challenging task that involves both social and technical aspects. The intersection of these two aspects is a key part of source separation i.e., separating the waste where it is generated.

Some Swedish stakeholders, including recycling companies and landlords, felt that material recycling was challenging due to the lack of an analysis tool that can help improve the source separation system at the local level. The central point of this thesis focuses on creating an analysis tool that can be used to improve sorting at the source in any waste collection system. The objectives were to answer the following research questions: 1) How can the current sorting behaviour be evaluated? 2) How can appropriate interventions for improving this behaviour be identified? 3) How can the effect of interventions be assessed?

A participatory research based on Action Research framework (Look, Think, Act) was designed in a pilot area in the city of Borås, Sweden. Data was collected based on a combination of quantitative and qualitative methods since recycling behaviour is a complex phenomenon. Pick analysis (waste composition study) was found to be a relevant method to evaluate the recycling behaviour and assess the effects of interventions in a recycling scheme. Similarly, using interviews strengthened the understanding of the need of inhabitants to participate in the source separation system in order to identify effective interventions. A Recycling Behaviour Transition (RBT) procedure was developed based on the pilot study. The RBT procedure proved to be an effective tool for improving recycling behaviour. Also, the procedure was designed for being adaptable for any waste separation system. It was concluded that the factors influencing recycling behaviour are complex and sensitive to local circumstances and individual perceptions. Therefore, the RBT procedure helps identify the appropriate interventions based on the local context.

Data collected in the pilot area revealed that the convenience factors in recycling behaviour are crucial for improving participation in the recycling scheme. By placing recycling containers close to the vicinity of waste generation, which is a convenience factor,

iii the miss-sorted fraction decreased by up to 28% (wt%). Similarly, easy access to relevant Preface information is another intervention that significantly improved the recycling behaviour in the pilot area. Furthermore, it was concluded that Action Research, as a participatory methodology, is particularly appropriate when seeking the link between social and technical Nearing completion, I thought why did I choose this topic and what was my thought factors in waste management. processing, when I wanted to focus this particular area of research. The answer lies within my roots when I started my first summer job in my father’s factory. I was a practitioner, learning Keywords: recycling behaviour, source separation, waste management, resource recovery, different types of machinery and finding solutions to complex problems was sowed in when I household waste, waste composition study, waste sorting, waste collection, was 12. Later, I became an entrepreneur after graduating from Mechanical Engineering; separate collection, municipal solid waste, circular economy designing and construction of boilers, tanks and steel structures. The engineer in me learnt to

solve and analyse problems from real-life experiences. For personal reasons, I had shut the

business and moved to Sweden to start schooling again after 10 years. Moving to Sweden changed my life radically where I found myself as a changer.

My master thesis reviewed solid waste management system in Borås that opened a new

window of opportunities. I concluded that waste sorting at home requires a system/tool that involves the end users i.e., inhabitants for an efficient operation. Personally, I like to solve a problem and see the changes in the society than just identifying it. Kurt Lewin, the German- American psychologist in Action Research once said: “No research without action, no action

without research” and this motto represented my research. The different hats which I wore in the early part of my life including the roles of practitioner, engineer and changer helped me in grounding this research. I used existing methodologies to solve a particular problem, while

I tried understanding different theories from a scientific perspective.

This thesis was performed wearing those different hats that present the rationale for this study, justification of methods, followed by how to conduct them, and a discussion of

important finding with remarks and conclusion. Some by-products were obtained during the

course of this thesis including the pilot area got new garbage bins, a facility for waste sorting closer to their apartment building, barbeque place and play area for children. The stakeholders including property owner, waste management company and inhabitants have started to harvest the fruits out of this research. Moreover, neighbouring cities also started using the results as a

part of their waste management activity.

This study equally impacted me as a researcher learning qualitative methods, analysing data from a perspective I never looked into. I’m not a social scientist, however; I tried understanding their methods and draw an applicable conclusion out of it. The engineer in me

v iv

Preface

Nearing completion, I thought why did I choose this topic and what was my thought processing, when I wanted to focus this particular area of research. The answer lies within my roots when I started my first summer job in my father’s factory. I was a practitioner, learning different types of machinery and finding solutions to complex problems was sowed in when I was 12. Later, I became an entrepreneur after graduating from Mechanical Engineering; designing and construction of boilers, tanks and steel structures. The engineer in me learnt to solve and analyse problems from real-life experiences. For personal reasons, I had shut the business and moved to Sweden to start schooling again after 10 years. Moving to Sweden changed my life radically where I found myself as a changer.

My master thesis reviewed solid waste management system in Borås that opened a new window of opportunities. I concluded that waste sorting at home requires a system/tool that involves the end users i.e., inhabitants for an efficient operation. Personally, I like to solve a problem and see the changes in the society than just identifying it. Kurt Lewin, the German- American psychologist in Action Research once said: “No research without action, no action without research” and this motto represented my research. The different hats which I wore in the early part of my life including the roles of practitioner, engineer and changer helped me in grounding this research. I used existing methodologies to solve a particular problem, while I tried understanding different theories from a scientific perspective.

This thesis was performed wearing those different hats that present the rationale for this study, justification of methods, followed by how to conduct them, and a discussion of important finding with remarks and conclusion. Some by-products were obtained during the course of this thesis including the pilot area got new garbage bins, a facility for waste sorting closer to their apartment building, barbeque place and play area for children. The stakeholders including property owner, waste management company and inhabitants have started to harvest the fruits out of this research. Moreover, neighbouring cities also started using the results as a part of their waste management activity.

v learnt the social science that skilled and excelled my abilities to a new maximum. I’m an early List of Publications researcher, however, this study had given me the confidence to do multidisciplinary research combing different methods and find solutions to more complicated problems. This study is a win-win for different stakeholders including me as a researcher, waste management company I. Rousta K, Ordoñez I, Bolton K, Dahlén L, (2017) Support for Designing Waste Sorting Systems: A Mini Review. Waste management & Research 35(11): 1099-1111. and property owner who got the tools/system for effective waste management, and inhabitants who got better facilities. As a researcher, you want to see your solutions work out of the II. Rousta K, Ekström K.M, (2013) Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City. Sustainability 5(10): 4349-4361. laboratory and the satisfaction you get when it is welcomed by all is priceless and I’m proud of it. III. Rousta K, Bolton K, Lundin M, Dahlén L, (2015) Quantitative Assessment of Distance to Collection Point and Improved Sorting Information on Source Separation of Household Waste. Waste Management 40: 22-30. IV. Rousta K, Bolton K, Dahlén L, (2016) A Procedure to Transform Recycling Behavior for Source Separation of Household Waste. Recycling 1(1): 147-165

Kamran Rousta

vii vi

List of Publications

I. Rousta K, Ordoñez I, Bolton K, Dahlén L, (2017) Support for Designing Waste Sorting Systems: A Mini Review. Waste management & Research 35(11): 1099-1111.

II. Rousta K, Ekström K.M, (2013) Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City. Sustainability 5(10): 4349-4361.

III. Rousta K, Bolton K, Lundin M, Dahlén L, (2015) Quantitative Assessment of Distance to Collection Point and Improved Sorting Information on Source Separation of Household Waste. Waste Management 40: 22-30.

IV. Rousta K, Bolton K, Dahlén L, (2016) A Procedure to Transform Recycling Behavior for Source Separation of Household Waste. Recycling 1(1): 147-165

vii

Statement of Contributions

Paper I: I and Isabel Ordoñez had equal responsibility for the idea, literature review, data analyses, manuscript preparation and its revision.

Paper II: I was responsible for the idea, experimental design, data collection, statistical analyses, manuscript preparation and its revision.

Paper III: I was responsible for the idea, experimental design, data collection, interpreting the results, manuscript preparation and its revision.

Paper IV: I was responsible for the idea, data collection, analyses, manuscript preparation and its revision.

ix viii

Statement of Contributions

Paper I: I and Isabel Ordoñez had equal responsibility for the idea, literature review, data analyses, manuscript preparation and its revision.

Paper II: I was responsible for the idea, experimental design, data collection, statistical analyses, manuscript preparation and its revision.

Paper III: I was responsible for the idea, experimental design, data collection, interpreting the results, manuscript preparation and its revision.

Paper IV: I was responsible for the idea, data collection, analyses, manuscript preparation and its revision.

Publications Not Included in This Thesis

Conference paper: . Rousta K, Kaeni Moghadam M, Haji Karimkhani P, Richards T, (2011) Evaluation the inhabitants’ participation in separation at source by waste characterization. WASTES: Solutions Treatments Opportunities Conference, Guimarães, Portugal.

Book chapters:

1. Rousta K, Dahlén L, (2015) Source separation of Household Waste: Technology and Social Aspects, in Resource Recovery to approach zero municipal wastes, M. J. Taherzadeh and T. Richards, 07/2015: chapter 4; CRC Press., ISBN: 9781482240351 .

2. Rousta K, Richards T, Taherzadeh M.J, (2015) An Overview of Solid Waste Management towards Zero Landfill: A Swedish Model, in Resource Recovery to approach zero municipal wastes, M. J. Taherzadeh and T. Richards, 07/2015: chapter 1; CRC Press., ISBN: 9781482240351.

Article:

. Brancoli P, Rousta K, Bolton K, (2017) Life cycle assessment of supermarket food waste. Resources, Conservation and Recycling 118:39-46.

xi x

Publications Not Included in This Thesis

Conference paper:

. Rousta K, Kaeni Moghadam M, Haji Karimkhani P, Richards T, (2011) Evaluation the inhabitants’ participation in separation at source by waste characterization. WASTES: Solutions Treatments Opportunities Conference, Guimarães, Portugal.

Book chapters:

Article:

. Brancoli P, Rousta K, Bolton K, (2017) Life cycle assessment of supermarket food waste. Resources, Conservation and Recycling 118:39-46.

Acknowledgment

I had succeeded many challenging tasks during my lifetime alone, nonetheless, the toughest was to finish PhD thesis that couldn’t have happened without the supporting pillars at difficult times. There are many people who supported me directly or indirectly in this journey that I take the opportunity to thank them for their help.

MY SPECIAL THANKS TO:

My supervisors, Kim Bolton and Lisa Dahlén who accepted to supervise my research.

My enthusiasm towards research accelerated exponentially over the years with you beyond horizons. I had difficulty finding words to express my gratitude and you’re a think-tank. I still cherish my memories, when I met Lisa for the first time at Luleå and she accepted to be my supervisor. I remember her words “Yes, I’ll supervise you” with a smile is evergreen. Thanks

for accepting me Lisa, your constructive feedback and the way you express to me, was so valuable that converted my confusion to confidence. Kim, your advice is invaluable that kept my research to the high standards. I’m fortunate to have you both as my PhD Supervisors and

your suggestions were crackerjack. I can’t envisage finishing this thesis without your support,

help, advice and encouragement. My examiner, Mohammad Taherzadeh who supported my proposal and pushed me to

kick-start PhD program. I appreciate the encouragement and advice you gave during my PhD,

which is precious. I couldn’t forget the day when I presented my idea in five minutes to you, and after some silent seconds, you said: “go and do it”.

My research and funding partner, Borås Energi och Miljö AB (BEM) who supported

my research in cash and in-kind. Some key people at BEM who supported this project include Hans Skoglund (retired), and Anna-Karin Schön. Thanks a million for the countless constructive discussions about the results of the project, and support. Pär Carlsson and co- workers at BEM trusted me and supported all the stages of my research is greatly

acknowledged. Working with you was a great honour for me.

Another research partner, AB Bostäder i Borås, (property owners of the pilot area), and their personnel who helped in applying changes and interventions in the pilot area.

Without your support, it was impossible to complete this study.

xiii xii

Acknowledgment

MY SPECIAL THANKS TO:

My supervisors, Kim Bolton and Lisa Dahlén who accepted to supervise my research. My enthusiasm towards research accelerated exponentially over the years with you beyond horizons. I had difficulty finding words to express my gratitude and you’re a think-tank. I still cherish my memories, when I met Lisa for the first time at Luleå and she accepted to be my supervisor. I remember her words “Yes, I’ll supervise you” with a smile is evergreen. Thanks for accepting me Lisa, your constructive feedback and the way you express to me, was so valuable that converted my confusion to confidence. Kim, your advice is invaluable that kept my research to the high standards. I’m fortunate to have you both as my PhD Supervisors and your suggestions were crackerjack. I can’t envisage finishing this thesis without your support, help, advice and encouragement.

My examiner, Mohammad Taherzadeh who supported my proposal and pushed me to kick-start PhD program. I appreciate the encouragement and advice you gave during my PhD, which is precious. I couldn’t forget the day when I presented my idea in five minutes to you, and after some silent seconds, you said: “go and do it”.

My research and funding partner, Borås Energi och Miljö AB (BEM) who supported my research in cash and in-kind. Some key people at BEM who supported this project include Hans Skoglund (retired), and Anna-Karin Schön. Thanks a million for the countless constructive discussions about the results of the project, and support. Pär Carlsson and co- workers at BEM trusted me and supported all the stages of my research is greatly acknowledged. Working with you was a great honour for me.

Another research partner, AB Bostäder i Borås, (property owners of the pilot area), and their personnel who helped in applying changes and interventions in the pilot area. Without your support, it was impossible to complete this study.

xiii

Other research partners, the tenants in the pilot area who participated in the My financiers, Sjuhärads kommunalförbund and Sparbanksstiftelsen Sjuhärad for interviews and all the stages of my research. Thanks for your kindness through email or in their partial funding of this research. person. Your engagement and participation in the research were the main reason for me to Finally, I want to have an unusual kudos to myself. Doing the PhD is a tough job work harder. Special thanks to Mona who helped me in finding the interviewees and Osman, especially when you have teaching and research load of 50-50. When I got burnout last year, who encouraged the inhabitants to use the Environmental Room in the pilot area to sort the only writing the thesis was left. I was so close to giving up because of my health conditions waste correctly. and I’m here today succeeding it. The ups and downs in the PhD gave life experiences which Mobile Info Center, Masoud, Esmieel and Ismet, who supported me in practical stages I can use at any point of time in my life. Thank you, Kamran for a good job! of my research in the pilot area.

Pick analysis group that helped me in conducting the pick analysis in addition to the constructive discussion after each pick analysis for the miss-sorted fractions. Your valuable activity was the most important part of this thesis. Thank you so much, guys! My research friends, Magnus Lundin and Isabel Ordoñez who shared their knowledge and supported me in my research. The never-ending discussions with you made me think outside the box. My immediate boss, Peter Therning who helped me in planning my time for both research and teaching.

Director of studies Resource Recovery Doctoral program, Tomas Wahnström and his administrative team that prepared lot of administration works and helped with the documentation for my PhD.

All my colleagues at the Engineering School, who encouraged me and discussed many aspects of my research in an inspiring way. Thanks to Anita Pettersson, for your support and encouragement, especially the time we shared our office.

All PhD students at Swedish Centre for Resource Recovery (SCRR), who graduated earlier and who are current working; thanks for your encouragement.

My most important in-direct supporters, my beautiful wife Mitra and my sweet daughters Sara and Dina; who let me use my time for this research instead to be with them. I am getting short of words to say thank you for your support and understanding. I hope to spend more time with you in the future and compensate this debt.

My lovely parents, Kazem and Mehri who taught me patience, perseverance and willingness, beyond the rest. You’re far away from me, but your energy was always with me.

xv xiv

My financiers, Sjuhärads kommunalförbund and Sparbanksstiftelsen Sjuhärad for their partial funding of this research.

Finally, I want to have an unusual kudos to myself. Doing the PhD is a tough job especially when you have teaching and research load of 50-50. When I got burnout last year, only writing the thesis was left. I was so close to giving up because of my health conditions and I’m here today succeeding it. The ups and downs in the PhD gave life experiences which I can use at any point of time in my life. Thank you, Kamran for a good job!

Abbreviations

BC Before Christ EU European Union FTI Förpackning & Tidning Insamlingen MOAB Motivation-Opportunity-Ability-Behaviour

MRF Material Recovery Facility OECD Organisation for Economic Co-operation and Development RBT Recycling Behaviour Transition SEK Swedish Krona SFS Svensk författningssamling

SMED Svenska MiljöEmissionsData TPB Theory of Planned Behaviour TRA Theory of Reasoned Action

UK United Kingdom

wt% Percentage by weight

xvii xvi

Abbreviations

BC Before Christ EU European Union FTI Förpackning & Tidning Insamlingen MOAB Motivation-Opportunity-Ability-Behaviour MRF Material Recovery Facility OECD Organisation for Economic Co-operation and Development RBT Recycling Behaviour Transition SEK Swedish Krona SFS Svensk författningssamling SMED Svenska MiljöEmissionsData TPB Theory of Planned Behaviour TRA Theory of Reasoned Action UK United Kingdom wt% Percentage by weight

xvii

Table of Contents

Abstract ...... iii

Preface ...... v

List of Publications ...... vii

Statement of Contributions ...... ix

Publications Not Included in This Thesis ...... xi

Acknowledgment ...... xiii

Abbreviations ...... xvii

Table of Contents ...... xix

1 Introduction ...... 1

1.1 The Issue of Waste ...... 1

1.2 Waste Management and Source Separation ...... 3

1.3 Understanding Recycling Behaviour ...... 7

1.4 Recycling Behaviour from a Theoretical Perspective ...... 8

1.5 Objectives ...... 12

2 Methods ...... 13

2.1 Methodology and Justification of the Selected Methods ...... 13

2.1.1 Why pick analysis? ...... 15

2.1.2 Why interview? ...... 16

2.2 Materials and Methods ...... 16

2.2.1 Pilot area ...... 17

2.2.2 Data collection methods ...... 18

2.2.2.1 Pick analysis ...... 18

2.2.2.2 Interview ...... 19 xix xviii

Table of Contents

Abstract ...... iii

Preface ...... v

List of Publications ...... vii

Statement of Contributions ...... ix

Publications Not Included in This Thesis ...... xi

Acknowledgment ...... xiii

Abbreviations ...... xvii

Table of Contents ...... xix

1 Introduction ...... 1

1.1 The Issue of Waste ...... 1

1.2 Waste Management and Source Separation ...... 3

1.3 Understanding Recycling Behaviour ...... 7

1.4 Recycling Behaviour from a Theoretical Perspective ...... 8

1.5 Objectives ...... 12

2 Methods ...... 13

2.1 Methodology and Justification of the Selected Methods ...... 13

2.1.1 Why pick analysis? ...... 15

2.1.2 Why interview? ...... 16

2.2 Materials and Methods ...... 16

2.2.1 Pilot area ...... 17

2.2.2 Data collection methods ...... 18

2.2.2.1 Pick analysis ...... 18

2.2.2.2 Interview ...... 19 xix

2.2.3 Design of study ...... 19 1 Introduction 3 Results and Discussion ...... 21

3.1 Development of a Tool to Improve Source Separation ...... 21 This chapter addresses the various concerns about waste1, which as general definition does 3.2 Evaluation of the Current Sorting Behaviour ...... 22 not have any economic or social value. Furthermore, it discusses the importance of source 3.3 Identifying the Appropriate Interventions for Improving Recycling Behaviour ...... 23 separation of waste in a waste management system, the theories of recycling behaviour that

3.4 Assessing the Effect of the Interventions ...... 25 can be used to understand this phenomenon, as well as the objectives and research questions of this study. 3.5 A Common Thread through Papers I-IV ...... 27

3.6 Sources of Error ...... 29 1.1 The Issue of Waste 4 Conclusions ...... 31 Waste is considered as any unwanted substance from human life. The term “waste” in 5 Outlook ...... 33 general, could be used in different descriptive terminologies such as garbage, trash, rubbish, References ...... 35 refuse, by-product, a rest product and discards which Mattsson (2003) has described in detail. No matter which terminology is used, waste is ever ingrained and inseparable from human life, and it requires proper handling for a better tomorrow. Barbalace (2003), in a report on the

history of waste, mentioned that native Americans generated 2.4 kg waste/day/person in 6500 BC, which were leftovers including bones from buffalos. This problem was solved by dumping the buffalos bones in nature at that time. Dumping the waste from nature to nature was a straightforward and reasonable solution. Nonetheless, waste became a problem after the modern day industrial revolutions with new urban life and civilisation, which have a different consumption pattern than the predecessor. The change in consumption patterns among humans led to the change in the amount of waste generated and, moreover, waste was generated from various streams such as packaging from plastic and paper, and hazardous and toxic waste. The transformation of the world is rapid and alarming in terms of population, consumption patterns, and socio-economic status based on the development in each country. The World Bank estimated that the municipal solid waste generation in the world will

1 Waste can be generated in different sources such as institutions, business, industries, households, construction and demolition sites, etc. Hoornweg and Bhada-Tata (2012) stated a defintion by OECD in their report that “Municipal waste is collected and treated by, or for municipalities. It covers waste from households, including bulky waste, similar waste from commerce and trade, office buildings, institutions and small businesses, yard and garden, street sweepings, contents of litter containers, and market cleansing. Waste from municipal sewage networks and treatment, as well as municipal construction and demolition is excluded”. The primary focus of this study is on household solid waste which is a part of municipal solid waste based on this definition. For this reason, general using of the word “waste” in this thesis refers to the “household solid waste”, otherwise it is identified clearly.

1 xx

1 Introduction

This chapter addresses the various concerns about waste1, which as general definition does not have any economic or social value. Furthermore, it discusses the importance of source separation of waste in a waste management system, the theories of recycling behaviour that can be used to understand this phenomenon, as well as the objectives and research questions of this study.

1.1 The Issue of Waste

Waste is considered as any unwanted substance from human life. The term “waste” in general, could be used in different descriptive terminologies such as garbage, trash, rubbish, refuse, by-product, a rest product and discards which Mattsson (2003) has described in detail. No matter which terminology is used, waste is ever ingrained and inseparable from human life, and it requires proper handling for a better tomorrow. Barbalace (2003), in a report on the history of waste, mentioned that native Americans generated 2.4 kg waste/day/person in 6500 BC, which were leftovers including bones from buffalos. This problem was solved by dumping the buffalos bones in nature at that time. Dumping the waste from nature to nature was a straightforward and reasonable solution. Nonetheless, waste became a problem after the modern day industrial revolutions with new urban life and civilisation, which have a different consumption pattern than the predecessor. The change in consumption patterns among humans led to the change in the amount of waste generated and, moreover, waste was generated from various streams such as packaging from plastic and paper, and hazardous and toxic waste. The transformation of the world is rapid and alarming in terms of population, consumption patterns, and socio-economic status based on the development in each country. The World Bank estimated that the municipal solid waste generation in the world will

1 increase from 3.5 million tonnes per day in 2010 to >6 million tonnes per day in 2025, which 1.2 Waste Management and Source Separation is almost doubled in a span of fifteen years and it needs acute attention (Hoornweg and A sustainable and integrated waste management system is a complicated process involving Bhada-Tata, 2012). Hoornweg et al. (2013) estimated that municipal solid waste generation various characteristics. These characteristics consist of a) reducing the environmental impacts; would rise from 9 to 12 million tonnes per day by 2100 which was based on civilisation b) assuring public health; c) developing techniques and treatments to encounter the waste growth. As Sub-Saharan Africa is yet to develop, it will be a major contributor. However, the issues that are economically feasible; d) organising appropriate infrastructure, service and waste generation in Europe and Central Asia is expected to remain constant (Hoornweg et al., communication channel for the inhabitants by understanding the needs of the society; e) 2013). Therefore, competent handling of wastes will be a vital part of infrastructures given the setting targets and goals for an efficient waste management planning and development fact that the amount and diversity of wastes are increasing dramatically. An efficient waste policies and regulations. Rousta et al. (2015) discussed the relationship between these aspects management is the need of the hour, and failing to do so will lead to severe consequences and their complexity in a book chapter in “Resource Recovery to Approach Zero Municipal including overloaded recipients of pollution (air, water, and soil), increasing depletion of Waste”. Developing a robust strategy is vital for a waste management system. In EU, this material resources, climate change, threat to energy supply, etc. strategy is defined as a waste hierarchy (Fig. 1), which determines the trajectories of the waste Global increase in household waste will increase the environmental impacts which management systems (EU., 2008). create new challenges for researchers to find solutions for those problems. It is worth mentioning that household solid waste is a minor fraction of the total waste generated. Nonetheless, it is a waste fraction that impacts the environment and must therefore be addressed. Swedish Environmental Protection Agency reported that 83% of the total waste generated in Sweden in 2014 were from mining industries and household wastes corresponds to negligible fraction (2.5%) (SMED, 2016). When mining wastes were kept aside, the household solid waste corresponds to 16% in the other waste category, which is the third biggest contributor within this category that needs attention. To supplement it, the diversity of the materials in the household waste makes the waste management complex. It is important to note the total amount of waste generated during goods production is significantly higher than the waste generated after its use. A report by Laurenti and Stenmarck (2015) on “products total waste” indicated the same, that the total amount of waste generated during production of Fig. 1. Schematics of a waste hierarchy model. a product in its life cycle is enormous compared to the waste generated after its use. For example, 85 kg worth of materials were wasted in the product chain to produce a mobile The most preferred option in a waste management system is to minimise the amount of waste phone worth of 169 g in weight. Similarly, although household waste seems a small fraction generated while the least option is landfilling. The intermediate options depend on the type of of the total waste, reducing and recycling it has a huge impact on the overall waste generated. waste that can be treated in the other steps of the hierarchy. All steps within the waste This could be mainly attributed to the large amounts of processed materials representing the hierarchy fulfil an important function. When it is possible to recycle certain types of waste, “tip of an iceberg” regarding resource depletion. This is why the household solid waste is an recycling is the preferred option in comparison with energy production methods such as important issue in the total material recirculation and circular economy in the world. This combustion or anaerobic digestion. The reason behind that is material recycling has positive shows how important it is for every municipality to plan, seek goals, and develop strategies impacts on climate change which was evidenced from a report by Nordic countries (Hillman and actions on handling and treating wastes. et al., 2015). This is similar to avoiding landfills for organic wastes as more valuable products

3 2

1.2 Waste Management and Source Separation

A sustainable and integrated waste management system is a complicated process involving various characteristics. These characteristics consist of a) reducing the environmental impacts; b) assuring public health; c) developing techniques and treatments to encounter the waste issues that are economically feasible; d) organising appropriate infrastructure, service and communication channel for the inhabitants by understanding the needs of the society; e) setting targets and goals for an efficient waste management planning and development policies and regulations. Rousta et al. (2015) discussed the relationship between these aspects and their complexity in a book chapter in “Resource Recovery to Approach Zero Municipal Waste”. Developing a robust strategy is vital for a waste management system. In EU, this strategy is defined as a waste hierarchy (Fig. 1), which determines the trajectories of the waste management systems (EU., 2008).

Fig. 1. Schematics of a waste hierarchy model.

The most preferred option in a waste management system is to minimise the amount of waste generated while the least option is landfilling. The intermediate options depend on the type of waste that can be treated in the other steps of the hierarchy. All steps within the waste hierarchy fulfil an important function. When it is possible to recycle certain types of waste, recycling is the preferred option in comparison with energy production methods such as combustion or anaerobic digestion. The reason behind that is material recycling has positive impacts on climate change which was evidenced from a report by Nordic countries (Hillman et al., 2015). This is similar to avoiding landfills for organic wastes as more valuable products

3 could be obtained through composting or anaerobic digestion processes. Landfilling is not always a bad choice and for some waste types, proper landfilling is the best and preferred treatment available. For example, some types of hazardous waste must be safely stored forever in controlled and closed landfills. For such materials, landfilling cannot be avoided as it is the final and only treatment at present in the circular economy for continuous detoxifying of resources. The challenge in the waste management system is to channel every single type of waste to its optimal level in the waste hierarchy. In addition, it is critical to ensure that no wastes end up as uncontrolled dumping or spreading, which is outside the waste hierarchy. The key point in this strategy is to convert the linear and traditional economy, i.e., production-consumption-disposal, to a circular economy which encourages recycling and reusing materials and recovering energy in their economic cycle as long as possible. In 2015, European Union developed an action plan to encourage circular economy among EU members by establishing goals from production to end of life cycle for a particular product including waste handling and management (European Commission, 2015). In many developed cities across the globe, waste hierarchy has been used as a policy Fig. 2. The interrelationship between functional elements in a municipal solid waste management, instrument to be recognised and applied efficiently. Tchobanoglous et al. (1993) defined the modified from Tchobanoglous et al. (1993). waste hierarchy in four levels from top to down: source reduction; recycling; waste As shown in Fig. 2, the source separation is the function when the user is involved in the transformation involving physical, chemical and biological treatment of the waste; and waste management system where the user contributes to the waste sorting system. It is landfilling. For implementation purposes, Tchobanoglous et al. (1993) introduced six therefore important to understand the importance of the source separation function both from functional elements within the integrated solid waste management. The interrelationship user and technical perspective and the importance of this function in the waste management between these six functional elements, from waste generation to disposal (Fig. 2), is crucial to system. Rousta et al. (2015) compared the oil refinery and waste refinery system to show how reach the top in a waste hierarchy. One may ask why this old model of waste management is important is to separate the waste, that it adds value in the further processes as well as still interesting. The answer is, from experience, different systems in Swedish municipalities, minimise the landfilling. They explained how the separation of waste can change a “waste” to incorporated this model either fully or partially and it is still implemented. In this model, the a value added “resource” (Rousta et al., 2015). waste separation at source appears to be the crucial part to reach preferred options in the There are different ways to separate different fractions in a waste stream. For example, waste hierarchy i.e., Recycle, Reuse and perhaps even Reduce. Waste separation at the source, sorting can be implemented in material recovery facilities (MRF) by sorting machines or also called source separation, is the process of separating different fractions of waste at the sorting several waste fractions at home by inhabitants (Rousta and Dahlén, 2015). Among place where it is generated, i.e., at home. these, waste separation at the source is, in principal, an easy and efficient way for solid waste management systems (Tchobanoglous et al., 1993). Looking at the waste management across the globe indicated that the cities which applied waste sorting system at the source decreased the landfilling dramatically and increased the recycling rate (Xevgenos et al., 2015). Within source separation, there are different systems such as kerbside collection, bring system or

5 4

Fig. 2. The interrelationship between functional elements in a municipal solid waste management, modified from Tchobanoglous et al. (1993).

As shown in Fig. 2, the source separation is the function when the user is involved in the waste management system where the user contributes to the waste sorting system. It is therefore important to understand the importance of the source separation function both from user and technical perspective and the importance of this function in the waste management system. Rousta et al. (2015) compared the oil refinery and waste refinery system to show how important is to separate the waste, that it adds value in the further processes as well as minimise the landfilling. They explained how the separation of waste can change a “waste” to a value added “resource” (Rousta et al., 2015).

There are different ways to separate different fractions in a waste stream. For example, sorting can be implemented in material recovery facilities (MRF) by sorting machines or sorting several waste fractions at home by inhabitants (Rousta and Dahlén, 2015). Among these, waste separation at the source is, in principal, an easy and efficient way for solid waste management systems (Tchobanoglous et al., 1993). Looking at the waste management across the globe indicated that the cities which applied waste sorting system at the source decreased the landfilling dramatically and increased the recycling rate (Xevgenos et al., 2015). Within source separation, there are different systems such as kerbside collection, bring system or

5 drop-off and property close collection system (Rousta and Dahlén, 2015). In all of these the people and their behaviour in source separation, alternatively known as recycling systems, the aim is to collect the recyclable materials from the waste stream to avoid behaviour, is vital to understand. inappropriate burning and landfilling. The effective participation of citizens in the system is required to achieve this goal. Rousta and Dahlén (2015) explained the type of source 1.3 Understanding Recycling Behaviour separation system comprehensively and the importance of the waste separation role in a waste management system. Source separation is a common method which was used in Sweden as Studying the recycling behaviour (the act of participation of people in a recycling program) well. The waste statistics in Sweden indicate that landfilling of municipal solid waste has has been conducted since a few decades ago when the concern about environmental issues decreased from 62% in 1975 to less than 1% in 2016 and recycling has increased from 6% to was raised. Investigating the factors influencing recycling behaviour was usually done by 35% (Fig. 3) during the same period (Swedish Waste Management Association, 2017). researchers in psychological, sociological and environmental science disciplines. It was interesting for researchers to find out why some people are recycler and others non-recycler. In one of the earliest study, Reid et al. (1976) examined the effect of prompting and proximity of containers in increasing the paper collection for recycling. In the literature review (Paper I), it was shown that different factors influence the recycling behaviour, of which convenient factors, such as easy access and to be close to sorting infrastructure, was the most important regardless the geographical locations and socio-demographic factors.

Rousta and Dahlén (2015) divided these factors into three broad categories: internal, external, and socio-demographic factors. Internal factors can influence attitude and intention

towards recycling. Examples of internal factors that have been discussed are pro-recycling Fig. 3. Comparison of primary treatment of household waste in Sweden in 1975 and 2016. (Tonglet et al., 2004), ascription of responsibility (Guagnano et al., 1995), attitudes towards Source separation plays a vital role in Swedish waste management. Source separation is recycling (Oskamp et al., 1991; Tucker and Speirs, 2003; Meneses and Palacio, 2005; Martin common in many other countries, including developing countries where they are setting up et al., 2006; Jesson, 2009) and environmental concern (Guerin et al., 2001; Meneses and goals to achieve 3Rs2 (Reduce, Reuse, and Recycle). In the cities where the wastes are not Palacio, 2005; Miafodzyeva and Brandt, 2013). separated then all wastes are typically transported to a landfill. That is why landfill is On the other hand, external factors are those that facilitate the knowledge and action converted to a place for scavengers to find materials to make a living. This causes a lot of towards recycling. Convenient infrastructure for recycling, access to recycling facilities, environmental and social problems in these cities. There are many studies about the issues of distance to the recycling stations, having enough space at home, (McCarty and Shrum, 1994; waste management in developing countries which mention the need for an efficient source Guagnano et al., 1995; Thøgersen, 1997; Tonglet et al., 2004; González-Torre and Adenso- separation (e.g. (Henry et al., 2006; Chen et al., 2010; Tai et al., 2011; Rajamanikam et al., Díaz, 2005; Martin et al., 2006; Dahlén et al., 2009; Jesson, 2009; Bernstad, 2014), 2014). Therefore, studying how to establish and improve the waste separation at source is knowledge about recycling (Tonglet et al., 2004), prompt and providing rewards and feedback always important in integrated solid waste management. For this reason, the participation of (Schultz et al., 1995) are identified as external factors in some studies.

Socio-demographic factors such as gender, age and income are also investigated in

2 3Rs (Reduce, Reuse, and Recycle) is common expression in many countries in waste management. Lagerkvist (2017) developed 7Rs to some studies. Ando and Gosselin (2005) and Beigl et al. (2008) stated that these factors play a explore the 3R for a sustainable waste management. He added four other Rs: 1) “Recover”: for example recover materials from old landfills; 2) “Remediate”: recreate the fresh environment; 3) “Retain”: safe final disposal; 4) “Release”: Spread / revert at a rate that nature tolerates major role in recycling behaviour, but how it affects this behaviour is still unclear. In a meta- (Lagerkvist, 2017).

7 6

the people and their behaviour in source separation, alternatively known as recycling behaviour, is vital to understand.

1.3 Understanding Recycling Behaviour

Studying the recycling behaviour (the act of participation of people in a recycling program) has been conducted since a few decades ago when the concern about environmental issues was raised. Investigating the factors influencing recycling behaviour was usually done by researchers in psychological, sociological and environmental science disciplines. It was interesting for researchers to find out why some people are recycler and others non-recycler. In one of the earliest study, Reid et al. (1976) examined the effect of prompting and proximity of containers in increasing the paper collection for recycling. In the literature review (Paper I), it was shown that different factors influence the recycling behaviour, of which convenient factors, such as easy access and to be close to sorting infrastructure, was the most important regardless the geographical locations and socio-demographic factors.

Rousta and Dahlén (2015) divided these factors into three broad categories: internal, external, and socio-demographic factors. Internal factors can influence attitude and intention towards recycling. Examples of internal factors that have been discussed are pro-recycling (Tonglet et al., 2004), ascription of responsibility (Guagnano et al., 1995), attitudes towards recycling (Oskamp et al., 1991; Tucker and Speirs, 2003; Meneses and Palacio, 2005; Martin et al., 2006; Jesson, 2009) and environmental concern (Guerin et al., 2001; Meneses and Palacio, 2005; Miafodzyeva and Brandt, 2013).

On the other hand, external factors are those that facilitate the knowledge and action towards recycling. Convenient infrastructure for recycling, access to recycling facilities, distance to the recycling stations, having enough space at home, (McCarty and Shrum, 1994; Guagnano et al., 1995; Thøgersen, 1997; Tonglet et al., 2004; González-Torre and Adenso- Díaz, 2005; Martin et al., 2006; Dahlén et al., 2009; Jesson, 2009; Bernstad, 2014), knowledge about recycling (Tonglet et al., 2004), prompt and providing rewards and feedback (Schultz et al., 1995) are identified as external factors in some studies.

Socio-demographic factors such as gender, age and income are also investigated in some studies. Ando and Gosselin (2005) and Beigl et al. (2008) stated that these factors play a major role in recycling behaviour, but how it affects this behaviour is still unclear. In a meta-

7 analysis of various studies, Miafodzyeva and Brandt (2013) identified that there are some is not performed. The brief explanation reveals the diversity of the models and research doubts on how the socio-demographic factors influence recycling behaviour. It is important to conducted in this field. Among all, there are three important models and frameworks note that most of the research in this field was based on case studies in a limited geographical developed specially for source separation. They are explained briefly below. and cultural area. Therefore, it is not simple to generalise the factors that influence recycling Motivation-Opportunity-Ability-Behaviour (MOAB) model: behaviour. However, the study on the factors mentioned above gives a broad understanding of The MOAB model was developed by Ölander and Thøgersen (1995) using case studies from recycling behaviour. It is worth mentioning that none of these factors can support recycling Denmark. This model (Fig. 4) explains that beliefs and evaluation of outcomes as well as behaviour alone. For example, one person is fully motivated and educated for recycling and social norms bring in the attitude towards recycling. It’s collectively called the Motivation there is no system to separate the waste, how can we expect better recycling of wastes? In box. The intention for recycling is the immediate antecedent for the behaviour in this model. another scenario, a gentleman with high income who lives in a city with enough Two other elements, ability and opportunity, were considered as moderators to convert infrastructures for sorting the waste does not sort his waste because he does not care. There intention in recycling into action. A return connection was also established between behaviour are thousands of scenarios with different factors that influence recycling behaviour which can and belief directly or via ability which is shown in Fig. 4 by dashed lines. According to this make a person a recycler or a non-recycler. For understanding more about the relationship model, when behaviour becomes a habit, the belief and evaluation of outcomes of this between these factors, the theoretical perspective of recycling behaviour is discussed in the behaviour can be changed towards stronger attitudes. next section.

1.4 Recycling Behaviour from a Theoretical Perspective

Is attitude (feeling about something) towards recycling an intentions factor to perform behaviour such as recycling? Do subjective norms and social pressures play a role? Theory of reasoned action (TRA) was developed by Ajzen and Fishbein (1980) who found positive answers to the above questions when they theorised the relation between attitude and behaviour. It was later found that in addition to attitude and subjective norms, how one perceives the difficulty of the behaviour can influence the intention to act (perceived behavioural control). This fact led to the theory of planned behaviour (TPB) (Ajzen, 1985). A large number of studies have examined these two theories, employed them directly or expanded the TPB to understand recycling behaviour (e.g. (Goldenhar and Connell, 1993; Park and Levine, 1999; Park, 2000; Kim et al., 2012; Largo-Wight et al., 2012). Tucker

(2001) described different models which have been improved based on TPB including Taylor Fig. 4. The MOAB model, modified from Ölander and Thøgersen (1995). and Todd (1995); model of Jones (1989); model of Goldenhar and Connell (1993) and model of Boldero (1995). He also explained some cause-effect models such as model of Altruistic Conceptual framework of environmental behaviour: Behaviour after Schwartz (1977); model of McCarty and Shrum (1994); model of Guagnano Barr (2002) developed a similar framework for environmental behaviour (Fig. 5) based on the et al. (1995) and model of Jackson et al. (1993). The A-B-C model by Guagnano et al. (1995) case studies in Exeter, England. Behavioural intention is influenced by environmental values, which says when A+C>0 i.e., if the sum of attitude (A) and external conditions (C) that psychological variables and situational variables. He counted psychological variables as support the behaviour were positive then the behaviour will be performed and when A+C<0 it 9 8

is not performed. The brief explanation reveals the diversity of the models and research conducted in this field. Among all, there are three important models and frameworks developed specially for source separation. They are explained briefly below.

Motivation-Opportunity-Ability-Behaviour (MOAB) model:

The MOAB model was developed by Ölander and Thøgersen (1995) using case studies from Denmark. This model (Fig. 4) explains that beliefs and evaluation of outcomes as well as social norms bring in the attitude towards recycling. It’s collectively called the Motivation box. The intention for recycling is the immediate antecedent for the behaviour in this model. Two other elements, ability and opportunity, were considered as moderators to convert intention in recycling into action. A return connection was also established between behaviour and belief directly or via ability which is shown in Fig. 4 by dashed lines. According to this model, when behaviour becomes a habit, the belief and evaluation of outcomes of this behaviour can be changed towards stronger attitudes.

Fig. 4. The MOAB model, modified from Ölander and Thøgersen (1995).

Conceptual framework of environmental behaviour:

Barr (2002) developed a similar framework for environmental behaviour (Fig. 5) based on the case studies in Exeter, England. Behavioural intention is influenced by environmental values, psychological variables and situational variables. He counted psychological variables as

9 motivators/barriers and the situational variables as enablers/disablers. These factors can combination of several factors and too complex to define in a rigid model. However, these directly influence the behaviour or behaviour intention. Categorising different factors in this three models strike consistently that intention to recycle is not sufficient for recycling framework differed from the MOAB model. For example, in the Barr (2002) model, the behaviour. In other words, if one is motivated to sort the waste, in the absence of an adequate logistics means recycling facilities, which was placed in psychological variables equivalent to system, participation would be missed. intrinsic factors and social norms whereas in the MOAB model, these factors were categorised as situational factors which is in opportunity element of the model.

Fig. 5. The conceptual framework of environmental behaviour, modified from Barr (2002).

Model of recycling participation: Fig. 6. The model of recycling participation, modified from Tucker (2001).

Another interesting model in recycling behaviour was developed by Tucker (2001) and it is By analysing these models and related literature of recycling behaviour, there is no doubt that known as a model of recycling participation. He developed a participation model (Fig. 6) understanding the numerous factors influencing recycling behaviour is vital to create and based on a comprehensive study for improving newspaper recycling in different cities in the develop recycling schemes. However, the most important thing is how to investigate and UK. He put the attitude, social norms and specific barriers antecedent to recycling intention. evaluate these factors in any context. Some studies aimed to use mixed methods (combination He also claimed that personal and system difficulties influence the presence of behaviour even of quantitative and qualitative methods) to evaluate the different interventions for recycling when the intention was available. He explained that if the sum of attitudes and norms schemes, but none of them aimed to introduce a common tool for evaluating a waste sorting becomes bigger than the barriers, then the household will participate in recycling. system. For example, Pieters (1989) used self-reported questionnaires, direct observation and

The models discussed above were developed based on case studies in various regions partial waste composition study methods. His study objectives were “acquiring insight in the with different conditions. In addition, researcher background and the aims of the research led motives of citizens to participate or not in the (recycling) program; insight that cannot be to the differences in the recycling behaviour models. Considering the theory and framework missed to correct the program, and to inform the public” (Pieters, 1989). In the same way, of recycling behaviour, the process of promoting recycling behaviour is dependent on a Tucker (2001) combined attitude-behaviour surveys, waste composition study, monitoring 11 10

combination of several factors and too complex to define in a rigid model. However, these three models strike consistently that intention to recycle is not sufficient for recycling behaviour. In other words, if one is motivated to sort the waste, in the absence of an adequate system, participation would be missed.

Fig. 6. The model of recycling participation, modified from Tucker (2001).

By analysing these models and related literature of recycling behaviour, there is no doubt that understanding the numerous factors influencing recycling behaviour is vital to create and develop recycling schemes. However, the most important thing is how to investigate and evaluate these factors in any context. Some studies aimed to use mixed methods (combination of quantitative and qualitative methods) to evaluate the different interventions for recycling schemes, but none of them aimed to introduce a common tool for evaluating a waste sorting system. For example, Pieters (1989) used self-reported questionnaires, direct observation and partial waste composition study methods. His study objectives were “acquiring insight in the motives of citizens to participate or not in the (recycling) program; insight that cannot be missed to correct the program, and to inform the public” (Pieters, 1989). In the same way, Tucker (2001) combined attitude-behaviour surveys, waste composition study, monitoring

11 and case study investigation for changes to develop a better understanding of recycling 2 Methods behaviour, particularly in newspaper recycling. The main purpose of his study was “to structure this understanding into a formal decision support tool, to help waste managers optimise newspaper recoveries to provide the maximum technical, social, environmental, and This section is divided into two main parts. The first part discusses the methodology and economic benefits”. The literature study (Paper I) revealed a lack of research on how to find justification of the selected methods whereas, the second part explains the materials and interventions for the recycling program and quantified the response to the interventions in a methods included in the geographical area of study (pilot area), and the design of the study waste management system. At the same time, there are a lot of recycling programs in the step by step in a time line. developed and developing countries which need to be evaluated and improved in order to increase material recovery. The need for a procedure for evaluating and improving the 2.1 Methodology and Justification of the Selected Methods recycling programs is essential for waste managers and policy makers. This means that the waste management systems need procedures to investigate and implement the interventions in Different methods have been used to study recycling behaviour, where the methods are either a recycling program and tools to analyse the effectiveness of the interventions. Hence, this qualitative or quantitative or a mix of both (Paper I). Recycling behaviour is a complex research aimed to create a procedure to investigate and assess the interventions in a recycling phenomenon and single methods cannot deliver reliable results for improving such systems. scheme which can be used in different waste management system in varying local Therefore, choice of methods as well as data collection techniques, are vital for making the circumstances. study successful. For a waste management system to function effectively, different actors involved in the system should work together to improve it. Inhabitants play a significant role in a waste management system, where they have multiple roles, being a consumer, performer 1.5 Objectives and customer (Rousta and Dahlén, 2015). Other players include the property owners and local The overall aim of this study was to contribute to sustainable use of material recycling for a authorities who are responsible for preparing infrastructure and providing services to sustainable society. This study mainly aimed to develop an analysis tool in order to create, inhabitants. Every participant within this waste management system has their own objectives. test and evaluate continuous improvement of any source separation system for household For example, the municipality aims to increase recycling rate within waste management waste. The main focus of this study was to answer the following research questions: system while the intention of a property owner is to ensure a good environment for the tenants. A research methodology and design should consider and involve these players and 1. How can the current sorting behaviour be evaluated? their goals. Moreover, it needs to be simple yet effective when considering their daily 2. How can appropriate interventions for improving this behaviour be identified? routines. That is why participatory research method, known as Action Research, was 3. How can the effect of interventions be assessed? implemented as the principal method in this study. According to Myers (2009), Action

Research methodology is used to solve a practical problem by expanding the scientific knowledge. Action Research analyses organisational change and study a process

simultaneously. Similarly, Stringer (2014) defined Action Research as “a systematic approach to investigation that enables people to find an effective solution to problems they confront in their everyday lives”. Many studies have used Action Research in educational research (e.g. (Kemmis, 1980; Katsenou et al., 2013)). In addition, Action Research has been used to solve

the community problems (Park, 1999) or community development (Stringer, 2014). Though

13 12

2 Methods

This section is divided into two main parts. The first part discusses the methodology and justification of the selected methods whereas, the second part explains the materials and methods included in the geographical area of study (pilot area), and the design of the study step by step in a time line.

2.1 Methodology and Justification of the Selected Methods

Different methods have been used to study recycling behaviour, where the methods are either qualitative or quantitative or a mix of both (Paper I). Recycling behaviour is a complex phenomenon and single methods cannot deliver reliable results for improving such systems. Therefore, choice of methods as well as data collection techniques, are vital for making the study successful. For a waste management system to function effectively, different actors involved in the system should work together to improve it. Inhabitants play a significant role in a waste management system, where they have multiple roles, being a consumer, performer and customer (Rousta and Dahlén, 2015). Other players include the property owners and local authorities who are responsible for preparing infrastructure and providing services to inhabitants. Every participant within this waste management system has their own objectives. For example, the municipality aims to increase recycling rate within waste management system while the intention of a property owner is to ensure a good environment for the tenants. A research methodology and design should consider and involve these players and their goals. Moreover, it needs to be simple yet effective when considering their daily routines. That is why participatory research method, known as Action Research, was implemented as the principal method in this study. According to Myers (2009), Action Research methodology is used to solve a practical problem by expanding the scientific knowledge. Action Research analyses organisational change and study a process simultaneously. Similarly, Stringer (2014) defined Action Research as “a systematic approach to investigation that enables people to find an effective solution to problems they confront in their everyday lives”. Many studies have used Action Research in educational research (e.g. (Kemmis, 1980; Katsenou et al., 2013)). In addition, Action Research has been used to solve the community problems (Park, 1999) or community development (Stringer, 2014). Though

Action Research is common in educational research, in environmental science and waste behaviours was identified, and socio-demographic factors were collected. This is referred as management it is uncommon and only a few studied have used this methodology (e.g.(Fahy the "Look" phase where data on the waste sorting behaviour was obtained: How good/bad is and Davies, 2007; Gutberlet, 2008)). Action Research has immense potential to solve the sorting behaviour in this area? Is there another perspective to show how important problems related to waste management. participation is in recycling schemes other than environmental issues? The gathered data

Many researchers described Action Research in multiple ways which are as follows: revealed that the problem (low sorting rate) was obvious which needs a thinking to identify Myers (2009) explained it in five steps: 1) diagnosing, 2) action planning 3) action taking, 4) solutions. The "Think" phase was intended to find an answer to the following questions: What evaluation and 5) specifying learning, whereas Stringer (2014) introduced Action Research as is happening here? Why is the rate of sorting so low? What do the inhabitants need to a systematic process for inquiry that can find appropriate solutions for a problem under local improve participation in the recycling schemes? The responses were critical in finding circumstances which includes three main routines: 1) Look, 2) Think, 3) Act (Fig. 7). answers to the questions that will be in the "Act" phase. Finally, the "Act" phase included the following questions: Is there anything that can be done to improve the sorting behaviour? How should the interventions be? How can the effectiveness of the interventions be assessed? The researcher’s role was to facilitate different actors/stakeholders involved in this process to find a solution.

To answer the various questions raised, different methods need to be used. The complexity of the recycling behaviour demands a combination of different methods. A mix-

method was recommended because the goal is to study a complex system (Tashakkori and Fig. 7. The Action Research interaction elements, modified from Stringer (2014). Teddlie, 2010). Both quantitative and qualitative methods were required for reliable data collection and analysis. The two data collection methods were pick analysis (waste This process is a continuous cyclic activity where "Look" phase is the step to define the composition study by manual sorting) and interview, which were found suitable to answer the situation and gather data; "Think" phase is to explore and interpret the phenomenon; "Act" research questions during different steps of Action Research. The study was designed in a phase is to identify the appropriate action, implement it and examine its effectiveness. It is way which was started with a pick analysis. The pick analysis was followed by an interview important to note that there is a back and forth relation between "Look", "Think" and "Act", and after applying the intervention for improving the waste sorting system, another pick illustrated by the dashed lines in Fig. 7. analysis was conducted. A complete design of the study is presented in Section 2.2.3. The The different phases of the Action Research, i.e., "Look, Think, Act", were adapted choice of data collection methods is discussed below. based on research questions in this study. The first step was to identify the stakeholders involved in the issue and their interest in the research. In this case, three actors were identified 2.1.1 Why pick analysis? (excluding research institution): 1) the waste management company (Goal: to reduce the Studying garbage is not new, and many researchers in consumer behaviour and waste amount of the waste and increase the collection of the recyclable materials), 2) the property management have studied it before. Rathje (1984) studied food consumption for more than a owner (Goal: to provide service for their clients) and 3) the inhabitants (Goal: to get rid of decade by studying the content of the waste in a specific locale in the US defined as Garbage waste in a convenient way and participate in the recycling schemes). Not all people Archaeology. Looking through garbage gives an insight of what people do, and possibility participated in the waste sorting, and different areas in the city had different waste sorting allows one to compare self-reported data to actual behaviour (Rathje, 1984). Similarly, Cote rate. Places with lowest sorting rates needed to be identified and intervention needed to be et al. (1985) used Garbology analysis to study the food consumption pattern. The waste implemented to improve the sorting rate. A pilot area including areas of lowest sorting

15 14

behaviours was identified, and socio-demographic factors were collected. This is referred as the "Look" phase where data on the waste sorting behaviour was obtained: How good/bad is the sorting behaviour in this area? Is there another perspective to show how important participation is in recycling schemes other than environmental issues? The gathered data revealed that the problem (low sorting rate) was obvious which needs a thinking to identify solutions. The "Think" phase was intended to find an answer to the following questions: What is happening here? Why is the rate of sorting so low? What do the inhabitants need to improve participation in the recycling schemes? The responses were critical in finding answers to the questions that will be in the "Act" phase. Finally, the "Act" phase included the following questions: Is there anything that can be done to improve the sorting behaviour? How should the interventions be? How can the effectiveness of the interventions be assessed? The researcher’s role was to facilitate different actors/stakeholders involved in this process to find a solution.

To answer the various questions raised, different methods need to be used. The complexity of the recycling behaviour demands a combination of different methods. A mix- method was recommended because the goal is to study a complex system (Tashakkori and Teddlie, 2010). Both quantitative and qualitative methods were required for reliable data collection and analysis. The two data collection methods were pick analysis (waste composition study by manual sorting) and interview, which were found suitable to answer the research questions during different steps of Action Research. The study was designed in a way which was started with a pick analysis. The pick analysis was followed by an interview and after applying the intervention for improving the waste sorting system, another pick analysis was conducted. A complete design of the study is presented in Section 2.2.3. The choice of data collection methods is discussed below.

2.1.1 Why pick analysis?

Studying garbage is not new, and many researchers in consumer behaviour and waste management have studied it before. Rathje (1984) studied food consumption for more than a decade by studying the content of the waste in a specific locale in the US defined as Garbage Archaeology. Looking through garbage gives an insight of what people do, and possibility allows one to compare self-reported data to actual behaviour (Rathje, 1984). Similarly, Cote et al. (1985) used Garbology analysis to study the food consumption pattern. The waste

15 contents in garbage bins reveals how people consume and how they sort their materials in 2.2.1 Pilot area relation to recycling schemes. This means that by looking in the garbage bins and identifying The pilot area is a residential apartment building which has 208 flats. These flats are located the fractions indicates which material was sorted right or wrong, based on the recycling in Borås, a city in the south-west of Sweden. This area was chosen for various reasons, schemes. This method measures the actual waste sorting behaviour of the inhabitants, including the diversity of socio-demographic background of inhabitants (more than two-third irrespective of the answers obtained in self-report surveys. A study about waste inhabitants born outside Sweden, three out of five had income <15,600 Euro/year and one- attitude/behaviour in Scotland and North-West England, indicates an inconsistency between third did not have a car (Swedish Statistisc, 2012)). Secondly, a prior pick analysis conducted self-report and observations where 15% of self-reports made false claims (Tucker and Speirs, in Borås showed that this area had the highest miss-sorted fraction in their waste compared 2003). Similarly, there are other studies that indicated the difference between what people say with other parts of the city. Also, the stakeholders, including property owners and waste in self-reports and what they actually do (Barker et al., 1994; Corral-Verdugo, 1997). handling companies, were interested in improving the recycling behaviour in this area. Therefore, measuring inhabitants’ behaviour in sorting the waste by pick analysis was Finally, from technical point of view, sampling of the waste for pick analysis and controlling identified as a suitable method for the purpose of this study. waste streams in this area was possible.

2.1.2 Why interview? At the time, the city of Borås had a source separation system for recyclable waste which was based on a bring-system. In a bring-system, packaging and recyclable materials such as A part of this study tries to understand the needs of inhabitants to improve the recycling newsprints, papers, plastics, metals and glass should be brought to the closest recycling behaviour. Interview gives deep insights into how inhabitants understand and perceive the station (drop-off point) and sorted into the proper containers. Special wastes including waste sorting system from their point of view. For this reason, qualitative methods such as hazardous waste, bulky waste and garden waste should be taken to one of five recycling interviews and focus groups were recommended as a data collection methods (Kvale and centres located in different parts of the city. For the remaining household waste, bags in two Brinkmann, 2009; Myers, 2009). Briefly, this method involves talking to people, asking them different colours were used to collect, 1) food waste in black bags, 2) other residuals and/or what happened, why it happened, how and when it happened (Myers, 2009). This could be combustible waste in white bags. Fig. 8 shows the waste collection system in this city (Rousta considered as an inter-view meaning that the interaction between two views may contribute to et al., 2015). understand a phenomenon (Kvale and Brinkmann, 2009). Qualitative methods mostly involve The pilot area was part of the waste management system described above. This area had interaction between researchers and the target group. One may talk to people (interview), containers to collect newsprints and glass packaging at the back of the resident building. observe them (field study), or even practise and facilitate a discussion in a group (focus However, the inhabitants had to transport the other materials about two kilometres to reach group). This interview helps in interpreting how people understand the waste sorting system the nearest recycling stations and centres. and what they perceive is needed to improve their participation in the system. To interpret the data from the pilot area, a reference area was chosen in the same city. No interventions were made in the reference area during the study. The reference area 2.2 Materials and Methods included 299 apartments that had two recycling stations for collecting recyclables with a This section focusses on describing how the study was conducted, including the pilot area, distance about 400 m from the buildings. The results from the reference area were used only how to use pick analysis, interview data collection and finally the design of the study. to interpret the changes in the pilot area.

17 16

2.2.1 Pilot area

The pilot area is a residential apartment building which has 208 flats. These flats are located in Borås, a city in the south-west of Sweden. This area was chosen for various reasons, including the diversity of socio-demographic background of inhabitants (more than two-third inhabitants born outside Sweden, three out of five had income <15,600 Euro/year and one- third did not have a car (Swedish Statistisc, 2012)). Secondly, a prior pick analysis conducted in Borås showed that this area had the highest miss-sorted fraction in their waste compared with other parts of the city. Also, the stakeholders, including property owners and waste handling companies, were interested in improving the recycling behaviour in this area. Finally, from technical point of view, sampling of the waste for pick analysis and controlling waste streams in this area was possible.

At the time, the city of Borås had a source separation system for recyclable waste which was based on a bring-system. In a bring-system, packaging and recyclable materials such as newsprints, papers, plastics, metals and glass should be brought to the closest recycling station (drop-off point) and sorted into the proper containers. Special wastes including hazardous waste, bulky waste and garden waste should be taken to one of five recycling centres located in different parts of the city. For the remaining household waste, bags in two different colours were used to collect, 1) food waste in black bags, 2) other residuals and/or combustible waste in white bags. Fig. 8 shows the waste collection system in this city (Rousta et al., 2015).

The pilot area was part of the waste management system described above. This area had containers to collect newsprints and glass packaging at the back of the resident building. However, the inhabitants had to transport the other materials about two kilometres to reach the nearest recycling stations and centres.

To interpret the data from the pilot area, a reference area was chosen in the same city. No interventions were made in the reference area during the study. The reference area included 299 apartments that had two recycling stations for collecting recyclables with a distance about 400 m from the buildings. The results from the reference area were used only to interpret the changes in the pilot area.

sampling size and the analysis have been described in Papers III and IV. In total, three pick analyses have been conducted in the pilot and the reference areas. The samples were collected in a 660-liter wheeled bin. The sampling scheme in the pilot area is shown in Table 1, and is reprinted from Paper IV.

Table 1. Details of sampling for pick analysis in the pilot area.

Date Unit of sample No. of samples per week Total samples Total weight (kg) Nov. 2011 Waste in a 660 litre bin 7 28 1732.9 Nov. 2013 Waste in a 660 litre bin 7 28 1699.0 Nov. 2015 Waste in a 660 litre bin 7 28 1592.5

2.2.2.2 Interview

A semi-structured interview and a structured short-interview were designed, and Paper IV

describes them in detail. For semi-structured interviews a theme was fixed while questions Fig. 8. Waste collection system in the city of Borås modified from Rousta et al. (2015). were not (Kvale and Brinkmann, 2009). Based on the responses from the participants, probing questions were asked to learn the interviewees perceptions during the interview (Myers, 2.2.2 Data collection methods 2009). The theme of the interview was to understand inhabitants’ needs to participate in the waste sorting system. The main challenge was to find the interviewees, and eight interviews This section summarizes the data collection methods that have been used in this study. were performed. The interviewees were selected using convenience sampling by writing a Comprehensive information about the data collection methods can be obtained from letter to the target group (inhabitants in the pilot area) or asking a contact person in the pilot Paper IV. area to find them. All interviews were performed based on the principles described by Kvale 2.2.2.1 Pick analysis and Brinkmann (2009) and Myers (2009) at the University of Borås. Each interview lasted Dahlén and Lagerkvist (2008) reviewed different methods of waste composition study across between 50 and 65 minutes and was recorded. The recordings were transcribed, analysed and the globe and developed a manual on how to conduct a pick analysis. It includes planning, interpreted to find the needs of inhabitants to participate in the waste sorting system. The sampling (sample size, number of samples and location), manual sorting and evaluating the results from these interviews were used to perform structured and short-interviews with a data by considering the aim, type and waste components in the study. The Swedish Waste bigger population in the pilot area (Paper IV). The short interviews have been carried out in Management Association instructed all municipalities to use this manual for pick analyses the front of the apartment buildings in the pilot area. (Swedish Waste Management Association, 2005; Swedish Waste Management Association, 2013). According to this manual, statistics on pick analysis should be based on the aim of the 2.2.3 Design of study study and the quantity of the data for each waste component. Some suggestions were made by The study was designed for a period of four years and was based on Action Research. Fig. 9 Lagerkvist et al. (2011) on how to implement and analyse the data from pick analyses. Based shows the time line where different activities were carried out in this study. on the instructions from the Swedish Waste Management Association, sampling and quantifying different waste fractions in a pick analysis was carried out. Then, the data were analysed using statistical software (Minitab) by comparing two groups of data in a t-test. The

19 18

Table 1. Details of sampling for pick analysis in the pilot area.

2.2.2.2 Interview

A semi-structured interview and a structured short-interview were designed, and Paper IV describes them in detail. For semi-structured interviews a theme was fixed while questions were not (Kvale and Brinkmann, 2009). Based on the responses from the participants, probing questions were asked to learn the interviewees perceptions during the interview (Myers, 2009). The theme of the interview was to understand inhabitants’ needs to participate in the waste sorting system. The main challenge was to find the interviewees, and eight interviews were performed. The interviewees were selected using convenience sampling by writing a letter to the target group (inhabitants in the pilot area) or asking a contact person in the pilot area to find them. All interviews were performed based on the principles described by Kvale and Brinkmann (2009) and Myers (2009) at the University of Borås. Each interview lasted between 50 and 65 minutes and was recorded. The recordings were transcribed, analysed and interpreted to find the needs of inhabitants to participate in the waste sorting system. The results from these interviews were used to perform structured and short-interviews with a bigger population in the pilot area (Paper IV). The short interviews have been carried out in the front of the apartment buildings in the pilot area.

2.2.3 Design of study

The study was designed for a period of four years and was based on Action Research. Fig. 9 shows the time line where different activities were carried out in this study.

3 Results and Discussion

This section synthesises the results presented in Papers I-IV answers the research questions and discusses the Recycling Behaviour Transition procedure.

3.1 Development of a Tool to Improve Source Separation

Recycling Behaviour Transition (RBT), shown in Fig. 11 and explained in detail in Paper IV. Fig. 9. Design of the study showing the pilot and reference area with the differences between them. RBT can be used in any source separation system for continuous improvement. RBT was The pilot area and reference area are shown above and below of the time line in Fig. 9 created based on the trials conducted in the pilot area mentioned in Section 2.2.3. respectively. The first pick analysis was conducted in both areas and acts as the reference point for comparison. Then, the interviews were conducted to find appropriate interventions in the pilot area. The interventions were implemented in two-stages in the pilot area. After implementations, a new pick analysis was conducted to assess the effect of these interventions. No interventions were made in the reference area during this study. Fig. 10 shows the tasks conducted in the pilot area during the four years (Papers III and IV).

Fig. 11. The Recycling Behaviour Transition (RBT) procedure (Paper IV).

RBT was developed through a combination of qualitative and quantitative methods that have been employed within the Action Research framework. Pick analyses and interviews were used as data collection techniques in RBT that was thoroughly examined in the pilot area. The mix of methods used in RBT helped in answering different research questions. According to Fig. 10. The time line of different activities performed in this study. Interventions were: a) placing the previous literature (Paper I), most studies in waste management are case specific, i.e. stickers on black bag holders to sort food waste; b) building a collection system close to the pilot area; conducted in a limited geographical location with specific social and technical background. c) distributing different types of information about waste sorting; d) using informal conversation to Generalising these results is an impossible task due to the variation in local circumstances and promote recycling.

21 20

3 Results and Discussion

This section synthesises the results presented in Papers I-IV answers the research questions and discusses the Recycling Behaviour Transition procedure.

3.1 Development of a Tool to Improve Source Separation

The main objective of this research was to develop an analysis tool that will test and evaluate any source separation system for household waste. The methods described above are employed within the Action Research methodology resulted in a procedure known as Recycling Behaviour Transition (RBT), shown in Fig. 11 and explained in detail in Paper IV. RBT can be used in any source separation system for continuous improvement. RBT was created based on the trials conducted in the pilot area mentioned in Section 2.2.3.

Fig. 11. The Recycling Behaviour Transition (RBT) procedure (Paper IV).

21 several interdisciplinary factors involved. However, the RBT procedure can be used as a 3.3 Identifying the Appropriate Interventions for Improving common and general tool that can be applied in any waste source separation system, Recycling Behaviour regardless of the technical and social background. This is because the procedure itself To analyse the results from the first pick analysis and to answer several research questions, explores and evaluates the local situation. This procedure can help waste managers and including why the inhabitants did not participate in recycling scheme and what they need for stakeholders involved in the waste management system to investigate and apply the effective participation, semi-structured interviews and structured short interview were conducted. The interventions that include the cultural and socio-demographic factors related to the specific results of these interviews can be found in Paper IV. The basic finding revealed that the municipality/region. RBT may be used as a standard tool and would offer a useful perspective motivation for recycling was a varying parameter among the inhabitants in the pilot area. The for comparing the factors and interventions involved in the specific waste sorting system. inhabitants could be classified into three categories based on their recycling attitudes: 1) Those who care about recycling do it properly, 2) Those who do not care never recycle 3.2 Evaluation of the Current Sorting Behaviour regardless of the system, and 3) Those who stand between the other two groups. The main focus of the interviews was to find the interventions to change the behaviour of the third. The first pick analysis was conducted in the pilot area to evaluate the miss-sorted fraction in Some people knew how to sort the waste fractions, but the distance to the recycling station white bags (intended for combustible waste and excludes packaging) and black bags (intended was an obstacle. On the other hand, some got wrong information, and they believed that they for food waste). were sorting correctly. The wrong information was communicated via informal channels The results of this pick analysis showed that the average proportion of miss-sorted including friends and family members, or they perceived the right information in a wrong way 3 fractions in both white and black bags was 55% (Paper II). The miss-sorted fraction in white because of lack of knowledge in language and/or complexity in the type of information. This bags was about 70%, while in black bags it was about 30%. The most miss-sorted materials in indicates that understanding of the system is based on how a person perceives and interprets the white bags were packaging and newsprints. These miss-sorted materials form the it. The necessity for appropriate information and its dissemination were obvious among community costs 13 Million SEK for Borås, which has only 105,000 inhabitants. interviewees. This indicates the importance of the diversity of information, both in type, time Approximately 28% of this cost has been paid to the FTI (Förpackning & Tidning of distribution and communication channels. Insamlingen), that is responsible for the collection of packaging and newspaper in Sweden. The analysis from the semi-structured interviews showed that a) lack of knowledge and By the producer responsibility law (SFS1994a; SFS1994b) they are obliged to collect it, but information about sorting, and b) distance to the recycling station were the main reasons for they cannot be done due to the miss-sorting. The analysis in Paper I helps policy makers and miss-sorting of the waste. The structured short interviews confirmed these findings. These other actors in waste management re-think the responsibility of collection of the packaging two reasons formed the base of the interventions that were identified and employed in materials. The results from first pick analysis showed that this method is suitable to evaluate different ways in the pilot area during the course of this study (see Section 3.4). the recycling behaviour. The rich data collected through pick analysis create opportunities for analysing the waste sorting system from different perspectives. The next interesting finding from the interviews was the impact of the first impression on the inhabitants’ behaviour. Some interviewees claimed that their first impression of the sorting system had a deep impact and it was hard to change their perception after a while. This can be interpreted as task knowledge in the MOAB (Motivation-Opportunity-Ability- Behaviour) theory which was discussed in Section 1.4. The findings from the interview data also corresponded to the different theories mentioned in Section 1.4. It was mentioned earlier that attitude is not the only reason for the intention in recycling behaviour. Additional factors 3 All percentages in this text are percentage by weight, unless stated otherwise.

23 22

3.3 Identifying the Appropriate Interventions for Improving Recycling Behaviour

To analyse the results from the first pick analysis and to answer several research questions, including why the inhabitants did not participate in recycling scheme and what they need for participation, semi-structured interviews and structured short interview were conducted. The results of these interviews can be found in Paper IV. The basic finding revealed that the motivation for recycling was a varying parameter among the inhabitants in the pilot area. The inhabitants could be classified into three categories based on their recycling attitudes: 1) Those who care about recycling do it properly, 2) Those who do not care never recycle regardless of the system, and 3) Those who stand between the other two groups. The main focus of the interviews was to find the interventions to change the behaviour of the third. Some people knew how to sort the waste fractions, but the distance to the recycling station was an obstacle. On the other hand, some got wrong information, and they believed that they were sorting correctly. The wrong information was communicated via informal channels including friends and family members, or they perceived the right information in a wrong way because of lack of knowledge in language and/or complexity in the type of information. This indicates that understanding of the system is based on how a person perceives and interprets it. The necessity for appropriate information and its dissemination were obvious among interviewees. This indicates the importance of the diversity of information, both in type, time of distribution and communication channels.

The analysis from the semi-structured interviews showed that a) lack of knowledge and information about sorting, and b) distance to the recycling station were the main reasons for miss-sorting of the waste. The structured short interviews confirmed these findings. These two reasons formed the base of the interventions that were identified and employed in different ways in the pilot area during the course of this study (see Section 3.4).

The next interesting finding from the interviews was the impact of the first impression on the inhabitants’ behaviour. Some interviewees claimed that their first impression of the sorting system had a deep impact and it was hard to change their perception after a while. This can be interpreted as task knowledge in the MOAB (Motivation-Opportunity-Ability- Behaviour) theory which was discussed in Section 1.4. The findings from the interview data also corresponded to the different theories mentioned in Section 1.4. It was mentioned earlier that attitude is not the only reason for the intention in recycling behaviour. Additional factors

23 including personal situations, convenience, task knowledge, opportunities and abilities 3.4 Assessing the Effect of the Interventions influence both behaviour and attitude. For example, placing the recycling container closer to After applying the interventions in the pilot area, the second and third pick analyses were the vicinity was a psychological variable in Barr (2002), personal difficulties in Tucker (2001) conducted to evaluate the miss-sorted fraction in the waste stream. In Paper III, a comparison and opportunity in Ölander and Thøgersen (1995). of the pick analyses has been done before and after interventions. The results showed that The important findings from the interviews are summarised below: both interventions (decrease in the distance to the collection point and information about how

 Finding 1: There are various motivations for recycling behaviour including to sort food waste) were effective. The interventions decreased the miss-sorted fraction of the contribution to the next generation, conservation of the environment, to keep the household waste from 55% to 39%. Simultaneously, data reliability and evaluation in the surrounding clean, pressure from society, and to respect regulations. changes of the waste fraction before and after interventions, and waste fractions of the  Finding 2: Convenience is a crucial factor for increasing participation in the recycling reference area have also been analysed. These investigations helped the interpretation of the scheme. results. A detailed analysis on the effects of the interventions was given in Paper III that  Finding 3: Convenience is relative, where different people perceive it differently. It is showed that the effects of the interventions were significant when compared with nuisance difficult to have a convenient waste sorting system that satisfies all people. At the factors. The important findings from the two pick analyses before and after the interventions same time, doing the task of sorting can change the attitude of how the system is (Fig. 10 a and b) were: convenient.  Finding 1: A closer distance to the recycling station improves the recycling rate. By  Finding 4: For increased participation in recycling behaviour, the design of placing recycling station closer to the premises, miss-sorted fractions decreased by information, types of the communication channels, and the timing of dissemination are 28% in the pilot area. vital.  Finding 2: Increasing the number of recyclable fractions that are collected at the drop  Finding 5: First impressions of the system had a deep influence to create inhabitants off points increased the participation in waste sorting. habits within recycling behaviour.  Finding 3: The easy access to correct information improves sorting of the waste  Finding 6: Waste sorting can change the beliefs of inhabitants about environmental fractions. By placing stickers on the black bags holders, the amount of miss-sorted issues. fraction, in this case diapers, decreased by 70% in the black bags in the pilot area.  Finding 7: Interviews are relevant methods to understand the needs of inhabitants for  Finding 4: Though the effects of interventions are quantifiable through pick analyses, an effective recycling behaviour. it is impossible to isolate the effect of each intervention from the others.

The semi-structured interviews provided an insight into the need of inhabitants to participate  Finding 5: The reference area was crucial in interpreting the effect of the interventions in source separation. However, there were some challenges in conducting interviews and finding out other reasons for differences. including finding the right interviewee who can explain their thoughts, designing the theme of  Finding 6: The longer the study, the more factors needs to be considered that could be interview to gather reliable data, and subjective analysis of data. For the interventions to be influential. accepted by the community, analysing the data in the interview is important. The pilot study The third and last pick analysis was conducted after one year of applying other interventions. yielded reliable data because it was based on local circumstances. Similarly for RBT, Comparing the miss-sorted waste fractions in the beginning of the study and after four-years qualitative data must be selected based on the local circumstances. The researcher/interviewer showed that miss-sorted fraction was significantly lower (Fig. 12). The two interventions at role in data collection and further analysis were crucial for this type of study. the beginning of this study (Fig. 10 a and b) had more effects than the interventions applied during the last year (Fig. 10 c and d). Decreasing the distance between the recycling

25 24

3.4 Assessing the Effect of the Interventions

After applying the interventions in the pilot area, the second and third pick analyses were conducted to evaluate the miss-sorted fraction in the waste stream. In Paper III, a comparison of the pick analyses has been done before and after interventions. The results showed that both interventions (decrease in the distance to the collection point and information about how to sort food waste) were effective. The interventions decreased the miss-sorted fraction of the household waste from 55% to 39%. Simultaneously, data reliability and evaluation in the changes of the waste fraction before and after interventions, and waste fractions of the reference area have also been analysed. These investigations helped the interpretation of the results. A detailed analysis on the effects of the interventions was given in Paper III that showed that the effects of the interventions were significant when compared with nuisance factors. The important findings from the two pick analyses before and after the interventions (Fig. 10 a and b) were:

 Finding 1: A closer distance to the recycling station improves the recycling rate. By placing recycling station closer to the premises, miss-sorted fractions decreased by 28% in the pilot area.  Finding 2: Increasing the number of recyclable fractions that are collected at the drop off points increased the participation in waste sorting.  Finding 3: The easy access to correct information improves sorting of the waste fractions. By placing stickers on the black bags holders, the amount of miss-sorted fraction, in this case diapers, decreased by 70% in the black bags in the pilot area.  Finding 4: Though the effects of interventions are quantifiable through pick analyses, it is impossible to isolate the effect of each intervention from the others.  Finding 5: The reference area was crucial in interpreting the effect of the interventions and finding out other reasons for differences.  Finding 6: The longer the study, the more factors needs to be considered that could be influential.

The third and last pick analysis was conducted after one year of applying other interventions. Comparing the miss-sorted waste fractions in the beginning of the study and after four-years showed that miss-sorted fraction was significantly lower (Fig. 12). The two interventions at the beginning of this study (Fig. 10 a and b) had more effects than the interventions applied during the last year (Fig. 10 c and d). Decreasing the distance between the recycling

25 containers and the pilot area had a larger effect than informal information and feedbacks. The black bags increased. No changes in the reference area were expected as no interventions informal information didn’t decrease the miss-sorted waste significantly (during 2013 to were applied. Simultaneously, the proper interventions in the pilot area increased the 2015). No clear conclusion could be obtained on what would have happened if no information collection of recyclable. In addition, the results confirmed that the RBT procedure that was had been provided. Further work is necessary to find the interrelation between the information developed in this study is an effective tool. Re-calculation of the community cost described in and recycling behaviour patterns. Paper II reveals that the interventions decreased the community cost in the pilot area by 40%.

Pick analysis was found to be a reliable method for comparing the results before and wt% 80 after employing interventions to assess the effect of interventions, and evaluates the actual 70 behaviour of waste sorting. Nonetheless, there are challenges in execution and analysing the 60 data. It is important to have a large sample size, to avoid nuisance factors and have a 50 White bags statistically meaningful data. Also, the procedure of pick analysis should be objectively based 40 Black bags 30 on a clear instruction. To cut the costs of conducting pick analysis, it is good to analyse the Total Miss-sorted ratio Miss-sorted 20 data from pick analysis for other purposes than just improving the recycling behaviour at the 10 same time. For example, if a system distributes free plastic bags to collect the waste, it is 0 possible by pick analysis to evaluate the number of bags used per household that helps in 2011 2013 2015

further planning. The result of pick analyses can be used as an annual waste statistic for the Fig. 12. The proportion of the miss-sorted fractions in the white bags (intended for combustible waste) city. To combine different purposes by a pick analysis is a good idea that reduces the cost. For and black bags (intended for food waste) in the pilot area during four years of study. research purposes, having a reference area is recommended to interpret the results properly. To interpret the changes from pick analysis results and applied interventions, a reference area was examined during the same period of time (Fig. 13). 3.5 A Common Thread through Papers I-IV

wt% This thesis combines four publications with data collected at different stages. A common 80 thread was observed and maintained throughout the study to achieve the objectives. 70 Reviewing the literature (Paper I) was a continuous task during this PhD period. Although it 60 was published at the end, the task started at the beginning of the study. Different theories, 50 White bags methods and results from other studies were identified that provided a broad knowledge for 40 Black bags the researcher to design this study, select methods and to analyse the data. The literature 30 Total Miss-sorted Miss-sorted ratio 20 review (Paper I) revealed that multidisciplinary research with complementary methods is 10 required to study recycling behaviour, since it is a complex phenomenon. The research should 0 focus on how to improve the waste source separation system, rather than only to identify the 2010 2013 2015 factors influencing the system. Furthermore, it was revealed that generalising result from one Fig. 13. The same as Fig. 12 but for the reference area. study to another is extremely difficult because each study has been conducted in the specific local situation. It was also found that there was no analysis tool to identify the intervention in No particular trend was observed in the reference area, where no intervention was applied a source separation system and to assess its effectiveness. This led to the choice of Action (Fig. 13). Miss-sorting in the white bags remained at the same level while miss-sorting in 27 26

black bags increased. No changes in the reference area were expected as no interventions were applied. Simultaneously, the proper interventions in the pilot area increased the collection of recyclable. In addition, the results confirmed that the RBT procedure that was developed in this study is an effective tool. Re-calculation of the community cost described in Paper II reveals that the interventions decreased the community cost in the pilot area by 40%.

Pick analysis was found to be a reliable method for comparing the results before and after employing interventions to assess the effect of interventions, and evaluates the actual behaviour of waste sorting. Nonetheless, there are challenges in execution and analysing the data. It is important to have a large sample size, to avoid nuisance factors and have a statistically meaningful data. Also, the procedure of pick analysis should be objectively based on a clear instruction. To cut the costs of conducting pick analysis, it is good to analyse the data from pick analysis for other purposes than just improving the recycling behaviour at the same time. For example, if a system distributes free plastic bags to collect the waste, it is possible by pick analysis to evaluate the number of bags used per household that helps in further planning. The result of pick analyses can be used as an annual waste statistic for the city. To combine different purposes by a pick analysis is a good idea that reduces the cost. For research purposes, having a reference area is recommended to interpret the results properly.

3.5 A Common Thread through Papers I-IV

This thesis combines four publications with data collected at different stages. A common thread was observed and maintained throughout the study to achieve the objectives. Reviewing the literature (Paper I) was a continuous task during this PhD period. Although it was published at the end, the task started at the beginning of the study. Different theories, methods and results from other studies were identified that provided a broad knowledge for the researcher to design this study, select methods and to analyse the data. The literature review (Paper I) revealed that multidisciplinary research with complementary methods is required to study recycling behaviour, since it is a complex phenomenon. The research should focus on how to improve the waste source separation system, rather than only to identify the factors influencing the system. Furthermore, it was revealed that generalising result from one study to another is extremely difficult because each study has been conducted in the specific local situation. It was also found that there was no analysis tool to identify the intervention in a source separation system and to assess its effectiveness. This led to the choice of Action

Research, with a combination of a qualitative and quantitative method to conduct this study. Table 2. Primary aim of different paper published from this thesis. After identifying the pilot area and the actors involved, the first pick analysis was conducted. Paper Aim of the publication Mainly related to Paper II presents the results of the first pick analysis that assesses the importance of source separation in the Swedish waste management system from the national economic perspective. To identify the lack of knowledge in recycling Supporting the study from scientific behaviour field. and theoretical perspectives. The results indicate that the miss-sorting behaviour yields a considerable community cost. I To understand the theories and framework in research

on recycling behaviour.

The main purpose of Paper II was: 1) To use the data from pick analysis in another way than To choose the appropriate methodology. just calculating the percentages and characterization of the waste fractions; 2) To quantify the To show the importance of effective source separation How can the current sorting miss-sorting in a source separation system; 3) To reveal the importance of effective waste in Swedish waste management systems. behaviour be evaluated? To show another perspective of data analysis from pick sorting system to the stakeholders (waste managers, politicians, and consumers). The results II analysis and the importance of this method to measure of the Paper II provide a strong motivation for various actors involved in the research to recycling behaviour. continue the study that was vital for long-term participatory research. Moreover, it revealed To assess the effect of the intervention using statistics How can the effect of interventions that pick analysis is a proper tool to measure the participation in the recycling schemes. of pick analysis. be assessed? Assessment of the interventions in an accurate way was the aim of the work presented in III To identify the pick analysis as proper tool for quantifying the effect of the interventions. Paper III. The lack of knowledge about how to quantify the effect of the interventions that

Creating a tool to improve recycling was apparent from the literature was explained in this paper. By doing the second pick To use a qualitative method, interview, for investigation behaviour. of interventions. analysis and a systematic statistical analysis of the data, it was possible to quantify the effect IV How can appropriate interventions to To describe the RBT procedure, which is the main improve this behaviour be identified? result of this study. of interventions. In addition, work presented in Paper III revealed the barriers and factors that influence data analysis.

To summarize, Papers II and III showed that analysing a rich set of data generated 3.6 Sources of Error from pick analysis is suitable for answering the research questions. The aim of Paper IV was The design of the study, including the implementation of the data collection methods, was to introduce the RBT procedure as the main result of the entire study. This paper covers all done in a way to ensure valid answers to the research questions. In this way, using aspects, including how to find the intervention with qualitative methods and how to place the triangulation (to use two or more methods for data collection to check the results of the same series of the data collections methods in an Action Research framework to generate the subject), such as comparing the results and observation of the fractions in pick analysis, and analysis tool for improving the source separation. The common thread through publication the findings in the interviews, helped the researcher to ensure both validity and reliability of was obvious from start to end and this could help researchers and various actors involved in to the research findings. Several tasks were conducted to minimise errors and collect reliable use the results in their activities. A summary of the aims of the publication is shown in data. These are: Table 2. The third column in this table lists the research question that this work tries to  The sample size for all three pick analyses was almost the same, and it was selected answer. There was some overlap in the publications to answer different research questions, randomly from the pilot area as shown in Table 1. but only mentioned the main research question that the paper aims to answer is shown in the  The researcher was involved personally in data collection and analysis. column.  The same scale, with the same calibration, same team, and same instructions was used

in all three pick analyses to reduce the human errors in the measurements.

29 28

Table 2. Primary aim of different paper published from this thesis.

Paper Aim of the publication Mainly related to

To identify the lack of knowledge in recycling Supporting the study from scientific behaviour field. and theoretical perspectives. I To understand the theories and framework in research

on recycling behaviour.

To choose the appropriate methodology.

To show the importance of effective source separation How can the current sorting in Swedish waste management systems. behaviour be evaluated? To show another perspective of data analysis from pick II analysis and the importance of this method to measure recycling behaviour.

To assess the effect of the intervention using statistics How can the effect of interventions of pick analysis. be assessed? III To identify the pick analysis as proper tool for quantifying the effect of the interventions.

Creating a tool to improve recycling To use a qualitative method, interview, for investigation behaviour. of interventions. IV How can appropriate interventions to To describe the RBT procedure, which is the main improve this behaviour be identified? result of this study.

3.6 Sources of Error

The design of the study, including the implementation of the data collection methods, was done in a way to ensure valid answers to the research questions. In this way, using triangulation (to use two or more methods for data collection to check the results of the same subject), such as comparing the results and observation of the fractions in pick analysis, and the findings in the interviews, helped the researcher to ensure both validity and reliability of the research findings. Several tasks were conducted to minimise errors and collect reliable data. These are:

 The sample size for all three pick analyses was almost the same, and it was selected randomly from the pilot area as shown in Table 1.  The researcher was involved personally in data collection and analysis.  The same scale, with the same calibration, same team, and same instructions was used in all three pick analyses to reduce the human errors in the measurements.

 Statistical analysis was carried out using Minitab statistical software and normality 4 Conclusions test. The two sample t-test and 95% confidence interval (P<0.05) were applied to determine if observed changes were statistically significant (Paper III).  Parallel to the pilot area, a reference area was analysed to see if there are any changes The recycling behaviour transition (RBT) procedure (Fig. 11) evaluated the current recycling in recycling behaviour based on interventions or any changes in consumption patterns behaviour, identified appropriate interventions and assessed the effect of these interventions in general. in the pilot area. The RBT procedure proved to be an efficient tool that improved recycling  The semi-structured interviews were recorded and transcribed in order to ensure significantly in the pilot area. Conclusions are listed as follows:

correct analysis of the original material.  The RBT procedure has a promising potential to be a standard tool for improving any In long-term studies there are nuisance factors that should be considered as possible sources waste sorting system, irrespective of the social and technical factors involved. of error. It was not possible to control these sources. However, the effect of them on the  A participatory research framework like Action Research has the potential to solve results and conclusions of the study were not considerable when the results were compared complex problems in a waste management system. and interpreted with the data from the reference area.  Using mixed methods was vital for the complex studies such as recycling behaviour where both technical and social aspects are involved. A combination of qualitative and  It is impossible to isolate the impact of different interventions on each other and assess quantitative methods reinforced the reliability of the data collected. the effectiveness of each intervention alone.  Pick analysis was a relevant method to measure the waste sorting behaviour that  Interview by itself is an intervention when it creates an extra awareness in the pilot quantified the effect of interventions in the system. area. This impact is hard to be evaluated.  Semi-structured interviews are found as a relevant method to identify interventions for  Moving tenants from/to the pilot area created uncertainties when comparing results improving recycling behaviour. from pick analyses done during study period.  Decreasing the distance to the recycling containers and providing easy access to  Small fractions of waste, for example hazardous waste, could not be identified using information were two effective interventions in the pilot area. pick analysis as the amount is low and not sufficient for a statistically meaningful  Improvement of recycling behaviour cannot be expected without appropriate comparison. interventions.

 Convenient infrastructure for recycling, and access to information and communication channels are key factors for improving recycling behaviour. Out of them, the convenience factor was the most effective.  The factors influencing recycling behaviour are perceived differently by different

people. Hence, the interventions should be investigated, designed and applied so that

they are proper for the local circumstances. In this way, the majority of inhabitants accept and follow the interventions.

31 30

4 Conclusions

The recycling behaviour transition (RBT) procedure (Fig. 11) evaluated the current recycling behaviour, identified appropriate interventions and assessed the effect of these interventions in the pilot area. The RBT procedure proved to be an efficient tool that improved recycling significantly in the pilot area. Conclusions are listed as follows:

 The RBT procedure has a promising potential to be a standard tool for improving any waste sorting system, irrespective of the social and technical factors involved.  A participatory research framework like Action Research has the potential to solve complex problems in a waste management system.  Using mixed methods was vital for the complex studies such as recycling behaviour where both technical and social aspects are involved. A combination of qualitative and quantitative methods reinforced the reliability of the data collected.  Pick analysis was a relevant method to measure the waste sorting behaviour that quantified the effect of interventions in the system.  Semi-structured interviews are found as a relevant method to identify interventions for improving recycling behaviour.  Decreasing the distance to the recycling containers and providing easy access to information were two effective interventions in the pilot area.  Improvement of recycling behaviour cannot be expected without appropriate interventions.  Convenient infrastructure for recycling, and access to information and communication channels are key factors for improving recycling behaviour. Out of them, the convenience factor was the most effective.  The factors influencing recycling behaviour are perceived differently by different people. Hence, the interventions should be investigated, designed and applied so that they are proper for the local circumstances. In this way, the majority of inhabitants accept and follow the interventions.

5 Outlook

Doing this research and using the different methods gave new insights about waste source separation systems. During the interviews, there were several citations from the inhabitants

that declared sorting the waste was not only a duty for them; it was something that they did to take care of the environment. Some of them expressed that their environmental awareness

began due to the act of waste sorting. This changing attitude raised the following question: Is it possible to use this change to minimise waste generation, which is the uppermost level in the waste hierarchy? In my opinion, the only way to achieve it is changing the attitude towards this question. It needs some action and this act can be waste sorting. So the

hypothesis for the future research is: The act of waste sorting is a tool towards waste

reduction.

The waste sorting is a low cost task in waste management but it needs a lot of effort to consider the social aspects. For this reason, there are some doubts among engineers if waste

sorting at the source is preferred to the sorting the mixed waste by a machine. According to thermodynamics, it is known that mixing increases the entropy. For decreasing disorder in a

system, energy is needed. Therefore, it would be interesting to compare source separation and waste sorting by machines in the terms of thermodynamics, where can be seen in term of energy and entropy changes.

The RBT procedure needs to be tested in other cities with different circumstances. It would be interesting to develop this procedure, reducing the cost of its implementation and

complement it with other tools such life cycle assessment. It would also be interesting to

improve the RBT procedure for creating a waste sorting system from scratch. This is suggested for future work.

A large part of the waste that should be sorted and recycled is from packaging. There

are some packaging industries that already work on the design of packaging so that it can be recycled, thereby decreasing the environmental impact. In the next step, the design of

packaging should assist the user to know how and why they should sort the packaging.

Source separation and waste sorting at home are vital for a sustainable waste management system where the user gets involved in the technical system. The knowledge from this study showed that this user involvement can have impacts on social, environmental

33 32

5 Outlook

Doing this research and using the different methods gave new insights about waste source separation systems. During the interviews, there were several citations from the inhabitants that declared sorting the waste was not only a duty for them; it was something that they did to take care of the environment. Some of them expressed that their environmental awareness began due to the act of waste sorting. This changing attitude raised the following question: Is it possible to use this change to minimise waste generation, which is the uppermost level in the waste hierarchy? In my opinion, the only way to achieve it is changing the attitude towards this question. It needs some action and this act can be waste sorting. So the hypothesis for the future research is: The act of waste sorting is a tool towards waste reduction.

The waste sorting is a low cost task in waste management but it needs a lot of effort to consider the social aspects. For this reason, there are some doubts among engineers if waste sorting at the source is preferred to the sorting the mixed waste by a machine. According to thermodynamics, it is known that mixing increases the entropy. For decreasing disorder in a system, energy is needed. Therefore, it would be interesting to compare source separation and waste sorting by machines in the terms of thermodynamics, where can be seen in term of energy and entropy changes.

The RBT procedure needs to be tested in other cities with different circumstances. It would be interesting to develop this procedure, reducing the cost of its implementation and complement it with other tools such life cycle assessment. It would also be interesting to improve the RBT procedure for creating a waste sorting system from scratch. This is suggested for future work.

A large part of the waste that should be sorted and recycled is from packaging. There are some packaging industries that already work on the design of packaging so that it can be recycled, thereby decreasing the environmental impact. In the next step, the design of packaging should assist the user to know how and why they should sort the packaging.

33 and economic perspectives which are the three key elements in sustainable development. The References waste sorting task has an evolutionary impact on inhabitants’ attitudes and behaviour towards recycling and environmental issues. In the beginning, they may see sorting as a duty or as a respect of regulations. After a while, they find it as contribution to, and protection of the Ajzen, I. (1985). From Intentions to Actions: A Theory of Planned Behavior. Action Control: From Cognition to Behavior. J. Kuhl and J. Beckmann. Berlin, Heidelberg, Springer Berlin environment for developing a sustainable society. Involving the people in the system is not an Heidelberg: 11-39. easy task, but it is a necessary investment for a long-term sustainable society. Therefore, it is Ajzen, I. and M. Fishbein (1980) Understanding attitudes and predicting social behaviour. Englewood Cliffs, NJ: Prentice-Hall. recommended to policy makers, waste managers and stakeholders in a waste management Ando, A. W. and A. Y. Gosselin (2005) Recycling in multifamily dwellings: does convenience system to set long-term goals in their plan, and put the source separation as the core of the matter? Economic inquiry 43(2): 426-438. waste management system that they want to launch. Barbalace, R. C. (2003) The History of Waste: Do you want to be a garbologist? The History of Waste: Do You Want to Be a Garbologist.

Barker, K., L. Fong, S. Grossman, C. Quin and R. Reid (1994) Comparison of self-reported recycling attitudes and behaviors with actual behavior. Psychological Reports 75(1): 571-577.

Barr, S. (2002). Household Waste in Social Perspective; Values, attitudes, situation and behaviour. Burlington, Ashgate. Beigl, P., S. Lebersorger and S. Salhofer (2008) Modelling municipal solid waste generation: A review. Waste management 28(1): 200-214. Bernstad, A. (2014) Household food waste separation behavior and the importance of convenience.

Waste Management 34(7): 1317-1323. Boldero, J. (1995) The prediction of household recycling of newspapers: The role of attitudes, intentions, and situational factors1. Journal of Applied Social Psychology 25(5): 440-462. Chen, X., Y. Geng and T. Fujita (2010) An overview of municipal solid waste management in China. Waste management 30(4): 716-724.

Corral-Verdugo, V. c. (1997) Dual ‘realities’ of conservation behavior: self-reports vs observation of re-use and recycling behavior. Journal of environmental psychology 17(2): 135-145.

Cote, J. A., J. McCullough and M. Reilly (1985) Effects of unexpected situations on behavior- intention differences: A garbology analysis. Journal of Consumer Research 12(2): 188-194. Dahlén, L. and A. Lagerkvist (2008) Methods for household waste composition studies. Waste Management 28(7): 1100-1112.

Dahlén, L., H. Åberg, A. Lagerkvist and P. E. O. Berg (2009) Inconsistent pathways of household waste. Waste management 29(6): 1798-1806. EU. (2008). Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on Waste and Repealing Certain Directives, European Commission, Brussels, Belgium. European Commission (2015). Closing the loop - An EU plan for the Circular Economy, Brussels. Fahy, F. and A. Davies (2007) Home improvements: Household waste minimisation and action research. Resources, Conservation and Recycling 52(1): 13-27. Goldenhar, L. M. and C. M. Connell (1993) Understanding and predicting recycling behavior: An application of the theory of reasoned action. Journal of Environmental Systems 22: 91-91.

González-Torre, P. L. and B. Adenso-Díaz (2005) Influence of distance on the motivation and frequency of household recycling. Waste management 25(1): 15-23. Guagnano, G. A., P. C. Stern and T. Dietz (1995) Influences on attitude-behavior relationships a natural experiment with curbside recycling. Environment and behavior 27(5): 699-718.

35 34

References

Ajzen, I. (1985). From Intentions to Actions: A Theory of Planned Behavior. Action Control: From Cognition to Behavior. J. Kuhl and J. Beckmann. Berlin, Heidelberg, Springer Berlin Heidelberg: 11-39. Ajzen, I. and M. Fishbein (1980) Understanding attitudes and predicting social behaviour. Englewood Cliffs, NJ: Prentice-Hall. Ando, A. W. and A. Y. Gosselin (2005) Recycling in multifamily dwellings: does convenience matter? Economic inquiry 43(2): 426-438. Barbalace, R. C. (2003) The History of Waste: Do you want to be a garbologist? The History of Waste: Do You Want to Be a Garbologist. Barker, K., L. Fong, S. Grossman, C. Quin and R. Reid (1994) Comparison of self-reported recycling attitudes and behaviors with actual behavior. Psychological Reports 75(1): 571-577. Barr, S. (2002). Household Waste in Social Perspective; Values, attitudes, situation and behaviour. Burlington, Ashgate. Beigl, P., S. Lebersorger and S. Salhofer (2008) Modelling municipal solid waste generation: A review. Waste management 28(1): 200-214. Bernstad, A. (2014) Household food waste separation behavior and the importance of convenience. Waste Management 34(7): 1317-1323. Boldero, J. (1995) The prediction of household recycling of newspapers: The role of attitudes, intentions, and situational factors1. Journal of Applied Social Psychology 25(5): 440-462. Chen, X., Y. Geng and T. Fujita (2010) An overview of municipal solid waste management in China. Waste management 30(4): 716-724. Corral-Verdugo, V. c. (1997) Dual ‘realities’ of conservation behavior: self-reports vs observation of re-use and recycling behavior. Journal of environmental psychology 17(2): 135-145. Cote, J. A., J. McCullough and M. Reilly (1985) Effects of unexpected situations on behavior- intention differences: A garbology analysis. Journal of Consumer Research 12(2): 188-194. Dahlén, L. and A. Lagerkvist (2008) Methods for household waste composition studies. Waste Management 28(7): 1100-1112. Dahlén, L., H. Åberg, A. Lagerkvist and P. E. O. Berg (2009) Inconsistent pathways of household waste. Waste management 29(6): 1798-1806. EU. (2008). Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on Waste and Repealing Certain Directives, European Commission, Brussels, Belgium. European Commission (2015). Closing the loop - An EU plan for the Circular Economy, Brussels. Fahy, F. and A. Davies (2007) Home improvements: Household waste minimisation and action research. Resources, Conservation and Recycling 52(1): 13-27. Goldenhar, L. M. and C. M. Connell (1993) Understanding and predicting recycling behavior: An application of the theory of reasoned action. Journal of Environmental Systems 22: 91-91. González-Torre, P. L. and B. Adenso-Díaz (2005) Influence of distance on the motivation and frequency of household recycling. Waste management 25(1): 15-23. Guagnano, G. A., P. C. Stern and T. Dietz (1995) Influences on attitude-behavior relationships a natural experiment with curbside recycling. Environment and behavior 27(5): 699-718.

Guerin, D., J. Crete and J. Mercier (2001) A multilevel analysis of the determinants of recycling McCarty, J. A. and L. J. Shrum (1994) The recycling of solid wastes: Personal values, value behavior in the European countries. Social Science Research 30(2): 195-218. orientations, and attitudes about recycling as antecedents of recycling behavior. Journal of Business Research 30(1): 53-62. Gutberlet, J. (2008) Empowering collective recycling initiatives: Video documentation and action research with a recycling co-op in Brazil. Resources, Conservation and recycling 52(4): 659- Meneses, G. D. and A. B. Palacio (2005) Recycling behavior a multidimensional approach. 670. Environment and Behavior 37(6): 837-860. Henry, R. K., Z. Yongsheng and D. Jun (2006) Municipal solid waste management challenges in Miafodzyeva, S. and N. Brandt (2013) Recycling Behaviour Among Householders: Synthesizing developing countries–Kenyan case study. Waste management 26(1): 92-100. Determinants Via a Meta-analysis. Waste and Biomass Valorization 4(2): 221-235. Hillman, K., A. Damgaard, O. Eriksson, D. l Jonsson and L. Fluck (2015). Climate Benefits of Myers, M. D. (2009). Qualitative research in business and management. London, Sage. Material Recycling, Inventory of Average Greenhouse Gas Emissions for Denmark, Norway Oskamp, S., M. J. Harrington, T. C. Edwards, D. L. Sherwood, S. M. Okuda and D. C. Swanson and Sweden. Nordic Council of Ministers, Denmark. (1991) Factors influencing household recycling behavior. Environment and behavior 23(4): Hoornweg, D. and P. Bhada-Tata (2012). What a waste: a global review of solid waste management. 494-519. Urban development series knowledge papers. W. Bank, Washington, DC. 15: 1-98. Park, H. S. (2000) Relationships among attitudes and subjective norms: Testing the theory of reasoned Hoornweg, D., P. Bhada-Tata and C. Kennedy (2013) Environment: Waste production must peak this action across cultures. Communication Studies 51(2): 162-175. century. Nature 502(7473): 615. Park, H. S. and T. R. Levine (1999) The theory of reasoned action and self construal: Evidence from Hoornweg, D., P. Bhada-Tata and C. Kennedy (2013) Waste production must peak this century. three cultures. Communication Monographs 66(3): 199-218. ‐ Nature 502(7473): 615-617. Park, P. (1999) People, knowledge, and change in participatory research. Management learning 30(2): Jackson, A. L., J. E. Olsen, K. L. Granzin and A. C. Burns (1993) An investigation of determinants of 141-157. recycling consumer behavior. NA-Advances in Consumer Research Volume 20: 481-487. Pieters, R. G. M. (1989). Attitudes and behavior in a source separation program: A Garbology Jesson, J. (2009) Household waste recycling behavior: A market segmentation model. Social Approach. Doctoral Thesis, Delft: Eurobon. Marketing Quarterly 15(2): 25-38. Rajamanikam, R., G. Poyyamoli and S. Kumar (2014) The role of non-governmental organizations in Jones, P. R. E. (1989) Understanding paper recycling in an institutionally supportive setting: An residential solid waste management: A case study of Puducherry, a coastal city of India. Waste application of the theory of reasoned action. Journal of Environmental Systems 19(4): 307- Management & Research 32(9): 867-881. 321. Rathje, W. L. (1984) The garbage decade. American Behavioral Scientist 28(1): 9-29. Katsenou, C., E. Flogaitis and G. Liarakou (2013) Exploring pupil participation within a sustainable Reid, D. H., P. D. Luyben, R. J. Rawers and J. S. Bailey (1976) Newspaper recycling behavior: The school. Cambridge journal of education 43(2): 243-258. effects of prompting and proximity of containers. Environment and Behavior 8(3): 471-482. Kemmis, S. (1980) Action Research in Retrospect and Prospect. Annual Meeting for the Australian Rousta, K. and L. Dahlén (2015). Source Separation of Household Waste; Technology and Social Association for Research in Education. Aspects. Resource Recovery to Approach Zero Municipal Waste. M. J. Taherzadeh and T. Kim, S., S.-H. Jeong and Y. Hwang (2012) Predictors of pro-environmental behaviors of American Richards. Boca Raton, CRC Press: 61-77. and Korean students: The application of the theory of reasoned action and protection Rousta, K., T. Richards and M. J. Taherzadeh (2015). An overview of solid waste management toward motivation theory. Science Communication: 1075547012441692. zero landfill: a swedish model. Resource Recovery to Approach Zero Municipal Waste. M. J. Kvale, S. and S. Brinkmann (2009). Den kvalitativa forskningsintervjun. Lund, Studentlitteratur AB. Taherzadeh and T. Richards. Boca Raton, CRC Press: 1-22. Lagerkvist, A. (2017) Course material A0013B Waste Sience and Technology. Luleå University of Schultz, P., S. Oskamp and T. Mainieri (1995) Who recycles and when? A review of personal and Technology. situational factors. Journal of environmental psychology 15(2): 105-121. Lagerkvist, A., H. Ecke and T. H. Christensen (2011). Waste characterization: approaches and Schwartz, S. H. (1977) Normative influences on altruism. Advances in experimental social psychology methods. SolidWaste Technology & Management. T. H. Christensen. Malaysia, Wiley: 63-84. 10: 221-279. Largo-Wight, E., H. Bian and L. Lange (2012) An empirical test of an expanded version of the Theory SFS1994a Förordningen om producentansvar för returpapper [The Ordinance of Producers’ of Planned Behavior in predicting recycling behavior on campus. American Journal of Health Responsibility for Waste Paper]. Swedish legislation. SFS 1994:1205 (in Swedish). Education 43(2): 66-73. SFS1994b Förordningen om producentansvar för förpackningar [The Ordinance of Producers’ Laurenti, R. and Å. Stenmarck (2015). Produkters totala avfall - studie om avfallets fotavtryck och Responsibility for Packaging Materials]. Swedish legislation. SFS 1994:1235 (in Swedish). klimatkostnader [Products total waste- study on waste footprint and cilmate costs] Swedish SMED (2016). Avfall i Sverige 2014 [Waste in Sweden 2014]. Swedish Environmental Protection Waste Management Association, Malmö (in Swedish). Agency, Stockholm (in Swedish). Martin, M., I. D. Williams and M. Clark (2006) Social, cultural and structural influences on household Stringer, E. T. (2014). Action Research. Los Angeles, SAGE. waste recycling: A case study. Resources, conservation and recycling 48(4): 357-395. Swedish Statistisc (2012). Socio-economic statistics. SCB, Stockholm (in Swedish). Mattsson, C. (2003). Waste management and producer responsibility in Sweden. Licentiate Thesis, Chalmers University of Technology, Gothenburg. 37 36

McCarty, J. A. and L. J. Shrum (1994) The recycling of solid wastes: Personal values, value orientations, and attitudes about recycling as antecedents of recycling behavior. Journal of Business Research 30(1): 53-62. Meneses, G. D. and A. B. Palacio (2005) Recycling behavior a multidimensional approach. Environment and Behavior 37(6): 837-860. Miafodzyeva, S. and N. Brandt (2013) Recycling Behaviour Among Householders: Synthesizing Determinants Via a Meta-analysis. Waste and Biomass Valorization 4(2): 221-235. Myers, M. D. (2009). Qualitative research in business and management. London, Sage. Oskamp, S., M. J. Harrington, T. C. Edwards, D. L. Sherwood, S. M. Okuda and D. C. Swanson (1991) Factors influencing household recycling behavior. Environment and behavior 23(4): 494-519. Park, H. S. (2000) Relationships among attitudes and subjective norms: Testing the theory of reasoned action across cultures. Communication Studies 51(2): 162-175. Park, H. S. and T. R. Levine (1999) The theory of reasoned action and self construal: Evidence from three cultures. Communication Monographs 66(3): 199-218. ‐ Park, P. (1999) People, knowledge, and change in participatory research. Management learning 30(2): 141-157. Pieters, R. G. M. (1989). Attitudes and behavior in a source separation program: A Garbology Approach. Doctoral Thesis, Delft: Eurobon. Rajamanikam, R., G. Poyyamoli and S. Kumar (2014) The role of non-governmental organizations in residential solid waste management: A case study of Puducherry, a coastal city of India. Waste Management & Research 32(9): 867-881. Rathje, W. L. (1984) The garbage decade. American Behavioral Scientist 28(1): 9-29. Reid, D. H., P. D. Luyben, R. J. Rawers and J. S. Bailey (1976) Newspaper recycling behavior: The effects of prompting and proximity of containers. Environment and Behavior 8(3): 471-482. Rousta, K. and L. Dahlén (2015). Source Separation of Household Waste; Technology and Social Aspects. Resource Recovery to Approach Zero Municipal Waste. M. J. Taherzadeh and T. Richards. Boca Raton, CRC Press: 61-77. Rousta, K., T. Richards and M. J. Taherzadeh (2015). An overview of solid waste management toward zero landfill: a swedish model. Resource Recovery to Approach Zero Municipal Waste. M. J. Taherzadeh and T. Richards. Boca Raton, CRC Press: 1-22. Schultz, P., S. Oskamp and T. Mainieri (1995) Who recycles and when? A review of personal and situational factors. Journal of environmental psychology 15(2): 105-121. Schwartz, S. H. (1977) Normative influences on altruism. Advances in experimental social psychology 10: 221-279. SFS1994a Förordningen om producentansvar för returpapper [The Ordinance of Producers’ Responsibility for Waste Paper]. Swedish legislation. SFS 1994:1205 (in Swedish). SFS1994b Förordningen om producentansvar för förpackningar [The Ordinance of Producers’ Responsibility for Packaging Materials]. Swedish legislation. SFS 1994:1235 (in Swedish). SMED (2016). Avfall i Sverige 2014 [Waste in Sweden 2014]. Swedish Environmental Protection Agency, Stockholm (in Swedish). Stringer, E. T. (2014). Action Research. Los Angeles, SAGE. Swedish Statistisc (2012). Socio-economic statistics. SCB, Stockholm (in Swedish).

Swedish Waste Management Association (2005). Manual för plockanalys av hushållsavfall [Manual for Pick Analysis of Houshold Waste]. Malmö (in Swedish). Swedish Waste Management Association (2013). Manual för plockanalys av hushållsavfall [Manual for Pick Analysis of Houshold Waste]. Malmö (in Swedish). Swedish Waste Management Association (2017). Hushållsavfall i siffror- Kommun- och länsstatistik 2017 [Household waste statistics in Sweden 2017]. Malmö (in Swedish). Tai, J., W. Zhang, Y. Che and D. Feng (2011) Municipal solid waste source-separated collection in China: A comparative analysis. Waste management 31(8): 1673-1682. Tashakkori, A. and C. Teddlie (2010). Sage handbook of mixed methods in social & behavioral research. California, Sage. Taylor, S. and P. Todd (1995) An integrated model of waste management behavior: A test of household recycling and composting intentions. Environment and behavior 27(5): 603-630. Tchobanoglous, G., H. Theisen and S. A. Vigil (1993). Integrated solid waste management : engineering principles and management issues. New York, McGraw-Hill. Thøgersen, J. (1997) Facilitating Recycling: Reverse—Distribution Channel Design for Participation and Support. Social Marketing Quarterly 4(1): 42-55. Tonglet, M., P. S. Phillips and A. D. Read (2004) Using the Theory of Planned Behaviour to investigate the determinants of recycling behaviour: a case study from Brixworth, UK. Resources, Conservation and Recycling 41(3): 191-214. Tucker, P. (2001) Understanding recycling behaviour. Paper Technology 42(9): 51-54. Tucker, P. and D. Speirs (2003) Attitudes and behavioural change in household waste management behaviours. Journal of environmental planning and management 46(2): 289-307. Tucker, P. and D. Speirs (2003) Count me as doing my bit: An appraisal of the accuracy of waste attitude/behaviour surveys. CIWM Sci. Tech. Rev 4: 2003. Xevgenos, D., C. Papadaskalopoulou, V. Panaretou, K. Moustakas and D. Malamis (2015) Success stories for recycling of MSW at municipal level: a review. Waste and Biomass Valorization 6(5): 657-684. Ölander, F. and J. Thøgersen (1995) Understanding of consumer behaviour as a prerequisite for environmental protection. Journal of Consumer Policy 18(4): 345-385.

Paper I

Rousta, K., Ordoñez, I., Bolton, K., Dahlén, L., (2017)

Support for Designing Waste Sorting Systems: A Mini Review

Waste management & Research 35(11): 1099-1111 PAPER I

WMR0010.1177/0734242X17726164Waste Management & ResearchRousta et al. 726164research-article2017

Mini-review Article

Waste Management & Research 2017, Vol. 35(11) 1099 –1111 Support for designing waste sorting © The Author(s) 2017 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav systems: A mini review https://doi.org/10.1177/0734242X17726164DOI: 10.1177/0734242X17726164 journals.sagepub.com/home/wmr

Kamran Rousta1, Isabel Ordoñez2, Kim Bolton1 and Lisa Dahlén3

Abstract This article presents a mini review of research aimed at understanding material recovery from municipal solid waste. It focuses on two areas, waste sorting behaviour and collection systems, so that research on the link between these areas could be identified and evaluated. The main results presented and the methods used in the articles are categorised and appraised. The mini review reveals that most of the work that offered design guidelines for waste management systems was based on optimising technical aspects only. In contrast, most of the work that focused on user involvement did not consider developing the technical aspects of the system, but was limited to studies of user behaviour. The only clear consensus among the articles that link user involvement with the technical system is that convenient waste collection infrastructure is crucial for supporting source separation. This mini review reveals that even though the connection between sorting behaviour and technical infrastructure has been explored and described in some articles, there is still a gap when using this knowledge to design waste sorting systems. Future research in this field would benefit from being multidisciplinary and from using complementary methods, so that holistic solutions for material recirculation can be identified. It would be beneficial to actively involve users when developing sorting infrastructures, to be sure to provide a waste management system that will be properly used by them.

Keywords Source separation, household waste, recycling behaviour, resource recovery, sustainability, waste management, sorting infrastructure

Received 9th January 2017, accepted 18th July 2017 by Associate Editor Rodrigo Navia.

Introduction The European Commission has stated their intention of moving a sustainable municipal solid WM system has six main functions: from a linear production model to a circular one (European (1) waste generation, (2) waste handling and separation at the Commission, 2014, 2015). This means, among other things, that source, (3) collection, (4) transfer and transport, (5) processing and the Commission has set targets for recycling, with a minimum of transformation of waste in material recovery facilities and (6) dis- 65% recycling of municipal waste, and a reduction of landfilling of posal. This is illustrated in Figure 1, where treatment corresponds all waste to less than 10% by 2030. During 2014, European coun- to the processing and transformation of waste to useful resources. tries recycled and composted 44% of their municipal waste, and This may be done in several ways, through material recovery, recy- landfilled 28%. There are, however, positive trends in European cling, incineration with energy recovery, biogas production and waste management (WM), with a continuous reduction of the frac- composting. It is the treatment function of WM that links back to tion of waste that is landfilled, and increasing rates of recycling, material production, in a ‘new’ waste to resource stage. However, composting and incineration since 1995 (Eurostat, 2016). In order WM systems can only recirculate the materials that they receive, to achieve the mentioned targets, the action plan developed by the Commission states that waste separation and collection systems 1 have to be improved to minimise recycling and reuse costs Swedish Centre for Resource Recovery, University of Borås, Borås, Sweden (European Commission, 2015). Also, to achieve the goal of a cir- 2 Division of Design & Human Factors, Chalmers University of cular economy, the action plan suggests measures at all stages of Technology, Göteborg, Sweden 3 material use, that is, production, consumption, WM and ‘from Waste Science and Technology, Luleå University of Technology, Luleå, Sweden waste to resources’ (European Commission, 2015). It seems clear that current WM systems have to be transformed into structures Corresponding author: that enable material recirculation (i.e. reuse, re-manufacturing and Isabel Ordoñez, Division of Design & Human Factors, Department of Industrial and Materials Science, Chalmers University of Technology, recycling) by improving all stages of the WM system, from waste 412 96 Göteborg, Sweden. generation to disposal. According to Tchobanoglous et al. (1993), Email: [email protected] WMR0010.1177/0734242X17726164Waste Management & ResearchRousta et al. 726164research-article2017

Mini-review Article

Kamran Rousta1, Isabel Ordoñez2, Kim Bolton1 and Lisa Dahlén3

Keywords Source separation, household waste, recycling behaviour, resource recovery, sustainability, waste management, sorting infrastructure

Received 9th January 2017, accepted 18th July 2017 by Associate Editor Rodrigo Navia.

Introduction The European Commission has stated their intention of moving a sustainable municipal solid WM system has six main functions: from a linear production model to a circular one (European (1) waste generation, (2) waste handling and separation at the Commission, 2014, 2015). This means, among other things, that source, (3) collection, (4) transfer and transport, (5) processing and the Commission has set targets for recycling, with a minimum of transformation of waste in material recovery facilities and (6) dis- 65% recycling of municipal waste, and a reduction of landfilling of posal. This is illustrated in Figure 1, where treatment corresponds all waste to less than 10% by 2030. During 2014, European coun- to the processing and transformation of waste to useful resources. tries recycled and composted 44% of their municipal waste, and This may be done in several ways, through material recovery, recy- landfilled 28%. There are, however, positive trends in European cling, incineration with energy recovery, biogas production and waste management (WM), with a continuous reduction of the frac- composting. It is the treatment function of WM that links back to tion of waste that is landfilled, and increasing rates of recycling, material production, in a ‘new’ waste to resource stage. However, composting and incineration since 1995 (Eurostat, 2016). In order WM systems can only recirculate the materials that they receive, to achieve the mentioned targets, the action plan developed by the Commission states that waste separation and collection systems 1 have to be improved to minimise recycling and reuse costs Swedish Centre for Resource Recovery, University of Borås, Borås, Sweden (European Commission, 2015). Also, to achieve the goal of a cir- 2 Division of Design & Human Factors, Chalmers University of cular economy, the action plan suggests measures at all stages of Technology, Göteborg, Sweden 3 material use, that is, production, consumption, WM and ‘from Waste Science and Technology, Luleå University of Technology, Luleå, Sweden waste to resources’ (European Commission, 2015). It seems clear that current WM systems have to be transformed into structures Corresponding author: that enable material recirculation (i.e. reuse, re-manufacturing and Isabel Ordoñez, Division of Design & Human Factors, Department of Industrial and Materials Science, Chalmers University of Technology, recycling) by improving all stages of the WM system, from waste 412 96 Göteborg, Sweden. generation to disposal. According to Tchobanoglous et al. (1993), Email: [email protected] 1100 Waste Management & Research 35(11) Rousta et al. 1101

was limited to peer-reviewed scientific articles to ensure an Most of the articles were authored by researchers from acceptable quality. Key phrases were chosen to find articles pre- European countries (28 articles), followed by North America senting research on recycling behaviour and sorting and collec- (12 articles). Of the remaining 11 articles, nine were from Asia tion systems, as well as the link between them. Other WM and two compared sorting and collection systems from loca- functions were excluded. The key phrase ‘recycling behaviour’ tions on different continents. The European articles were evenly (including American spelling) was used for finding articles about divided between engineering and social sciences disciplines, the behavioural aspects of this phenomenon, resulting in approxi- whereas Asian articles were mostly from the engineering field mately 290 articles. The key phrases used to search for articles (seven articles, with two from social sciences) and the North about sorting and collection systems were ‘waste collection sys- American articles were mainly from social sciences (10 in tem’, ‘source separation’ and ‘recycling’, with the condition that social sciences and two in engineering). European authors they all appeared together. This resulted in 70 articles. Abstracts wrote both of the multidisciplinary articles. The most frequent from both searches were screened, choosing the articles that ful- journals where these articles were published were environment filled two criteria: (1) they focused on household waste or munic- and behaviour (8 articles), waste management and research (7), ipal solid waste, and (2) they investigated user participation in waste management (5), resource conservation and recycling recycling programmes and/or evaluated the sorting and collec- (5), environmental management (5), social marketing quarterly tion systems. The search for articles continued using the snow- (2) and waste biomass valour (2). The remaining 17 articles balling method as discussed by Hagen-Zanker and Mallet (2013) were published in 17 different journals. Figure 1. Functions of household WM, modified from Tchobanoglous et al. (1993). The review focused on the part of the to find more articles within the criteria of the study. Although one Of the articles reviewed, 29 articles addressed WM in urban system marked by dashed lines. cannot guarantee that all relevant articles were found, it is environments, 18 referred to both rural and urban contexts, while expected that these key phrases found a representative sample of only one article studied WM in a purely rural setting. In addition, making the collection and separation functions of WM systems behaviour and the technical aspects of the waste sorting and col- relevant articles in these fields. two articles obtained their data from WM of entire countries, and crucial for achieving high recirculation rates. The quality of the lection infrastructure. The infrastructure available for waste sort- The screening identified 51 articles that were relevant for fur- one literature review did not specify the (urban or rural) context secondary raw materials obtained from municipal waste depends ing and collection generates an environment where users can ther reviewing. Two aspects of the articles were of particular of the articles it addressed. initially on the quality of the source-sorted materials, as well as execute their behaviour. The way that this environment is config- interest for the review: (1) the results that were obtained and (2) how these materials are collected and transported and, perhaps ured influences this behaviour, whether it is planned or not the research methods that were used. Summary of the results most importantly, on how they are processed at the treatment facil- (Thaler et al., 2014). Often called nudging, designing for an The articles were summarised, listing authors, article title and ities. Separation and collection are two distinct functions of the intended behaviour could be a useful tool to engage inhabitants in journal name, scientific discipline of the first author, year of pub- The results found in the articles and that were relevant to this WM system that are in practise very closely related and are there- increasing the fractions of sorted waste. Such an approach to lication, methods used, results obtained, waste fractions consid- study were organised into an affinity diagram. As seen in Figure fore difficult to analyse separately. These two terms are used inter- designing waste sorting systems would need support from cogni- ered, how they describe the user’s role, the design of the system 2, the mini review revealed that the results could be grouped into changeably in this article. tive-psychology experts, but it should also be based on the ample and if they corresponded to urban or rural contexts. It was consid- five main categories, of which one had subcategories. Figure 2 Within Europe, the Swedish WM system is considered as a research available on waste sorting systems. Therefore, it is rel- ered relevant to make the distinction between urban and rural, also shows the number of articles in each category. The results good example of how policy decisions can help achieve WM tar- evant to review what has been found in previous research, both since this context may affect recycling behaviour and the collec- from each of these categories are discussed in detail below. gets. Landfilling of municipal solid waste has reduced dramati- regarding the socio-psychological and technical aspects at the tion system that is implemented. The quality of the articles was Differences between urban and rural contexts are presented at the cally, from almost 62% of the total household waste in 1975 (i.e. waste sorting stage. appraised by evaluating the consistency between the results and end of this section. It is important to note that the reviewed arti- 1,500,000 t) to less than 1% in 2014 (i.e. 32,900 t). This has been This article reviews literature on recycling behaviour and the methodology used, how the study design was described and if cles have additional results that have not been discussed here, achieved by landfill bans of organic waste and landfill tax on waste sorting systems for household solid waste to obtain results it was relevant to the results obtained. All of the results were since they were not relevant to this review. inorganic waste, which supports increased material recycling, that can support the design of future systems that facilitate the grouped into common topics and subsequently categorised into energy recovery and biological treatment of waste (Swedish recovery of a larger fraction of material resources. It does not an affinity diagram. Factors affecting recycling behaviour. As shown in Figure 2, Waste Management Association, 2015). Although the Swedish focus on specific material recycling technologies, nor does it the category that was discussed in most articles was factors that WM system is considered to function well, there is much room intend to review all factors that affect waste sorting behaviour. Results and discussion affect recycling behaviour (i.e. waste separation at the source). for improvement. Despite increases in material recycling, it is Rather the intention is to highlight aspects that can be used when The subcategory that was discussed in most of these articles (25 estimated that 70% of the recyclable and biodegradable waste is designing sorting infrastructure, to improve resource recovery. Not surprisingly, the 51 articles came from two broad research articles) was physical infrastructure, of which 14 conclude that collected unsorted and subsequently incinerated, regardless of The review is based on research published since the early 1990s fields: Social sciences (which focused mostly on users’ behav- convenient infrastructure improves recycling. Refsgaard and the ample sorting infrastructure available in the country (Swedish and focuses on two fields: Recycling behaviour/waste sorting at iour) and engineering (focused mostly on the sorting infrastruc- Magnussen (2009) even state that it affects attitudes and behav- Waste Management Association, 2015). Inhabitants’ participa- the source and collection systems. These areas were chosen so ture). Of these articles, 26 were from social sciences, 23 were iour of the users, and Porter et al. (1995) conclude that changes in tion in waste separation schemes seems crucial for the improve- that the relation between the users and the infrastructure that they from engineering and two were multidisciplinary studies. Even infrastructure have long-term effects on recycling behaviour. ments needed to achieve more ambitious material recycling goals interact with could be studied. Also, these studies are expected to though the articles fit mostly into one of the research fields, this Physical infrastructure was not only considered as collection (Rousta and Dahlén, 2015). However, it is unclear how to design shed light on the importance of this relation for efficient material does not mean that all of the articles have exclusively focused on points or the presence of kerbside collection, but also available waste sorting and collection systems that engage inhabitants to recovery. This review should be of interest both to researchers user behaviour or infrastructure. In fact, almost half of the articles space within each household, which was found to be a determin- increase their waste sorting. Therefore, it is of interest to review and other stakeholders involved in developing WM systems. reviewed discussed a combination of both social and technical ing factor for participation in recycling schemes (Ando and Gos- the literature focused on improving household waste sorting, by aspects (a total of 24 articles; including the two multidisciplinary selin, 2005; Jesson, 2009; McDonald and Oates, 2003). addressing the link between source sorting behaviour and the col- studies, seven articles from engineering and 15 from social sci- Methods In this subcategory, seven articles studied the role of distance lection system that supports it (as marked in Figure 1). ences). The articles were published between 1991 and 2016, with to the collection point as an influential factor in recycling behav- There is a large amount of research on how to improve waste Once the purpose of the review had been established, the search a uniform distribution over this period for both the articles from iour. A large distance between the collection point and the source sorting, both regarding the socio-psychological factors of sorting for articles started with the Web of Science database. The review the social sciences and the engineering disciplines. of waste generation reduces the collection of recyclable materials Rousta et al. 1101 was limited to peer-reviewed scientific articles to ensure an Most of the articles were authored by researchers from acceptable quality. Key phrases were chosen to find articles pre- European countries (28 articles), followed by North America senting research on recycling behaviour and sorting and collec- (12 articles). Of the remaining 11 articles, nine were from Asia tion systems, as well as the link between them. Other WM and two compared sorting and collection systems from loca- functions were excluded. The key phrase ‘recycling behaviour’ tions on different continents. The European articles were evenly (including American spelling) was used for finding articles about divided between engineering and social sciences disciplines, the behavioural aspects of this phenomenon, resulting in approxi- whereas Asian articles were mostly from the engineering field mately 290 articles. The key phrases used to search for articles (seven articles, with two from social sciences) and the North about sorting and collection systems were ‘waste collection sys- American articles were mainly from social sciences (10 in tem’, ‘source separation’ and ‘recycling’, with the condition that social sciences and two in engineering). European authors they all appeared together. This resulted in 70 articles. Abstracts wrote both of the multidisciplinary articles. The most frequent from both searches were screened, choosing the articles that ful- journals where these articles were published were environment filled two criteria: (1) they focused on household waste or munic- and behaviour (8 articles), waste management and research (7), ipal solid waste, and (2) they investigated user participation in waste management (5), resource conservation and recycling recycling programmes and/or evaluated the sorting and collec- (5), environmental management (5), social marketing quarterly tion systems. The search for articles continued using the snow- (2) and waste biomass valour (2). The remaining 17 articles balling method as discussed by Hagen-Zanker and Mallet (2013) were published in 17 different journals. to find more articles within the criteria of the study. Although one Of the articles reviewed, 29 articles addressed WM in urban cannot guarantee that all relevant articles were found, it is environments, 18 referred to both rural and urban contexts, while expected that these key phrases found a representative sample of only one article studied WM in a purely rural setting. In addition, relevant articles in these fields. two articles obtained their data from WM of entire countries, and The screening identified 51 articles that were relevant for fur- one literature review did not specify the (urban or rural) context ther reviewing. Two aspects of the articles were of particular of the articles it addressed. interest for the review: (1) the results that were obtained and (2) the research methods that were used. Summary of the results The articles were summarised, listing authors, article title and journal name, scientific discipline of the first author, year of pub- The results found in the articles and that were relevant to this lication, methods used, results obtained, waste fractions consid- study were organised into an affinity diagram. As seen in Figure ered, how they describe the user’s role, the design of the system 2, the mini review revealed that the results could be grouped into and if they corresponded to urban or rural contexts. It was consid- five main categories, of which one had subcategories. Figure 2 ered relevant to make the distinction between urban and rural, also shows the number of articles in each category. The results since this context may affect recycling behaviour and the collec- from each of these categories are discussed in detail below. tion system that is implemented. The quality of the articles was Differences between urban and rural contexts are presented at the appraised by evaluating the consistency between the results and end of this section. It is important to note that the reviewed arti- the methodology used, how the study design was described and if cles have additional results that have not been discussed here, it was relevant to the results obtained. All of the results were since they were not relevant to this review. grouped into common topics and subsequently categorised into an affinity diagram. Factors affecting recycling behaviour. As shown in Figure 2, the category that was discussed in most articles was factors that Results and discussion affect recycling behaviour (i.e. waste separation at the source). The subcategory that was discussed in most of these articles (25 Not surprisingly, the 51 articles came from two broad research articles) was physical infrastructure, of which 14 conclude that fields: Social sciences (which focused mostly on users’ behav- convenient infrastructure improves recycling. Refsgaard and iour) and engineering (focused mostly on the sorting infrastruc- Magnussen (2009) even state that it affects attitudes and behav- ture). Of these articles, 26 were from social sciences, 23 were iour of the users, and Porter et al. (1995) conclude that changes in from engineering and two were multidisciplinary studies. Even infrastructure have long-term effects on recycling behaviour. though the articles fit mostly into one of the research fields, this Physical infrastructure was not only considered as collection does not mean that all of the articles have exclusively focused on points or the presence of kerbside collection, but also available user behaviour or infrastructure. In fact, almost half of the articles space within each household, which was found to be a determin- reviewed discussed a combination of both social and technical ing factor for participation in recycling schemes (Ando and Gos- aspects (a total of 24 articles; including the two multidisciplinary selin, 2005; Jesson, 2009; McDonald and Oates, 2003). studies, seven articles from engineering and 15 from social sci- In this subcategory, seven articles studied the role of distance ences). The articles were published between 1991 and 2016, with to the collection point as an influential factor in recycling behav- a uniform distribution over this period for both the articles from iour. A large distance between the collection point and the source the social sciences and the engineering disciplines. of waste generation reduces the collection of recyclable materials 1102 Waste Management & Research 35(11) Rousta et al. 1103

source (Iyer and Kashyap, 2007), but that these incentives do not that do not match the waste generated by the users, or the collec- have a sustained effect (Thøgersen, 1994). The cost of the WM tion systems may differ from one place to another, causing uncer- service for households should be at most 1%–2% of the house- tainty among the users (Henriksson et al., 2010). Åberg et al. hold income (Wilson et al., 2012) for the service to be economi- (1996) reported that the problems perceived by users before com- cally sustainable. Gellynck et al. (2011) argued that the cost of posting were not the same as those that were most decisive in collecting mixed waste is important for minimising this fraction, practice. Tucker and Speirs (2003) stated that first impressions whereas the study done by Åberg et al. (1996) indicated that a remain for a long time unless the user has sustained different reduced collection fee is less important than environmental con- experiences. Interaction between users and competent and skilled cern for changing behaviour. Pay-as-you-throw (PAYT) systems, instructors educated in WM improve recycling (Suttibak and which also provide economic incentives for sorting, are discussed Nitivattananon, 2008). later on in the section ‘Types of collection systems’. Alternative options for getting rid of materials were found to Even though socio-demographic factors are considered to affect recycling behaviour (Dahlén and Lagerkvist, 2010a; play an important role in recycling behaviour (Ando and Gosselin, McDonald and Oates, 2003). These options could be charity col- 2005), their affect remains unclear. Iyer and Kashyap (2007) lections, local bring facilities, shared sorting bins with neigh- argue that single demographic measures are weak predictors of bours or reused material for other purposes. Figure 2. An affinity diagram showing the categories and subcategories of the results. recycling behaviour, and propose a combined socio-demographic In summary, there is agreement that the factors that affect indicator. They combined income, education and occupation of recycling behaviour the most are appropriate physical infrastruc- (Dahlén et al., 2009; González-Torre and Adenso-Díaz, 2005; recycling behaviour (Do Valle et al., 2005; Gamba and Oskamp, both parents to make an indicator for social class, with which ture, shorter distance to collection points, reliable service, user Thøgersen, 1997). This has been observed in particular for light- 1994; Iyer and Kashyap, 2007; Jesson, 2009; Miafodzyeva and they conclude that people with lower social class recycle more. convenience and adequate information. Studies of demographic weight packaging (Gallardo Izquierdo et al., 2010), paper, metal Brandt, 2013), and more engaged and recurrent information strate- The combined indicator proposed in that study is different to the factors, such as income, gender and education, do not yield con- and plastic packaging (Dahlén et al., 2007) and from multi-fam- gies give better results (Hopper and Nielsen, 1991; McDonald and single indicators from other studies, making it difficult to consistent results. ily dwellings (Ando and Gosselin, 2005). González-Torre and Oates, 2003; Porter et al., 1995; Salhofer and Isaac, 2002; Schultz, clude if results from different studies support or contradict each Adenso-Díaz (2005) argue that decreasing this distance does not 1999). Awareness of the consequences of recycling (Hopper and other. Gender was found to influence recycling behaviour (Iyer Types of collection systems. Even though most of the reviewed encourage people to start sorting, but it does increase the volume Nielsen, 1991; Refsgaard and Magnussen, 2009) and the impor- and Kashyap, 2007), with Meneses and Palacio (2005) conclud- articles describe some type of collection system, six articles com- of material collected. The results presented in Thøgersen (1997) tance of one’s participation (Do Valle et al., 2005) are also impor- ing that women recycle more than men. Single family dwellings pared different types of collection systems (Dahlén and are based on a literature review where the author made use of tant. Thøgersen (1994) argues that management of the physical were reported to recycle more than multi-family dwellings (Ando Lagerkvist, 2010a; Dahlén et al., 2007; Gallardo Izquierdo et al., several studies aimed at improving the sorting infrastructure for infrastructure is as important as the management of information. and Gosselin, 2005; Oskamp et al., 1991). Dahlén and Lagerkvist 2010; Suttibak and Nitivattananon, 2008; Wilson et al., 2012; household waste collection in Denmark. Thøgersen (1997) draws Two articles discussed normative feedback, indicating that it (2010a) concluded from their literature review that waste genera- Xevgenos et al., 2015). Of these, four compared collection sys- conclusions on several aspects of the physical infrastructure and increases recycling more than information alone (Schultz, 1999), tion increases with increased household consumption, which tems (i.e. Gallardo Izquierdo et al. (2010) in Spain, Dahlén et al. concludes that the distribution channels for recyclables should but that goal-setting was even more efficient in improving recy- implies that a larger household income may increase waste gen- (2007) and Dahlén and Lagerkvist (2010a) in Sweden, and Sut- provide a service of at least as high quality as the one available cling behaviour (Porter et al., 1995). Schultz (1999) argues that eration. In contrast, another study concluded that high income tibak and Nitivattananon (2008)) focused only on collection of for mixed waste. This helps to avoid a large amount of recyclable normative feedback increases the recycling rate, but does not reduces waste generation (Gellynck et al., 2011) and increases recyclables in Thailand, while two compared the entire WM sys- material from being discarded in the mixed waste. reduce the contamination in the sorted waste. The type of informa- recycling (Gamba and Oskamp, 1994). If high-income house- tem. Xevgenos et al. (2015) compare 19 WM systems in different Several other convenience factors were discussed in the litera- tion delivered about recycling was also studied in some articles. holds recycle more because of buying more goods or if they are countries and that are considered to be best practices, while Wil- ture. Reliable service was seen to support household recycling Porter et al. (1995) and Hopper and Nielsen (1991) found that face- better in waste separation remains unclear. Dahlén and Lagerkvist son et al. (2012) compared 20 WM systems that were chosen in (Jesson, 2009; Martin et al., 2006) and increase material collec- to-face and block-leader strategies were effective ways to inform (2010a) stated that education affects waste sorting, whereas order to provide a good global representation. All articles except tion (Thøgersen, 1997). Additionally, households engage in sort- tenants, whereas Woodard et al. (2001) achieved increased food Meneses and Palacio (2005) argued that it does not lead to Xevgenos et al. (2015) used data from local authorities, which ing if it is perceived as requiring little effort (Do Valle et al., waste collection by using an information campaign with open dis- improved sorting. Meneses and Palacio (2005) found that people reflect municipal trends. Xevgenos et al. (2015) provided a com- 2005; Thøgersen, 1997; Tucker and Speirs, 2003). Institutional cussions before changing the collection system. from 46 to 60 years old with a primary education are the group of parison that is solely based on information collected by other lit- factors, both formal (laws, collection frequencies, who is respon- The second largest subcategory (19 articles) was intrinsic individuals that are most reluctant towards recycling. Woodard erature studies, which increases the difficulty of making sible for the quality of sorting, etc.) and informal (e.g. household factors that affect recycling behaviour. The most frequent con- et al. (2001) found that elderly sorted more paper packaging comparisons between these studies. routines and norms), were found to affect recycling attitudes and clusion in this subcategory is that attitudes towards recycling waste than younger individuals, and Yuan and Yabe (2014) stated The key aspects of the collection systems described in these behaviour (Åberg et al., 1996; Gellynck et al., 2011; Refsgaard predict the degree of engagement in the recycling system that the elderly were the only group interested in economic articles were: the waste fractions collected (e.g. mixed recycla- and Magnussen, 2009). Some studies found that having free bags (Jesson, 2009; Martin et al., 2006; Meneses and Palacio, 2005; incentives. bles, biodegradable, packaging waste), type of collection point and bins for collecting biodegradable waste increased sorting Oskamp et al., 1991; Tucker and Speirs, 2003). Apart from this, Some articles agree that social pressure can be used for engag- (e.g. door-to-door, curbside, bring systems) and aspects of the participation for this fraction (Suttibak and Nitivattananon, 2008; there was little consensus between the conclusions. For exam- ing users in waste sorting activities (Do Valle et al., 2005; Gamba system administration (e.g. collection frequency, school gar- Yuan and Yabe, 2014). Miafodzyeva and Brandt (2013) did a lit- ple, some articles claim that environmental concern encourages and Oskamp, 1994; Jackson et al., 1993; Meneses and Palacio, bage banks, user or provider inclusivity). Dahlén and Lagerkvist erature review of variables that affect recycling behaviour and people to recycle (Guerin et al., 2001; Meneses and Palacio, 2005; Oskamp et al., 1991; Thøgersen, 1994). However, it may be (2010a), Wilson et al. (2012), Xevgenos et al. (2015), and concluded that ‘convenience is highly important in devising new 2005; Miafodzyeva and Brandt, 2013), while others state that it a hinderance when there are conflicts within households regarding Suttibak and Nitivattananon (2008) also evaluated what sort of technological changes in source separation and collection has no direct effect on recycling behaviour (Derksen and responsibilities (Åberg et al., 1996). Other social activities include information the collection system used. All articles evaluated schemes for household waste, which must be affordable and Gartrell, 1993; Oskamp et al., 1991) or that environmental con- managing a cooperative organisation, such as a ‘garbage bank’, the performance of the collection systems by comparing the widely accepted by the public’. cern cannot be predicted based on users motivations for or and involving different actors to increase social pressure that sup- quantity of sorted material. Gallardo Izquierdo et al. (2010), Another aspect discussed concerning recycling behaviour was against recycling (Gamba and Oskamp, 1994). ports recycling (Suttibak and Nitivattananon, 2008). Dahlén et al. (2007) and Dahlén and Lagerkvist (2010a) used information and knowledge (16 articles). Authors agree that hav- Economic issues were also discussed. Economic incentives User expectations do not always match the reality of waste complementary data to assess the quality of the sorted material ing knowledge about the recycling system is crucial for initiating appear to have an immediate and dramatic effect on sorting at the sorting. For example, the collection system may have categories (e.g. the purity of the sorted fractions and what proportion of a Rousta et al. 1103 source (Iyer and Kashyap, 2007), but that these incentives do not that do not match the waste generated by the users, or the collec- have a sustained effect (Thøgersen, 1994). The cost of the WM tion systems may differ from one place to another, causing uncer- service for households should be at most 1%–2% of the house- tainty among the users (Henriksson et al., 2010). Åberg et al. hold income (Wilson et al., 2012) for the service to be economi- (1996) reported that the problems perceived by users before com- cally sustainable. Gellynck et al. (2011) argued that the cost of posting were not the same as those that were most decisive in collecting mixed waste is important for minimising this fraction, practice. Tucker and Speirs (2003) stated that first impressions whereas the study done by Åberg et al. (1996) indicated that a remain for a long time unless the user has sustained different reduced collection fee is less important than environmental con- experiences. Interaction between users and competent and skilled cern for changing behaviour. Pay-as-you-throw (PAYT) systems, instructors educated in WM improve recycling (Suttibak and which also provide economic incentives for sorting, are discussed Nitivattananon, 2008). later on in the section ‘Types of collection systems’. Alternative options for getting rid of materials were found to Even though socio-demographic factors are considered to affect recycling behaviour (Dahlén and Lagerkvist, 2010a; play an important role in recycling behaviour (Ando and Gosselin, McDonald and Oates, 2003). These options could be charity col- 2005), their affect remains unclear. Iyer and Kashyap (2007) lections, local bring facilities, shared sorting bins with neigh- argue that single demographic measures are weak predictors of bours or reused material for other purposes. recycling behaviour, and propose a combined socio-demographic In summary, there is agreement that the factors that affect indicator. They combined income, education and occupation of recycling behaviour the most are appropriate physical infrastruc- both parents to make an indicator for social class, with which ture, shorter distance to collection points, reliable service, user they conclude that people with lower social class recycle more. convenience and adequate information. Studies of demographic The combined indicator proposed in that study is different to the factors, such as income, gender and education, do not yield con- single indicators from other studies, making it difficult to consistent results. clude if results from different studies support or contradict each other. Gender was found to influence recycling behaviour (Iyer Types of collection systems. Even though most of the reviewed and Kashyap, 2007), with Meneses and Palacio (2005) conclud- articles describe some type of collection system, six articles com- ing that women recycle more than men. Single family dwellings pared different types of collection systems (Dahlén and were reported to recycle more than multi-family dwellings (Ando Lagerkvist, 2010a; Dahlén et al., 2007; Gallardo Izquierdo et al., and Gosselin, 2005; Oskamp et al., 1991). Dahlén and Lagerkvist 2010; Suttibak and Nitivattananon, 2008; Wilson et al., 2012; (2010a) concluded from their literature review that waste genera- Xevgenos et al., 2015). Of these, four compared collection sys- tion increases with increased household consumption, which tems (i.e. Gallardo Izquierdo et al. (2010) in Spain, Dahlén et al. implies that a larger household income may increase waste gen- (2007) and Dahlén and Lagerkvist (2010a) in Sweden, and Sut- eration. In contrast, another study concluded that high income tibak and Nitivattananon (2008)) focused only on collection of reduces waste generation (Gellynck et al., 2011) and increases recyclables in Thailand, while two compared the entire WM sys- recycling (Gamba and Oskamp, 1994). If high-income house- tem. Xevgenos et al. (2015) compare 19 WM systems in different holds recycle more because of buying more goods or if they are countries and that are considered to be best practices, while Wil- better in waste separation remains unclear. Dahlén and Lagerkvist son et al. (2012) compared 20 WM systems that were chosen in (2010a) stated that education affects waste sorting, whereas order to provide a good global representation. All articles except Meneses and Palacio (2005) argued that it does not lead to Xevgenos et al. (2015) used data from local authorities, which improved sorting. Meneses and Palacio (2005) found that people reflect municipal trends. Xevgenos et al. (2015) provided a com- from 46 to 60 years old with a primary education are the group of parison that is solely based on information collected by other lit- individuals that are most reluctant towards recycling. Woodard erature studies, which increases the difficulty of making et al. (2001) found that elderly sorted more paper packaging comparisons between these studies. waste than younger individuals, and Yuan and Yabe (2014) stated The key aspects of the collection systems described in these that the elderly were the only group interested in economic articles were: the waste fractions collected (e.g. mixed recycla- incentives. bles, biodegradable, packaging waste), type of collection point Some articles agree that social pressure can be used for engag- (e.g. door-to-door, curbside, bring systems) and aspects of the ing users in waste sorting activities (Do Valle et al., 2005; Gamba system administration (e.g. collection frequency, school gar- and Oskamp, 1994; Jackson et al., 1993; Meneses and Palacio, bage banks, user or provider inclusivity). Dahlén and Lagerkvist 2005; Oskamp et al., 1991; Thøgersen, 1994). However, it may be (2010a), Wilson et al. (2012), Xevgenos et al. (2015), and a hinderance when there are conflicts within households regarding Suttibak and Nitivattananon (2008) also evaluated what sort of responsibilities (Åberg et al., 1996). Other social activities include information the collection system used. All articles evaluated managing a cooperative organisation, such as a ‘garbage bank’, the performance of the collection systems by comparing the and involving different actors to increase social pressure that sup- quantity of sorted material. Gallardo Izquierdo et al. (2010), ports recycling (Suttibak and Nitivattananon, 2008). Dahlén et al. (2007) and Dahlén and Lagerkvist (2010a) used User expectations do not always match the reality of waste complementary data to assess the quality of the sorted material sorting. For example, the collection system may have categories (e.g. the purity of the sorted fractions and what proportion of a 1104 Waste Management & Research 35(11) Rousta et al. 1105 specific fraction is sorted). Even though all articles measured context in which the comparison was done was different. was discussed here separately because many of the reviewed and landfilling having a small influence on the LCA results (Song the quantity of the sorted material, they did so in different ways, Although some of the indicators relate to similar aspects (e.g. articles focused on it. et al., 2013). Material recovery is better from an environmental making the results difficult to compare. Other topics that were source-sorting ratio and percentage of recycled material; partici- perspective than incineration with energy recovery (Bernstad evaluated were economical aspects of the collection systems pation rate and percentage of households that use and pay for Strategies that support recycling. Some articles reported et al., 2011), while door-to-door collection is better than kerbside (Suttibak and Nitivattananon, 2008; Wilson et al., 2012; waste collection), the way in which they are measured differs, strategies that support material recycling but that are not related collection or drop-off centres (Yıldız-Geyhan et al., 2016). Xevgenos et al., 2015) and participation rate or coverage of the hence making them difficult to compare. to waste collection. Material recovery facilities (MRFs) that system (Suttibak and Nitivattananon, 2008; Wilson et al., 2012). The collection systems for recyclable material compared in complement sorting at the source may be crucial to increase Recommendations. The most recurrent recommendation that Dahlén and Lagerkvist (2010b) stated that it is important to Suttibak and Nitivattananon (2008) are examples of a combi- material recycling. Cimpan et al. (2015), based on a review of was observed in the present review was to improve the informa- measure participation rate, but could not report this in their nation of decentralised collection points and large centralised existing MRFs, propose that MRFs play a prominent role in the tion provided about recycling (Do Valle et al., 2005; Martin et al., study because of data limitations and quality. recovery and composting facilities. The collection points, separation of municipal solid waste; they can be complemen- 2006; Refsgaard and Magnussen, 2009; Salhofer and Isaac, 2002; Dahlén et al. (2007) and Woodward et al. (2001) found that called garbage banks (GBs), are organised either by communi- tary to source separation and are best suited for urban areas. Wilson et al., 2012), followed by improving sorting infrastruc- the separate collection of biodegradable waste appears to increase ties (CGB) or in schools (SGBs). These GBs increase recycling Flexibility in the operation of these facilities is key to providing ture to facilitate recycling (Do Valle et al., 2005; Miafodzyeva sorting of recyclables. PAYT systems were reported to reduce rates and complement the sorting that local authorities organ- a good service for the WM system (Cimpan et al., 2015; and Brandt, 2013; Refsgaard and Magnussen, 2009; Salhofer and mixed waste generation by 20% (Dahlén et al., 2009; Dahlén and ise. People bring recyclable material to the GB and get paid for Griffiths et al., 2010). However, robust data is required to Isaac, 2002). It is interesting to note that improving information Lagerkvist, 2010b) and 50% (Dahlén et al., 2007). However, the collected material, providing an economic incentive for dimension these facilities properly (Rajamanikam et al., 2014). was recommended more often than improving infrastructure, since the PAYT systems did not increase the amount of waste that participation. In their evaluation, Suttibak and Nitivattananon Louis and Shih (2007) proposed that storing secondary materi- despite the fact that results about convenient infrastructure was correctly sorted in any of the studies, it was unclear how the (2008) observed that SGBs had a higher participation rate than als could provide increased revenues from material recovery if seemed to be more conclusive than the results regarding informa- reduction in the mixed waste fraction was achieved. Xevgenos the community-based ones, but achieved slightly lower recy- material prices fluctuate or increase steadily. The cost effi- tion. The literature also clarified the need to include social aspects et al. (2015) concluded that a system that combines PAYT with cling rates. The CGBs with the highest performance offered ciency of this storage depends on the needs for transportation when studying recycling (Martin et al., 2006; Refsgaard and kerbside collection of several fractions results in high quality col- door-to-door collection. Another success factor for CGBs was and the sorting capacity (Louis and Shih, 2007). Dahlén et al. Magnussen, 2009; Wilson et al., 2012). Dong et al. (2013) sug- lection. This supports the conclusion that PAYT systems affect allowing its members to apply for loans. The perception of the (2009) stated that recycling centres receive a large part of the gest that sorting at the source should be increased as much as sorting, if they are used in a proper way. Xevgenos et al. (2015) administrator’s awareness of WM problems, as well as low household waste, which they expect will increase. possible, to obtain the largest possible environmental benefits also state that a combination of kerbside and drop-off collection investment costs, was considered by Suttibak and Regarding economic aspects, Xevgenos et al. (2015) argue from resource recovery. To increase separation, Xu et al. (2015) results in high recycling rates. Nitivattananon (2008) as two of the main factors influencing that landfill and incineration bans and taxes are powerful tools to tested an incentive-based source separation model, and recom- Both articles that did a global review (Wilson et al., 2012; the success of the GBs. support material recovery, but they have no direct effect on mend models that benefit all stakeholders compared with those Xevgenos et al., 2015) stress the fact that no one WM solution Similar to GBs, Xu et al. (2015) investigated the effects of an household behaviour. A life cycle assessment (LCA), which was that only benefit some actors. However, one should be cautious fits all geographic contexts. They highlight the variety of suc- incentive-based source separation model in China, where house- complemented with an economic evaluation, revealed that source with generic recommendations, since Xevgenos et al. (2015), cessful systems, and emphasize that a good understanding of the holds receive an economic incentive for separating biodegrada- separation of biodegradable waste and recyclables generated Salhofer and Isaac (2002) and Wilson et al. (2012) agree that local context is needed to achieve a functioning system. ble and recyclable material from other waste streams. Their study more profit than landfilling these fractions with gas collection there is no one solution that fits all for providing waste sorting Xevgenos et al. (2015) claim that smaller populations are easier showed an increase in resource recovery from 25% to 87%. This (Dong et al., 2013). Xevgenos et al. (2015) argued that producer and collection systems. This means that a good understanding of to engage to achieve high recycling rates. However, the study cost analysis showed that it is possible to compensate households responsibility is an effective tool to collect recyclable material, if the context is needed to determine how to develop a relevant lacks a detailed explanation of how the comparison between dif- while simultaneously creating economic gains for the WM sys- it is implemented properly. waste collection system. This, in turn, requires reliable and up-to- ferent cities was done. tem. Even though the entire system can yield a profit, centralised Iyer and Kashyap (2007) argued that interventions are vital to date data (Wilson et al., 2012). Eight studies analysed their data by defining waste indicators WM disposal companies have little interest in the proposed improve recycling. Even though the present review included (Dahlén and Lagerkvist, 2010a; Dahlén et al., 2009; González- changes and might even oppose them (Xu et al., 2015). only a few studies of direct interventions, one of these articles Urban and rural aspects. The articles reviewed in this study Torre and Adenso-Díaz, 2005; Guagnano et al., 1995; Guerin The article authored by Rajamanikam et al. (2014) is the only (Porter et al., 1995) was a literature study focused only on come from several locations around the world. There is no com- et al., 2001; Miafodzyeva and Brandt, 2013; Porter et al., 1995; one in this review that describes a bottom-up solution for waste behavioural interventions to increase recycling. That literature mon definition among these articles of what is considered as an Wilson et al., 2012). The fact that these studies used new indica- collection systems. A community-based non-Government organ- study presented studies from 27 other articles that together urban or rural environment, with many of the articles not even tors to evaluate the waste systems supports the hypothesis (pre- isation (NGO) was established to provide door-to-door collection described 31 experiments (Porter et al., 1995). The authors argue mentioning this context. For example, a location with 27,000 sented by Dahlén et al., 2009) that official waste data is services that were not provided by local authorities. They estab- that there were two main discussions regarding interventions: inhabitants in southern Norway is considered to be an urban insufficient for proper evaluation of different household WM lished well-accepted collection routines (87% participation, with Those that aim to motivate long-term changes in behaviour and area by Refsgaard and Magnussen (2009). Dahlén and systems. A study comparing waste systems from various Swedish 39% separation at the source) and provided a service that those designed to reduce the amount of waste. The authors also Lagerkvist (2010a) defined large cities as those with 50,000 to municipalities used the following indicators: Waste generation achieved 47% reduction of waste to landfill. However, they reported that all reviewed studies showed at least some success 200,000 inhabitants, while Gallardo Izquierdo et al. (2010), rate, ratios of specific materials in the residual waste, source sort- lacked the technical knowledge for proper handling of difficult in increasing recycling. However, they often lacked evidence for who focused on urban waste, surveyed municipalities that had ing ratio, ratio of miss-sorted materials and participation rate fractions (e.g. hazardous waste). In addition, the door-to-door long-term effects of the interventions. at least 50,000 inhabitants. Based on the articles reviewed, the (Dahlén and Lagerkvist, 2010a). A multidisciplinary and interna- collection was expensive and did not support waste minimisa- present mini-review considers locations with more than 50,000 tional study conducted by Wilson et al. (2012) compared waste tion. The NGO activities, which did not require a collection fee, Environmental impacts. Four articles performed LCA with dif- inhabitants to be urban areas. For completeness, smaller loca- systems from 20 cities using the following indicators: Percentage were therefore not economically sustainable. Nevertheless, ferent scenarios for waste collection and how the collected mate- tions are also reviewed to determine if they correspond to urban of collection and sweeping coverage, percentage of controlled Rajamanikam et al. (2014) suggest that similar NGO-based col- rials would subsequently be handled. Most results agree that the or rural settlements. disposal, percentage of recycled materials and organics, percent- lection services (with a better economic strategy) can be estab- environmental benefits increase with the recycling rates (Dong With this definition, the study by Tucker and Speirs (2003) is age of total households that use and pay for waste collection, lished to complement municipal collection when needed. et al., 2013; Song et al., 2013), especially when increasing recy- the only one that corresponds to a rural context. They present a degree of user inclusivity, degree of provider inclusivity and It is important to note that the type of collection system can cling of the metal and plastic fractions (Bernstad et al., 2011). case study focused entirely on household composting. In spite of degree of institutional coherence (Wilson et al., 2012). The indi- be categorised as a part of the physical infrastructure, discussed Source separation and incineration are the most influential fac- this, the study by Tucker and Speirs (2003) has similar conclu- cators used in these two studies differed from each other since the above separately as factors that affect recycling behaviour. It tors in the LCA, with the collection, transportation, composting sions to the studies that were performed in urban areas. Rousta et al. 1105 was discussed here separately because many of the reviewed and landfilling having a small influence on the LCA results (Song articles focused on it. et al., 2013). Material recovery is better from an environmental perspective than incineration with energy recovery (Bernstad Strategies that support recycling. Some articles reported et al., 2011), while door-to-door collection is better than kerbside strategies that support material recycling but that are not related collection or drop-off centres (Yıldız-Geyhan et al., 2016). to waste collection. Material recovery facilities (MRFs) that complement sorting at the source may be crucial to increase Recommendations. The most recurrent recommendation that material recycling. Cimpan et al. (2015), based on a review of was observed in the present review was to improve the informa- existing MRFs, propose that MRFs play a prominent role in the tion provided about recycling (Do Valle et al., 2005; Martin et al., separation of municipal solid waste; they can be complemen- 2006; Refsgaard and Magnussen, 2009; Salhofer and Isaac, 2002; tary to source separation and are best suited for urban areas. Wilson et al., 2012), followed by improving sorting infrastruc- Flexibility in the operation of these facilities is key to providing ture to facilitate recycling (Do Valle et al., 2005; Miafodzyeva a good service for the WM system (Cimpan et al., 2015; and Brandt, 2013; Refsgaard and Magnussen, 2009; Salhofer and Griffiths et al., 2010). However, robust data is required to Isaac, 2002). It is interesting to note that improving information dimension these facilities properly (Rajamanikam et al., 2014). was recommended more often than improving infrastructure, Louis and Shih (2007) proposed that storing secondary materi- despite the fact that results about convenient infrastructure als could provide increased revenues from material recovery if seemed to be more conclusive than the results regarding informa- material prices fluctuate or increase steadily. The cost effi- tion. The literature also clarified the need to include social aspects ciency of this storage depends on the needs for transportation when studying recycling (Martin et al., 2006; Refsgaard and and the sorting capacity (Louis and Shih, 2007). Dahlén et al. Magnussen, 2009; Wilson et al., 2012). Dong et al. (2013) sug- (2009) stated that recycling centres receive a large part of the gest that sorting at the source should be increased as much as household waste, which they expect will increase. possible, to obtain the largest possible environmental benefits Regarding economic aspects, Xevgenos et al. (2015) argue from resource recovery. To increase separation, Xu et al. (2015) that landfill and incineration bans and taxes are powerful tools to tested an incentive-based source separation model, and recom- support material recovery, but they have no direct effect on mend models that benefit all stakeholders compared with those household behaviour. A life cycle assessment (LCA), which was that only benefit some actors. However, one should be cautious complemented with an economic evaluation, revealed that source with generic recommendations, since Xevgenos et al. (2015), separation of biodegradable waste and recyclables generated Salhofer and Isaac (2002) and Wilson et al. (2012) agree that more profit than landfilling these fractions with gas collection there is no one solution that fits all for providing waste sorting (Dong et al., 2013). Xevgenos et al. (2015) argued that producer and collection systems. This means that a good understanding of responsibility is an effective tool to collect recyclable material, if the context is needed to determine how to develop a relevant it is implemented properly. waste collection system. This, in turn, requires reliable and up-to- Iyer and Kashyap (2007) argued that interventions are vital to date data (Wilson et al., 2012). improve recycling. Even though the present review included only a few studies of direct interventions, one of these articles Urban and rural aspects. The articles reviewed in this study (Porter et al., 1995) was a literature study focused only on come from several locations around the world. There is no com- behavioural interventions to increase recycling. That literature mon definition among these articles of what is considered as an study presented studies from 27 other articles that together urban or rural environment, with many of the articles not even described 31 experiments (Porter et al., 1995). The authors argue mentioning this context. For example, a location with 27,000 that there were two main discussions regarding interventions: inhabitants in southern Norway is considered to be an urban Those that aim to motivate long-term changes in behaviour and area by Refsgaard and Magnussen (2009). Dahlén and those designed to reduce the amount of waste. The authors also Lagerkvist (2010a) defined large cities as those with 50,000 to reported that all reviewed studies showed at least some success 200,000 inhabitants, while Gallardo Izquierdo et al. (2010), in increasing recycling. However, they often lacked evidence for who focused on urban waste, surveyed municipalities that had long-term effects of the interventions. at least 50,000 inhabitants. Based on the articles reviewed, the present mini-review considers locations with more than 50,000 Environmental impacts. Four articles performed LCA with dif- inhabitants to be urban areas. For completeness, smaller loca- ferent scenarios for waste collection and how the collected mate- tions are also reviewed to determine if they correspond to urban rials would subsequently be handled. Most results agree that the or rural settlements. environmental benefits increase with the recycling rates (Dong With this definition, the study by Tucker and Speirs (2003) is et al., 2013; Song et al., 2013), especially when increasing recy- the only one that corresponds to a rural context. They present a cling of the metal and plastic fractions (Bernstad et al., 2011). case study focused entirely on household composting. In spite of Source separation and incineration are the most influential fac- this, the study by Tucker and Speirs (2003) has similar conclu- tors in the LCA, with the collection, transportation, composting sions to the studies that were performed in urban areas. 1106 Waste Management & Research 35(11) Rousta et al. 1107

Of the 18 articles that describe both rural and urban environ- most frequent method used. Three of the studies, Guagnano et al. Table 1. Overview of methods used per article. ments, only five have found clear differences in the sorting out- (1995), Knussen and Yule (2008) and Knussen et al. (2004), used Articles Methods used comes. Specifically, Thøgersen (1994) found that urban questionnaires and performed a thorough statistical analysis of households participated more in recycling schemes than their the data to investigate different behavioural theories, using recy- Survey and Modelling Literature Qualitative Empirical data rural counterparts. Derksen and Gartrell (1993) support this find- cling behaviour as their case study. Six of the seven articles that questionnaire review approach analysis ing, stating that the increased opportunity for recycling in urban used a qualitative approach were from the field of social sciences. Åberg et al. (1996) X areas was a significant factor that supports recycling. In contrast, In contrast, 15 of 17 articles that used empirical data analysis or Ando and Gosselin (2005) X Miafodzyeva and Brandt (2013) reviewed results that reveal that modelling as the main method were from engineering fields. Bernstad et al. (2011) X increasing population density decreases recycling. Lavee (2007) Some studies used a combination of methods. Dahlén and reported that larger municipalities have economic advantages Lagerkvist (2010b) used surveys in addition to empirical data Cimpan et al. (2015) X when establishing recycling as a WM procedure, owing to a high analysis. Dahlén et al. (2009) used an empirical data analysis as Dahlén and Lagerkvist (2010a) X concentration of recoverable material. Xevgenos et al. (2015) the main method and combined it with a literature review. Dahlén and Lagerkvist (2010b) < X found that it is easier for smaller municipalities to generate Thøgersen (1994) had questionnaires as the main method but Dahlén et al. (2007) X awareness and active participation in recycling schemes among also included a qualitative approach, and Iyer and Kashyap Dahlén et al. (2009) < X the citizens. (2007) combined questionnaires with an empirical data analysis. The remaining articles that covered both urban and rural con- Three articles, Jesson (2009), Henriksson et al. (2010) and Martin Derksen and Gartell (1993) X texts did not analyse the differences between these types of loca- et al. (2006), combined a qualitative approach as the main method Do Valle et al. (2005) X tions. However, the data they presented indicate that there are no with the questionnaire and survey method, and Ramachandra and Dong et al. (2013) X relevant differences between urban and rural scenarios. For Bachamanda (2007) complemented a qualitative approach with Gallardo et al. (2010) X example, Dahlén et al. (2007) concluded that municipalities that an empirical data analysis. In total, nine articles used a combina- Gamba and Oskamp (1994) X support recycling with convenient sorting infrastructure obtained tion of methods. Most of the articles included a short literature Gellynck et al. (2011) X better waste sorting results, regardless of the population density review, which was not considered as a method used in combina- of these locations (i.e. whether they were rural or urban). tion with another method. González-Torre and Adenso-Díaz (2005) X Two articles that describe models for WM made assumptions It is evident from Table 1 that the majority of the articles from Griffiths et al. (2010) X about the differences between urban and rural environments. the social sciences used the questionnaire and survey approach, Guagnano et al. (1995) X Louis and Shih (2007) assumed that rural areas have lower labour and that only a few articles combined these methods with a quali- Guerin et al. (2001) X costs, while Gellynck et al. (2011) assumed that the degree of tative approach or an empirical data analysis. Similarly, several Henriksson et al. (2010) < X urbanisation increased the amount of the waste that is generated. studies from the engineering discipline used just one method, Neither of these articles compared the results between urban and such as modelling or empirical data analysis. Given the complex- Hopper and Nielsen (1991) X rural areas. ity of recycling behaviour and sorting systems, this single-method Iyer and Kashyap (2007) X < approach may be too narrow to fully understand problems that Jackson et al. (1993) X arise within the behavioural and technical aspects of the system. Jesson (2009) < X Methods used A combination of relevant methods, both from the social and Knussen and Yule (2008) X The review revealed that the articles used methods that can be engineering sciences, is required to collect data that is necessary Knussen et al. (2004) X categorised into the following groups. for improving the WM system. Multidisciplinary studies, where quantitative and qualitative methods are combined with empiri- Lavee (2007) X •• Questionnaire and survey: Such as personal survey, mail sur- cal data, would help to overcome the limitations of using only Louis and Shih (2007) X vey, telephone survey, national survey, post questionnaire, one method. Martin et al. (2006) < X structured interviews. Among the reviewed articles, five articles used literature McDonald and Oates (2003) X •• Modelling: Such as LCA, different WM models, social mod- review as their main method. Given that these articles present Meneses and Palacio (2005) X els and frameworks. condensed results from other articles, it was deemed relevant to •• Literature review: Structured literature review focused on a specify the topics addressed by these literature reviews, the num- Miafodzyeva and Brandt (2013) X specific topic or research question. ber of articles they reviewed and the years they cover. This is Oskamp et al. (1991) X •• Qualitative approach: Such as structured and semi-structured summarised in Table 2. The results of these reviews are presented Porter et al. (1995) X interviews, focus groups and observations. together with the rest of the results in section ‘Summary of Rajamanikam et al. (2014) < X •• Empirical data analysis: Such as data collection from munici- results’. Ramachandra and Bachamanda (2007) X < palities, waste composition studies, data collection from case Refsgaard and Magnussen (2009) X studies. Reflection about the findings in the Salholfer and Isaac (2002) X literature Table 1 presents a summary of the reviewed articles and the Schultz (1999) X methods used. The main methods are shown in Table 1 with ‘X’ Many social science studies analyse recycling behaviour by test- Song et al. (2013) X and the secondary method with ‘<’. A total of 21 articles, 18 from ing theories and hypotheses, using questionnaires for this purpose. Sundberg et al. (1994) X social sciences and three from engineering fields, used the ques- A few studies have also used qualitative approaches to understand tionnaires and surveys as their main method. This was also the this phenomenon. These types of methods usually analyse what (Continued) Rousta et al. 1107

Table 1. Overview of methods used per article.

Articles Methods used

Survey and Modelling Literature Qualitative Empirical data questionnaire review approach analysis Åberg et al. (1996) X Ando and Gosselin (2005) X Bernstad et al. (2011) X Cimpan et al. (2015) X Dahlén and Lagerkvist (2010a) X Dahlén and Lagerkvist (2010b) < X Dahlén et al. (2007) X Dahlén et al. (2009) < X Derksen and Gartell (1993) X Do Valle et al. (2005) X Dong et al. (2013) X Gallardo et al. (2010) X Gamba and Oskamp (1994) X Gellynck et al. (2011) X González-Torre and Adenso-Díaz (2005) X Griffiths et al. (2010) X Guagnano et al. (1995) X Guerin et al. (2001) X Henriksson et al. (2010) < X Hopper and Nielsen (1991) X Iyer and Kashyap (2007) X < Jackson et al. (1993) X Jesson (2009) < X Knussen and Yule (2008) X Knussen et al. (2004) X Lavee (2007) X Louis and Shih (2007) X Martin et al. (2006) < X McDonald and Oates (2003) X Meneses and Palacio (2005) X Miafodzyeva and Brandt (2013) X Oskamp et al. (1991) X Porter et al. (1995) X Rajamanikam et al. (2014) < X Ramachandra and Bachamanda (2007) X < Refsgaard and Magnussen (2009) X Salholfer and Isaac (2002) X Schultz (1999) X Song et al. (2013) X Sundberg et al. (1994) X (Continued) 1108 Waste Management & Research 35(11) Rousta et al. 1109

Table 1. (Continued) community involvement and not a single household endeavour •• Convenient infrastructure is crucial for increasing the collec- (Tsai, 2008). This is relevant to consider when designing a sort- tion of sorted waste materials, regardless of whether the Articles Methods used ing system that requires or enables contact among users belong- infrastructure is in a rural or urban environment. Reduced Survey and Modelling Literature Qualitative Empirical data ing to a social network. However, group sorting may be perceived distance to collection points has consistently had the effect of questionnaire review approach analysis by users as inconvenient, and existing systems enable sorting that increasing sorted volumes. Convenient infrastructure should Suttibak and Nitivattananon (2008) < < X can be done when it suites the individual. be affordable for the users to use and should be widely Tanskanen (2000) X In summary, the results presented in this review are diverse; accepted, providing a service that is at least as good as the they are based on different case studies, and focus on different service for mixed waste collection. Thøgersen (1994) X < aspects of waste sorting. This illustrates the complexity of the •• Information should be given recurrently in a way that engages Thøgersen (1997) X source separation system and its dependence on the local context. users. Engagement can be achieved through personal contact, Tucker and Speirs (2003) X However, results regarding physical infrastructure (i.e. conveni- two-sided communication channels and/or early inclusion of Wilson et al. (2012) X ent collection system) are consistent between several studies. The the users in dialogues about the waste sorting system. Woodard et al. (2001) < X degree of convenience is perceived as an essential factor affect- •• Economic incentives do not seem to be too relevant to sup- ing waste sorting behaviour. This consensus is applicable to dif- port waste sorting, as long as waste sorting and collection Xevgenos et al. (2015) X ferent contexts, as long as the degree of convenience is defined in efforts are economically sustainable and provide shared Xu et al. (2015) < < X the context of the system users. benefits. Yildiz-Geyhan et al. (2016) X Even though the reviewed articles can be grouped as belonging •• Socio-demographic factors are known to affect waste genera- Yuan and Yabe (2014) X to social or engineering sciences based on how they address the tion and therefore influence waste sorting. However, there is Total articles X: 21 7 4 7 12 subject, almost half of the articles address a combination of techni- little consensus on how these factors affect sorting. Therefore, cal and social aspects when describing waste sorting systems. This socio-demographic considerations may be useful only when <: 7 0 1 4 2 seems to be an improvement from the conclusion drawn in the lit- used within the specific contexts, making these factors highly Symbol description in the table: X for main method used, < for secondary method used, symbols with white background relate to articles in erature review by Miafodzyeva and Brandt (2013) that there was a case-specific. social science field, symbols with orange background relate to articles in engineering field, symbols with blue background relate to articles in lack of studies that combined socio-psychological and technical- •• Sorting behaviour is influenced by social pressure, so infra- multidisciplinary area. organisational aspects. They also state that since these aspects are structures that enable social contact might increase sorting if studied separately, it is difficult to understand the interrelations they are still perceived as convenient. Table 2. Summary of the topics addressed by the literature reviews. they may have to gain a better understanding of recycling behav- •• This review reveals that the differences between rural and Reference Topic addressed Reviewed Between iour. The present review still observes that there is a division urban locations are just one of many contextual aspects that Porter et al. (1995) Interventions targeting recycling behaviour 27 articles 1973–1995 between the research done from engineering and social sciences, need to be considered when designing waste sorting systems. Miafodzyeva and Brandt (2013) Variables that affect recycling behaviour 63 articles 1991–2010 but sees a slight increase in studies that report on both technical Dahlén and Lagerkvist (2010a) Factors affecting the output of source sorting 66 articles 1981–2008 and social aspects. However, reporting on both types of aspects Most of the articles reviewed were case studies, meaning that programmes does not necessarily mean that the relations between these aspects the results and conclusions are case-specific and context-depend- Thøgersen (1997) How the design of reverse distribution channels 34 articles 1957–1995 of waste sorting are understood. Only a few articles provide a ent. However, many studies conclude that convenience (adequate affects public participation in recycling deeper understanding of these relations, presenting a holistic view access to sorting facilities, good service, etc.) and knowledge and Cimpan et al. (2015) Central sorting and recovery of household More than 50 articles and 2000–2015 waste recyclable materials technical reports on waste sorting by using a combination of methods in their stud- information are the most relevant factors that encourage waste ies. Technical optimisation of sorting at the source can only go as sorting in households. This can be considered a general conclu- far as to offer a specific infrastructure. How that infrastructure is sion within the field, which needs to be further studied to deter- the users perceive, and not what they do. On the other hand, most rural environments offer. The present review reveals that the type perceived and how it influences users cannot be understood by mine how these factors can be quantified and fully understood. of the studies in the engineering fields focused on improving the of infrastructure that is required depends not only on whether the only evaluating the empirical data of the system. Understanding Convenient infrastructure for separation is an aspect of the source technical waste collection system by using empirical data analy- location is urban or rural, but that there are also large variations users’ motivation and behaviour for sorting is crucial to be able to separation system that corresponds to the users’ perceived ease of sis, not reflecting on the users’ opinions of these systems. within urban and rural locations. Therefore, analyses of the local design collection systems or interventions that may help increase interaction with the technical system. It is therefore impossible to The articles reviewed here presented data from different context are required when designing sorting infrastructure. users’ participation in the system. Even though users’ motivations design a convenient infrastructure without considering both tech- administration levels (i.e. neighbourhoods, municipalities, Articles that report research on social capital and its effect on have been studied by several of the reviewed articles, they yield nical and social factors. It is also important to understand how regions and countries) and almost all the reviewed studies have waste separation and sorting were not included in this review. little consensus, making these results difficult to include when these factors must be adapted to the situation (context) where been done in a specific geographical area and are based on case This is because, although social capital aspects (e.g. number of designing waste sorting infrastructure. Therefore, when designing they will be implemented. Convenience has different meanings studies. Despite differences, some common conclusions have citizens performing charitable work) may affect separation and sorting infrastructure, it might be beneficial to include the waste in different contexts and for different users. been found. The articles that compared waste systems in different sorting (Tsai, 2008; Videras et al., 2012), it is not directly related system users in the development process to provide the case-spe- Most reviewed articles that offered design guidelines for the locations highlight that there is no one solution to fit all waste to the technical system available for sorting and hence cannot be cific considerations needed to improve the systems for these users. waste system were based on optimisation of the technical aspects sorting needs, so solutions should be tailored for different con- used to identify an improvement of the system. Nonetheless, alone. In contrast, most of the work that focused on user involve- texts. This supports the findings by Miliute-Plepiene et al. (2016) some insights from this field are relevant. The fact that social ment did not consider developing the technical aspects of the sys- that compared two waste systems at different degrees of comple- cohesion influences recycling behaviour is supported by the Conclusions tem, limiting the studies to user behaviour. More work is needed tion, and showed that the importance of norms and values to research reported in some of the reviewed articles that call for This article reviewed research done on recycling behaviour and to link user involvement with the technical solutions that the motivate people to sort their waste is higher when the conveni- stakeholder participation or describe bottom-up solutions for waste sorting systems to find results that could guide the design waste system offers. Using participation tools to include the users ence of the waste sorting infrastructure is low. WM (Rajamanikam et al., 2014; Suttibak and Nitivattananon, of future waste sorting infrastructure. The main conclusions that in the development of their sorting infrastructure might be a strat- When authorities decide on the type of sorting infrastructure 2008; Wilson et al., 2012). The literature that relates social capi- can be drawn from this review are as follows. egy to help bridge this gap. to implement, they consider the different logistics that urban and tal to waste sorting states that recycling is a function of Rousta et al. 1109 community involvement and not a single household endeavour •• Convenient infrastructure is crucial for increasing the collec- (Tsai, 2008). This is relevant to consider when designing a sort- tion of sorted waste materials, regardless of whether the ing system that requires or enables contact among users belong- infrastructure is in a rural or urban environment. Reduced ing to a social network. However, group sorting may be perceived distance to collection points has consistently had the effect of by users as inconvenient, and existing systems enable sorting that increasing sorted volumes. Convenient infrastructure should can be done when it suites the individual. be affordable for the users to use and should be widely In summary, the results presented in this review are diverse; accepted, providing a service that is at least as good as the they are based on different case studies, and focus on different service for mixed waste collection. aspects of waste sorting. This illustrates the complexity of the •• Information should be given recurrently in a way that engages source separation system and its dependence on the local context. users. Engagement can be achieved through personal contact, However, results regarding physical infrastructure (i.e. conveni- two-sided communication channels and/or early inclusion of ent collection system) are consistent between several studies. The the users in dialogues about the waste sorting system. degree of convenience is perceived as an essential factor affect- •• Economic incentives do not seem to be too relevant to sup- ing waste sorting behaviour. This consensus is applicable to dif- port waste sorting, as long as waste sorting and collection ferent contexts, as long as the degree of convenience is defined in efforts are economically sustainable and provide shared the context of the system users. benefits. Even though the reviewed articles can be grouped as belonging •• Socio-demographic factors are known to affect waste genera- to social or engineering sciences based on how they address the tion and therefore influence waste sorting. However, there is subject, almost half of the articles address a combination of techni- little consensus on how these factors affect sorting. Therefore, cal and social aspects when describing waste sorting systems. This socio-demographic considerations may be useful only when seems to be an improvement from the conclusion drawn in the lit- used within the specific contexts, making these factors highly erature review by Miafodzyeva and Brandt (2013) that there was a case-specific. lack of studies that combined socio-psychological and technical- •• Sorting behaviour is influenced by social pressure, so infra- organisational aspects. They also state that since these aspects are structures that enable social contact might increase sorting if studied separately, it is difficult to understand the interrelations they are still perceived as convenient. they may have to gain a better understanding of recycling behav- •• This review reveals that the differences between rural and iour. The present review still observes that there is a division urban locations are just one of many contextual aspects that between the research done from engineering and social sciences, need to be considered when designing waste sorting systems. but sees a slight increase in studies that report on both technical and social aspects. However, reporting on both types of aspects Most of the articles reviewed were case studies, meaning that does not necessarily mean that the relations between these aspects the results and conclusions are case-specific and context-depend- of waste sorting are understood. Only a few articles provide a ent. However, many studies conclude that convenience (adequate deeper understanding of these relations, presenting a holistic view access to sorting facilities, good service, etc.) and knowledge and on waste sorting by using a combination of methods in their stud- information are the most relevant factors that encourage waste ies. Technical optimisation of sorting at the source can only go as sorting in households. This can be considered a general conclu- far as to offer a specific infrastructure. How that infrastructure is sion within the field, which needs to be further studied to deter- perceived and how it influences users cannot be understood by mine how these factors can be quantified and fully understood. only evaluating the empirical data of the system. Understanding Convenient infrastructure for separation is an aspect of the source users’ motivation and behaviour for sorting is crucial to be able to separation system that corresponds to the users’ perceived ease of design collection systems or interventions that may help increase interaction with the technical system. It is therefore impossible to users’ participation in the system. Even though users’ motivations design a convenient infrastructure without considering both tech- have been studied by several of the reviewed articles, they yield nical and social factors. It is also important to understand how little consensus, making these results difficult to include when these factors must be adapted to the situation (context) where designing waste sorting infrastructure. Therefore, when designing they will be implemented. Convenience has different meanings sorting infrastructure, it might be beneficial to include the waste in different contexts and for different users. system users in the development process to provide the case-spe- Most reviewed articles that offered design guidelines for the cific considerations needed to improve the systems for these users. waste system were based on optimisation of the technical aspects alone. In contrast, most of the work that focused on user involvement did not consider developing the technical aspects of the sys- Conclusions tem, limiting the studies to user behaviour. More work is needed This article reviewed research done on recycling behaviour and to link user involvement with the technical solutions that the waste sorting systems to find results that could guide the design waste system offers. Using participation tools to include the users of future waste sorting infrastructure. The main conclusions that in the development of their sorting infrastructure might be a strat- can be drawn from this review are as follows. egy to help bridge this gap. 1110 Waste Management & Research 35(11) Rousta et al. 1111

Future work in this field should be multidisciplinary, use com- Gellynck X, Jacobsen R and Verhelst P (2011) Identifying the key factors in Refsgaard K and Magnussen K (2009) Household behaviour and attitudes Thøgersen J (1994) A model of recycling behaviour, with evidence from plementary methods and focus on the interface between user increasing recycling and reducing residual household waste: A case study with respect to recycling food waste–experiences from focus groups. Danish source separation programmes. International Journal of Research of the Flemish region of Belgium. Journal of Environmental Management Journal of Environmental Management 90: 760–771. in Marketing 11: 145–163. involvement and the technical aspects of the waste separation 92: 2683–2690. Rousta K and Dahlén L (2015) Source separation of household waste; tech- Thøgersen J (1997) Facilitating recycling: Reverse-distribution channel design system if it expects to provide solutions that are better at engag- González-Torre PL and Adenso-Díaz B (2005) Influence of distance on the nology and social aspects. In: Taherzadeh MJ and Richards T (eds) for participation and support. Social Marketing Quarterly 4: 42–55. ing households in waste sorting. This, in the long term, may well motivation and frequency of household recycling. Waste Management Resource Recovery to Approach Zero Municipal Waste. Boca Raton: Tsai Th (2008) The impact of social capital on regional waste recycling. 25: 15–23. CRC Press, pp.61–77. Sustainable Development 16: 44–55. result in a user-centred resource management system capable of Griffiths AJ, Williams KP and Owen N (2010) Engineering value recovery Salhofer S and Isaac NA (2002) Importance of public relations in recycling Tucker P and Speirs D (2003) Attitudes and behavioural change in household supporting increasing resource recovery. from municipal solid waste. Proceedings of the Institution of Mechanical strategies: Principles and case studies. Environmental Management 30: waste management behaviours. Journal of Environmental Planning and Engineers, Part C: Journal of Mechanical Engineering Science 224: 68–76. Management 46: 289–307. Declaration of conflicting interest 559–570. Schultz PW (1999) Changing behavior with normative feedback interven- Videras J, Owen AL, Conover E, et al. (2012) The influence of social rela- Guagnano GA, Stern PC and Dietz T (1995) Influences on attitude-behavior tions: A field experiment on curbside recycling. Basic and Applied Social tionships on pro-environment behaviors. Journal of Environmental The authors declared no potential conflicts of interest with respect to relationships a natural experiment with curbside recycling. Environment Psychology 21: 25–36. Economics and Management 63: 35–50. the research, authorship, and/or publication of this article. and Behavior 27: 699–718. Song Q, Wang Z and Li J (2013) Environmental performance of municipal Wilson DC, Rodic L, Scheinberg A, et al. (2012) Comparative analysis of Guerin D, Crete J and Mercier J (2001) A multilevel analysis of the deter- solid waste strategies based on LCA method: A case study of Macau. solid waste management in 20 cities. Waste Management & Research Funding minants of recycling behavior in the European countries. Social Science Journal of Cleaner Production 57: 92–100. 30: 237–254. Research 30: 195–218. Sundberg J, Gipperth P and Wene CO (1994) A systems approach to Woodard R, Harder MK, Bench M, et al. (2001) Evaluating the perfor- The authors disclosed receipt of the following financial support for Hagen-Zanker J and Mallet R (2013) How to do a rigorous, evidence-focused municipal solid waste management: A pilot study of Goteborg. Waste mance of a fortnightly collection of household waste separated into the research, authorship, and/or publication of this article: The literature review in international development, A Guidance Note. London: Management & Research 12: 73–91. compostables, recyclates and refuse in the south of England. Resources, authors appreciate the support and funding from Borås Energi och Overseas Development Institute. Suttibak S and Nitivattananon V (2008) Assessment of factors influencing the Conservation and Recycling 31: 265–284. Miljö AB and Sparbanksstiftelsen Sjuhärad. Henriksson G, Åkesson L and Ewert S (2010) Uncertainty regarding waste performance of solid waste recycling programs. Resources, Conservation Xevgenos D, Papadaskalopoulou C, Panaretou V, et al. (2015) Success handling in everyday life. Sustainability 2: 2799–2813. and Recycling 53: 45–56. stories for recycling of MSW at municipal level: A review. Waste and References Hopper JR and Nielsen JM (1991) Recycling as altruistic behavior normative Swedish Waste Management Association (2015) Hushållsavfall i siffror- Biomass Valorization 6: 657–684. and behavioral strategies to expand participation in a community recy- kommun- och länsstatistik 2014 [Household waste statistics 2014]. Xu W, Zhou C, Lan Y, et al. (2015) An incentive-based source separation Åberg H, Dahlman S, Shanahan H, et al. (1996) Towards sound environmen- cling program. Environment and Behavior 23: 195–220. Malmö, Sweden (in Swedish). model for sustainable municipal solid waste management in China. Waste tal behaviour: Exploring household participation in waste management. Iyer ES and Kashyap RK (2007) Consumer recycling: Role of incen- Tanskanen J-H (2000) Strategic planning of municipal solid waste manage- Management & Research 33: 469–476. Journal of Consumer Policy 19: 45–67. tives, information, and social class. Journal of Consumer Behaviour ment. Resources, Conservation and Recycling 30: 111–133. Yıldız-Geyhan E, Yılan-Çiftçi G, Altun-Çiftçioğlu GA, et al. (2016) Ando AW and Gosselin AY (2005) Recycling in multifamily dwellings: 6: 32–47. Tchobanoglous G, Theisen H and Vigil SA (1993) Integrated Solid Waste Environmental analysis of different packaging waste collection systems Does convenience matter? Economic Inquiry 43: 426–438. Jackson AL, Olsen JE, Granzin KL, et al. (1993) An investigation of determi- Management: Engineering Principles and Management Issues. New for Istanbul–Turkey case study. Resources, Conservation and Recycling Bernstad AK, la Cour Jansen J and Aspegren H (2011) Life cycle assessment of a nants of recycling consumer behavior. Advances in Consumer Research York, McGraw-Hill. 107: 27–37. household solid waste source separation programme: A Swedish case study. 20: 481–487. Thaler RH, Sunstein CR and Balz JP (2014) Choice architecture. In: Shafir Yuan Y and Yabe M (2014) Residents’ preferences for household kitchen Waste Management & Research 29: 1027–1042. Jesson J (2009) Household waste recycling behavior: A market segmentation E (ed.) The Behavioral Foundations of Public Policy. Princeton, Oxford, waste source separation services in Beijing: A choice experiment Cimpan C, Maul A, Jansen M, et al. (2015) Central sorting and recovery model. Social Marketing Quarterly 15: 25–38. UK: Princeton University Press, pp.428–440. Available at: https://ssrn. approach. International Journal of Environmental Research and Public of MSW recyclable materials: A review of technological state-of-the- Knussen C and Yule F (2008) “I’m not in the habit of recycling” the role of com/abstract=2536504 or http://dx.doi.org/10.2139/ssrn.2536504 Health 12: 176–190. art, cases, practice and implications for materials recycling. Journal of habitual behavior in the disposal of household waste. Environment and Environmental Management 156: 181–199. Behavior 40: 683–702. Dahlén L, Åberg H, Lagerkvist A, et al. (2009) Inconsistent pathways of Knussen C, Yule F, MacKenzie J, et al. (2004) An analysis of intentions to household waste. Waste Management 29: 1798–1806. recycle household waste: The roles of past behaviour, perceived habit, Dahlén L and Lagerkvist A (2010a) Evaluation of recycling programmes in and perceived lack of facilities. Journal of Environmental Psychology household waste collection systems. Waste Management & Research 28: 24: 237–246. 577–586. Lavee D (2007) Is municipal solid waste recycling economically efficient? Dahlén L and Lagerkvist A (2010b) Pay as you throw: Strengths and weak- Environmental Management 40: 926–943. nesses of weight-based billing in household waste collection systems in Louis G and Shih J-S (2007) A flexible inventory model for municipal solid Sweden. Waste Management 30: 23–31. waste recycling. Socio-Economic Planning Sciences 41: 61–89. Dahlén L, Vukicevic S, Meijer J-E, et al. (2007) Comparison of differ- Martin M, Williams ID and Clark M (2006) Social, cultural and structural ent collection systems for sorted household waste in Sweden. Waste influences on household waste recycling: A case study. Resources, Management 27: 1298–1305. Conservation and Recycling 48: 357–395. Derksen L and Gartrell J (1993) The social context of recycling. American McDonald S and Oates C (2003) Reasons for non-participation in a kerb- Sociological Review 58: 434–442. side recycling scheme. Resources, Conservation and Recycling 39: Do Valle PO, Rebelo E, Reis E, et al. (2005) Combining behavioral theo- 369–385. ries to predict recycling involvement. Environment and Behavior 37: Meneses GD and Palacio AB (2005) Recycling behavior a multidimensional 364–396. approach. Environment and Behavior 37: 837–860. Dong J, Ni M, Chi Y, et al. (2013) Life cycle and economic assessment of Miafodzyeva S and Brandt N (2013) Recycling behaviour among household- source-separated MSW collection with regard to greenhouse gas emis- ers: Synthesizing determinants via a meta-analysis. Waste and Biomass sions: A case study in China. Environmental Science and Pollution Valorization 4: 221–235. Research 20: 5512–5524. Miliute-Plepiene J, Hage O, Plepys A, et al. (2016) What motivates house- European Commission (2014) Communication from the Commission to the holds recycling behaviour in recycling schemes of different maturity? European Parliament, the Council, the European Economic and Social Lessons from Lithuania and Sweden. Resources, Conservation and Committee and the Committee of the Regions Towards a Circular Recycling 113: 40–52. Economy: A zero waste programme for Europe. Oskamp S, Harrington MJ, Edwards TC, et al. (1991) Factors influencing European Commission (2015) Closing the loop – An EU plan for the Circular household recycling behavior. Environment and Behavior 23: 494–519. Economy. Brussels. Porter BE, Leeming FC and Dwyer WO (1995) Solid waste recovery a Eurostat (2016) Municipal waste statistics. Available at: http://ec.europa. review of behavioral programs to increase recycling. Environment and eu/eurostat/statistics-explained/index.php/Municipal_waste_statistics Behavior 27: 122–152. (accessed 1 November 2016). Rajamanikam R, Poyyamoli G, Kumar S, et al. (2014) The role of non- Gallardo Izquierdo A, Bovea Edo MD, Colomer Mendoza FJ, et al. (2010) governmental organizations in residential solid waste management: A Comparison of different collection systems for sorted household waste in case study of Puducherry, a coastal city of India. Waste Management & Spain. Waste Management 30: 2430–2439. Research 32: 67–881. Gamba RJ and Oskamp S (1994) Factors influencing community residents’ Ramachandra TV and Bachamanda S (2007) Environmental audit of munici- participation in commingled curbside recycling programs. Environment pal solid waste management. International Journal of Environmental and Behavior 26: 587–612. Technology and Management 7: 369–391. Rousta et al. 1111

Refsgaard K and Magnussen K (2009) Household behaviour and attitudes Thøgersen J (1994) A model of recycling behaviour, with evidence from with respect to recycling food waste–experiences from focus groups. Danish source separation programmes. International Journal of Research Journal of Environmental Management 90: 760–771. in Marketing 11: 145–163. Rousta K and Dahlén L (2015) Source separation of household waste; tech- Thøgersen J (1997) Facilitating recycling: Reverse-distribution channel design nology and social aspects. In: Taherzadeh MJ and Richards T (eds) for participation and support. Social Marketing Quarterly 4: 42–55. Resource Recovery to Approach Zero Municipal Waste. Boca Raton: Tsai Th (2008) The impact of social capital on regional waste recycling. CRC Press, pp.61–77. Sustainable Development 16: 44–55. Salhofer S and Isaac NA (2002) Importance of public relations in recycling Tucker P and Speirs D (2003) Attitudes and behavioural change in household strategies: Principles and case studies. Environmental Management 30: waste management behaviours. Journal of Environmental Planning and 68–76. Management 46: 289–307. Schultz PW (1999) Changing behavior with normative feedback interven- Videras J, Owen AL, Conover E, et al. (2012) The influence of social relations: A field experiment on curbside recycling. Basic and Applied Social tionships on pro-environment behaviors. Journal of Environmental Psychology 21: 25–36. Economics and Management 63: 35–50. Song Q, Wang Z and Li J (2013) Environmental performance of municipal Wilson DC, Rodic L, Scheinberg A, et al. (2012) Comparative analysis of solid waste strategies based on LCA method: A case study of Macau. solid waste management in 20 cities. Waste Management & Research Journal of Cleaner Production 57: 92–100. 30: 237–254. Sundberg J, Gipperth P and Wene CO (1994) A systems approach to Woodard R, Harder MK, Bench M, et al. (2001) Evaluating the perfor- municipal solid waste management: A pilot study of Goteborg. Waste mance of a fortnightly collection of household waste separated into Management & Research 12: 73–91. compostables, recyclates and refuse in the south of England. Resources, Suttibak S and Nitivattananon V (2008) Assessment of factors influencing the Conservation and Recycling 31: 265–284. performance of solid waste recycling programs. Resources, Conservation Xevgenos D, Papadaskalopoulou C, Panaretou V, et al. (2015) Success and Recycling 53: 45–56. stories for recycling of MSW at municipal level: A review. Waste and Swedish Waste Management Association (2015) Hushållsavfall i siffror- Biomass Valorization 6: 657–684. kommun- och länsstatistik 2014 [Household waste statistics 2014]. Xu W, Zhou C, Lan Y, et al. (2015) An incentive-based source separation Malmö, Sweden (in Swedish). model for sustainable municipal solid waste management in China. Waste Tanskanen J-H (2000) Strategic planning of municipal solid waste manage- Management & Research 33: 469–476. ment. Resources, Conservation and Recycling 30: 111–133. Yıldız-Geyhan E, Yılan-Çiftçi G, Altun-Çiftçioğlu GA, et al. (2016) Tchobanoglous G, Theisen H and Vigil SA (1993) Integrated Solid Waste Environmental analysis of different packaging waste collection systems Management: Engineering Principles and Management Issues. New for Istanbul–Turkey case study. Resources, Conservation and Recycling York, McGraw-Hill. 107: 27–37. Thaler RH, Sunstein CR and Balz JP (2014) Choice architecture. In: Shafir Yuan Y and Yabe M (2014) Residents’ preferences for household kitchen E (ed.) The Behavioral Foundations of Public Policy. Princeton, Oxford, waste source separation services in Beijing: A choice experiment UK: Princeton University Press, pp.428–440. Available at: https://ssrn. approach. International Journal of Environmental Research and Public com/abstract=2536504 or http://dx.doi.org/10.2139/ssrn.2536504 Health 12: 176–190.

Paper II

I Rousta, K., Ekström, K.M., (2013) Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City PAPER I Sustainability 5(10), 4349-4361

Paper II

I Rousta, K., Ekström, K.M., (2013) Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City PAPER I Sustainability 5(10), 4349-4361 Sustainability 2013, 5, 4349-4361; doi:10.3390/su5104349 OPEN ACCESS sustainability ISSN 2071-1050 www.mdpi.com/journal/sustainability Article Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City

Kamran Rousta 1,* and Karin M. Ekström 2

1 School of Engineering, University of Borås, Allégatan 1, Borås 50190, Sweden 2 School of Business and IT, University of Borås, Allégatan 1, Borås 50190, Sweden; E-Mail: [email protected]

* Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: +46-33-435-46-44; Fax: +46-33-435-40-08.

Received: 24 June 2013; in revised form: 25 September 2013 / Accepted: 26 September 2013 / Published: 2 October 2013

Abstract: Source separation is a common method for dealing with the increasing problem of Municipal Solid Waste (MSW) in society. The citizens are then responsible for separating waste fractions produced in their home. If the consumers fail to sort the waste according to the source separation scheme, it will lead to an ineffective system. The purpose of this paper is to analyze the environmental, economic and social aspects of incorrect waste sorting in a medium sized Swedish city that has established a source separation system. In order to determine the extent to which citizens correctly sort their waste, food waste (black bags) and combustible fraction (white bags), were collected randomly from a residential area and categorized in different waste fractions. The results show that approximately 68 wt% of the waste in the white and 29 wt% in the black bags were not sorted correctly. This incorrect sorting accrues over 13 million SEK per year cost for this community. In order to improve the inhabitants’ participation in the waste management system, it is necessary to change different factors such as convenience and easy access to the recycling stations in the local MSW management systems as well as to review current regulation and policy.

Keywords: solid waste management; source separation; Sweden; recycling behavior

Sustainability 2013, 5, 4349-4361; doi:10.3390/su5104349 OPEN ACCESS sustainability ISSN 2071-1050 www.mdpi.com/journal/sustainability Article Assessing Incorrect Household Waste Sorting in a Medium-Sized Swedish City

Kamran Rousta 1,* and Karin M. Ekström 2

1 School of Engineering, University of Borås, Allégatan 1, Borås 50190, Sweden 2 School of Business and IT, University of Borås, Allégatan 1, Borås 50190, Sweden; E-Mail: [email protected]

* Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: +46-33-435-46-44; Fax: +46-33-435-40-08.

Received: 24 June 2013; in revised form: 25 September 2013 / Accepted: 26 September 2013 / Published: 2 October 2013

Keywords: solid waste management; source separation; Sweden; recycling behavior

Sustainability 2013, 5 4350 Sustainability 2013, 5 4351

1. Introduction introduced in Sweden, the recyclable material such as newspaper, plastic-, paper-, metal- and glass packaging must all be separated by the citizens and transferred to recycling stations. Based on this law, Problems with household waste have increased over time, in parallel to the development of Packaging and Newspaper Collection (FTI) is the service organization that is responsible for managing consumer society. As consumption increases, waste is generally also increasing in quantity and variety. the collection and recycling the packaging. This organization is financed by packaging fees and the This means depletion of the resources and consequently, environmental problems such as global selling of the collected material [15]. For glass packaging, “SvenskGlasÅtervinning” is the company warming if there is no sustainable way to manage this issue. These environmental problems have which collects and recycles the glass packaging [16]. According to statistics from FTI, 92% glass, 74% significant social and economic impacts [1]. For instance, if waste is just dumped into the land, i.e., paper packaging, 68% metal, 26% plastic packaging and 92% newspaper (in 2009) were recycled from landfilling, it causes pollutions in the air, water and soil as well as health hazards [2]. Besides the Swedish households waste in 2011[17]. environmental impacts and high cost of operating the landfill, establishing a new landfill is not an easy Based on the different sorting activities in the Municipal Solid Waste Management (MSWM) task [3]. That is why the European Union (EU) has developed a waste strategy that advocates the system of Borås, Figure 1 illustrates the three streams of household waste, which are collected and Thematic Strategy on the Prevention and Recycling of Waste [4]. According to this strategy, a waste transferred to different places in this system. management hierarchy has been formed [5]. This implies that waste should first of all be prevented, then reused and thirdly, recycled. If there is no way to apply the first three steps, waste should be used Figure 1. Different streams of waste collection in Borås [18]. as a source of energy. Finally, in the event that the last four steps do not result in any solution, waste should be dumped in a controlled landfill. Sweden is one of the foremost EU countries in applying the waste management hierarchy. During 2011, 4.3 million tons municipal solid waste were generated from households in Sweden, which implies about 460 kg per person per year [6]. Fifty-one per cent of the household waste generated was used for energy recovery, 33% for recycling, 15% for biological treatment, and less than one per cent to landfill [6]. In 2011, material recycling increased about five times more than the amount in 1975 [6]. Apart from the different regulation and technological development, sorting the waste at the source by citizens is the key for this success. Sorting waste decreases the solid waste management cost and improves the efficiency of recycling and the quality of recovered material [3]. A higher rate of source separation will increase the rate of recycling, which will then result in efficient and economical treatments for waste incineration and biogas production as well as reduction in environmental impacts as a result of the waste [7,8]. However, to succeed in separation of household waste, inhabitants have to manage and separate waste in their homes where it is generated. It is, therefore, important to have a consumer perspective when developing such a system. A consumer perspective requires an understanding of the consumers’ residential area in order to make it easier for them to sort and separate waste there. Also, uncertainty about which material is packaging Flow 1: Inhabitants sort the food waste in black bags and the combustible waste in white bags, or non-packaging, recyclable or non-recyclable, increases the probability of incorrect sorting of waste which are then collected and transferred to MR/TF for further processing by “Borås Energi och Miljö” from a consumer’s perspective [9]. It is also not efficient if the consumer has to travel long distances to (BEM), the municipality company in Borås. bring her/his waste to a recycling station. The system should therefore be as simple and user-friendly Flow 2: The packaging material needs to be sorted by inhabitants and collected at recycling as possible in order to motivate the consumers to be participants [10]. stations. FTI and “SvenskGlasÅtervinning” are responsible for managing the collection and recycling In order to understand the imperative role of the citizens in the Municipal Solid Waste (MSW), a case this type of waste. is presented in this paper namely the municipality of Borås with a population around 105,000 [11], Flow 3: The other type of waste, such as bulky waste like furniture, white goods such as located in the southwest of Sweden. Since 1991, inhabitants in Borås have been requested to sort their refrigerators and garden waste, hazardous waste, electronic waste, etc. needs to be transferred to waste at home and as a result less than 0.4% of the total households’ waste generation was dumped in recycling centers by inhabitants. BEM is responsible for these centers. All the collected waste in these the landfill in 2009 [12]. The residents sort the food waste and combustible waste in black and white centers is then transferred to industries for further processes. For instance, SAKAB is one of the bags, respectively. Thereafter, this waste is transferred to Material Recovery Transformation Facilities biggest companies in Sweden that takes care of the hazardous waste. They take out the toxic (MR/TF). Food waste in black bags is converted into biogas to fuel cars and public transportation, and substances from the hazardous waste and then treat them thermally to produce heat and electricity [19]. the combustible parts in white bags are prepared as Refuse Derived Fuel (RDF) for power plants to produce heat and electricity. Since 1994 when the Ordinance on Producer Responsibility [13,14] was Sustainability 2013, 5 4351 introduced in Sweden, the recyclable material such as newspaper, plastic-, paper-, metal- and glass packaging must all be separated by the citizens and transferred to recycling stations. Based on this law, Packaging and Newspaper Collection (FTI) is the service organization that is responsible for managing the collection and recycling the packaging. This organization is financed by packaging fees and the selling of the collected material [15]. For glass packaging, “SvenskGlasÅtervinning” is the company which collects and recycles the glass packaging [16]. According to statistics from FTI, 92% glass, 74% paper packaging, 68% metal, 26% plastic packaging and 92% newspaper (in 2009) were recycled from Swedish households waste in 2011[17]. Based on the different sorting activities in the Municipal Solid Waste Management (MSWM) system of Borås, Figure 1 illustrates the three streams of household waste, which are collected and transferred to different places in this system.

Figure 1. Different streams of waste collection in Borås [18].

Flow 1: Inhabitants sort the food waste in black bags and the combustible waste in white bags, which are then collected and transferred to MR/TF for further processing by “Borås Energi och Miljö” (BEM), the municipality company in Borås. Flow 2: The packaging material needs to be sorted by inhabitants and collected at recycling stations. FTI and “SvenskGlasÅtervinning” are responsible for managing the collection and recycling this type of waste. Flow 3: The other type of waste, such as bulky waste like furniture, white goods such as refrigerators and garden waste, hazardous waste, electronic waste, etc. needs to be transferred to recycling centers by inhabitants. BEM is responsible for these centers. All the collected waste in these centers is then transferred to industries for further processes. For instance, SAKAB is one of the biggest companies in Sweden that takes care of the hazardous waste. They take out the toxic substances from the hazardous waste and then treat them thermally to produce heat and electricity [19]. Sustainability 2013, 5 4352 Sustainability 2013, 5 4353

As indicated above, the Borås model for waste management relies heavily on the separation of the A pick analysis was applied in the pilot area in order to identify the total waste generation and waste at the source. The efficiency of the different processes in this system is based on the quality of waste characterization. In addition, the results of the pick analysis can present qualitative information the material, i.e., the separated waste, which feed into the system. For instance, for better biogas about the available waste management system [28]. The pick analysis was conducted based on production, it is necessary that the black bags include only the food waste and not any other material. instructions from “Avfall Sverige” (The Swedish Waste Management Association) [29]. These Also, if the packaging materials instead of being transferred to the recycling stations are sorted instructions are common practice for pick analysis conducted in Sweden. It was done in November incorrectly in the white bags, they will be burned rather than recycled which is not an optimal solution. 2011, as there are no special holidays and national days in this month affecting the results. Therefore, Since the model relies on separation at the source, the participation of the citizens in the source the results of the pick analysis are representative of waste generation in this pilot area, in general. As separation scheme is crucial in order for the system to succeed. the waste for the pilot area is collected once a week, wheeled bins were selected randomly over a The Municipal Solid Waste Management (MSWM) is a complicated and sophisticated system since four-week period, based on the Independent Random Sample (IRS), in order to get a more random it involves environmental, economic, and social aspects and their effects on each other [20]. It is result that did not vary according to a specific time. The random bins were transferred to a specific necessary to consider environmental aspects in order to reduce environmental damage. Furthermore, a place that had necessary instruments to measure the different fractions of the waste. The measurements system that is not economically justifiable will not be successful in the long run. Moreover, a system were conducted by one of the authors and two assistants. Because of the health and safety concerns, for waste management will not work if it is not accepted and used by the citizens; therefore, the social special clothes, masks and suitable gloves were worn. Weight measurement of the waste fractions in aspects are imperative. Even though there are some studies comparing features of different scenarios in the white and black bags was conducted by using an electronic wave balance (Tanita, max. 50 kg) with waste management processes such as landfilling, incineration and recycling from environmental and a measurement accuracy of 0.05 kg. The waste was categorized into: food, combustible (such as economic point of view [7,21–23], there is still a need to consider the social aspects in solid waste envelope, paper towels, diapers), packaging, hazardous (such as glue, painting materials, batteries and management. There are also studies that compare the different scenarios in collection systems and the lights) and other. For further investigation, the packaging was sorted into four categories of packaging: role of the source separation scheme in these scenarios [24–26]. However, there has been no specific plastic, paper, metal and glass and one category for newspaper. The total waste samples weight was study that quantifies the role of consumers in the waste source separation scheme. In a MSWM based 1732.8 kg, which amounted to 28% of the waste generation for this area in four weeks. The total on separation at the source, like in the Borås model, it is crucial that the inhabitants sort the fractions selected samples from this pilot area included twenty-eight bins. The total selected samples included correctly. It is therefore of interest to determine how important it is that the citizens do a correct job in 1332 white bags with a weight of 1070.7 kg and 675 black bags with a weight of 662.1 kg. First, each a source separation scheme. This paper is the first part of a larger study on the social aspects in waste bin was weighed and then the white and black bags were weighed separately. All the waste fractions management, analyzing the inhabitants’ participation in waste separation at the source as well as the mentioned above were weighed and documented. The process of sorting and measurement was also interaction of the social aspects with the other aspects in waste management. This paper specifically documented by memos and by taking photographs. After completing the measurement, a statistical examines waste sorting in the municipality of Borås. The purpose of the paper is to show the effect of analysis was conducted with data analyzed by MINITAB, and the mean value (in per cent) of five separation at the source in the waste management system by quantifying the role of the inhabitants in categories of waste in white and black bags, respectively, was determined. The data were used to such a system. This paper assesses the community cost when the waste is not sorted correctly in a identify the ratio of incorrect sorting, estimation of the characterization of waste generation and the source separation scheme. Based on the results, suggestions are made on how to improve the system. calculated cost for incorrect waste sorting for this pilot area. For some of the assessments, information from the MSWM in Borås as well as some information 2. Materials and Methods from the companies who are involved in collecting the packaging material was needed. Therefore, a structural interview was conducted as well as e-mail communications with a business developer who The study was conducted in the western part of Borås in a residential area that is called the “pilot works in the municipality company, “Borås Energi och Miljö” (BEM), which is responsible for waste area” in this paper. The pilot area includes 447 inhabitants who live in 208 apartments in nine management in Borås. Furthermore, phone enquiries were conducted with the CEO of “Pressretur” apartment blocks where each has eight floors. The tenants are renters and represent a variety of (company responsible for newspaper recycling) as well as a person responsible for the statistics at nationalities with different socio-economic backgrounds. According to Swedish Statistics, about 33% “SvenskGlasÅtervinning” (a company that collects and recycles the glass packaging in Sweden). of the residents in this area are native and the rest are immigrants [27]. About 63% of the residents are The main reason for interviews was to gather facts about the cost of different processes as well as more low income households and about 42% have no upper secondary education [27]. The majority of the information for clarifying the processes. inhabitants have lived in this area for more than five years [27]. The closest recycling station (flow 2 in Figure 1) is located at the back of the apartment blocks, where there are containers for newspapers and 3. Results glass packaging only. The closest recycling center (flow 3 in Figure 1) is located 2 kilometers from this residential area. In this section, the results are illustrated and the ratio of incorrect sorting, the characterization of the waste and the cost for incorrect waste sorting for the pilot area are determined. Sustainability 2013, 5 4353

A pick analysis was applied in the pilot area in order to identify the total waste generation and waste characterization. In addition, the results of the pick analysis can present qualitative information about the available waste management system [28]. The pick analysis was conducted based on instructions from “Avfall Sverige” (The Swedish Waste Management Association) [29]. These instructions are common practice for pick analysis conducted in Sweden. It was done in November 2011, as there are no special holidays and national days in this month affecting the results. Therefore, the results of the pick analysis are representative of waste generation in this pilot area, in general. As the waste for the pilot area is collected once a week, wheeled bins were selected randomly over a four-week period, based on the Independent Random Sample (IRS), in order to get a more random result that did not vary according to a specific time. The random bins were transferred to a specific place that had necessary instruments to measure the different fractions of the waste. The measurements were conducted by one of the authors and two assistants. Because of the health and safety concerns, special clothes, masks and suitable gloves were worn. Weight measurement of the waste fractions in the white and black bags was conducted by using an electronic wave balance (Tanita, max. 50 kg) with a measurement accuracy of 0.05 kg. The waste was categorized into: food, combustible (such as envelope, paper towels, diapers), packaging, hazardous (such as glue, painting materials, batteries and lights) and other. For further investigation, the packaging was sorted into four categories of packaging: plastic, paper, metal and glass and one category for newspaper. The total waste samples weight was 1732.8 kg, which amounted to 28% of the waste generation for this area in four weeks. The total selected samples from this pilot area included twenty-eight bins. The total selected samples included 1332 white bags with a weight of 1070.7 kg and 675 black bags with a weight of 662.1 kg. First, each bin was weighed and then the white and black bags were weighed separately. All the waste fractions mentioned above were weighed and documented. The process of sorting and measurement was also documented by memos and by taking photographs. After completing the measurement, a statistical analysis was conducted with data analyzed by MINITAB, and the mean value (in per cent) of five categories of waste in white and black bags, respectively, was determined. The data were used to identify the ratio of incorrect sorting, estimation of the characterization of waste generation and the calculated cost for incorrect waste sorting for this pilot area. For some of the assessments, information from the MSWM in Borås as well as some information from the companies who are involved in collecting the packaging material was needed. Therefore, a structural interview was conducted as well as e-mail communications with a business developer who works in the municipality company, “Borås Energi och Miljö” (BEM), which is responsible for waste management in Borås. Furthermore, phone enquiries were conducted with the CEO of “Pressretur” (company responsible for newspaper recycling) as well as a person responsible for the statistics at “SvenskGlasÅtervinning” (a company that collects and recycles the glass packaging in Sweden). The main reason for interviews was to gather facts about the cost of different processes as well as more information for clarifying the processes.

3. Results

In this section, the results are illustrated and the ratio of incorrect sorting, the characterization of the waste and the cost for incorrect waste sorting for the pilot area are determined. Sustainability 2013, 5 4354 Sustainability 2013, 5 4355

3.1. Ratio of Incorrect Sorting 3.2. Waste Characterization

The results of the statistical analysis of the measurement as well as the mean value of the proportion The statistical analysis by ANOVA (Analysis of Variance) shows that the difference between the of the five categories are shown in Figures 2 and 3 for white and black bags, respectively, with a 95% four weeks sampling periods is not significant. This means that there is no difference in waste confidence interval. The results presented in Figure 2 show that about 32 wt% of the waste in the white characterization between the weeks in this pilot area. Thus, based on the results of the pick analysis, bags (combustible) was sorted correctly. The majority of the faulty sorting in these bags consisted of the estimated average amount of different fractions of waste in white and black bags per household per packaging materials, around 43 wt%, which should have been sorted at recycling stations. About 22 wt% year in this residential area is presented in Table 1. As illustrated in this table, in this pilot area, on of the waste in the white bags was food waste, which should have been sorted in black bags. average 53.8 kg food waste, 38.3 kg newspaper, 26.5 kg paper packaging and 23.7 kg plastic packaging are the fractions that were sorted most incorrectly in the white bags by each household per Figure 2. Proportion of different waste categories in white bags intended for combustible waste. year. In the black bags, the fractions that were sorted most incorrectly are combustible waste with 26.9 kg/household/year. It can also be calculated that the total waste generation of each household is 50 ) 43.3 average about 386.7 kg/year which corresponds to an average of 7.4 kg of waste generation per week per % ( range s

g 40 household in this residential area. a b

t 32.0 i h

w Table 1. The content of white bags intended for combustible waste and black bags

30 n i

y 22.5 intended for food waste in kg/household/year for the pilot area with 447 inhabitants. r o

g 20 e t

a Type of the waste White bags Black bags c

e t

s Plastic packaging 23.7 5.4

a 10

W Paper packaging 26.5 3.6 0.5 1.7 0 Metal packaging 5.3 1.6 Food Combustible Packaging Hazardous Other Glass packaging 9.7 2.5 Newspaper 38.3 2.4 The results presented in Figure 3 show that about 71 wt% of the black bags was food waste that was Food waste 53.8 104.5 sorted correctly. Also, the black bags included approximately 18 wt% combustible material and 10 wt% Combustible 76.5 26.9 packaging that were sorted incorrectly. Although the amount of hazardous waste in both white and Hazardous 1.2 0.3 black bags that should have been transferred to recycling centers was less than 0.5 wt% of total waste Other 4.1 0.6 generation, this little amount needs to be reduced because of health hazards. In summary, 68 wt% of Total 238.9 147.8 the waste in white and 29 wt% in black bags were not sorted correctly. According to the results, the 3.3. Cost Calculation ratio of incorrect sorting in this area was determined to be 53 wt%. In other words, 53 wt% of the total waste in this area had not been sorted correctly. If all the waste is sorted as shown in Figure 1, it can be an ideal scenario for waste sorting in Borås. This means that all the food waste is sorted in black bags, combustible waste in white bags, all the Figure 3. Proportion of different waste categories in black bags intended for food waste. packaging is sorted at recycling stations and hazardous waste and other types of waste are transferred 80 to the recycling centers. As a result, the waste tariff paid by the inhabitants to the municipality would

) 70.7 average

% 70 be appropriately used. Also, FTI, which is partly financed by producers through packaging fees, would (

range s g

a 60 collect all the packaging material. If this were the case, there would be no extra cost for the b

c community. Here, the community is defined as all the people who are living in the city and pay taxes a

l 50 b

i and different tariffs to finance their social services including the waste management system. In this 40 y r

o case, the community cost means that the taxes and tariffs which are paid by the community are not g

e 30 t a

c used in the appropriate way and therefore causes a waste of resources. For instance, recyclable 18.2

e 20 t s

a 10.5 fractions of waste are burnt even though a waste tariff has been paid by the community for recycling.

W 10 0.2 0.4 This means that the resources which are provided by the community are not used in the right way; this 0 Food Combustible Packaging Hazardous Other is the definition of the community cost in this paper.

Sustainability 2013, 5 4355

3.2. Waste Characterization

The statistical analysis by ANOVA (Analysis of Variance) shows that the difference between the four weeks sampling periods is not significant. This means that there is no difference in waste characterization between the weeks in this pilot area. Thus, based on the results of the pick analysis, the estimated average amount of different fractions of waste in white and black bags per household per year in this residential area is presented in Table 1. As illustrated in this table, in this pilot area, on average 53.8 kg food waste, 38.3 kg newspaper, 26.5 kg paper packaging and 23.7 kg plastic packaging are the fractions that were sorted most incorrectly in the white bags by each household per year. In the black bags, the fractions that were sorted most incorrectly are combustible waste with 26.9 kg/household/year. It can also be calculated that the total waste generation of each household is about 386.7 kg/year which corresponds to an average of 7.4 kg of waste generation per week per household in this residential area.

Table 1. The content of white bags intended for combustible waste and black bags intended for food waste in kg/household/year for the pilot area with 447 inhabitants. Type of the waste White bags Black bags Plastic packaging 23.7 5.4 Paper packaging 26.5 3.6 Metal packaging 5.3 1.6 Glass packaging 9.7 2.5 Newspaper 38.3 2.4 Food waste 53.8 104.5 Combustible 76.5 26.9 Hazardous 1.2 0.3 Other 4.1 0.6 Total 238.9 147.8

3.3. Cost Calculation

If all the waste is sorted as shown in Figure 1, it can be an ideal scenario for waste sorting in Borås. This means that all the food waste is sorted in black bags, combustible waste in white bags, all the packaging is sorted at recycling stations and hazardous waste and other types of waste are transferred to the recycling centers. As a result, the waste tariff paid by the inhabitants to the municipality would be appropriately used. Also, FTI, which is partly financed by producers through packaging fees, would collect all the packaging material. If this were the case, there would be no extra cost for the community. Here, the community is defined as all the people who are living in the city and pay taxes and different tariffs to finance their social services including the waste management system. In this case, the community cost means that the taxes and tariffs which are paid by the community are not used in the appropriate way and therefore causes a waste of resources. For instance, recyclable fractions of waste are burnt even though a waste tariff has been paid by the community for recycling. This means that the resources which are provided by the community are not used in the right way; this is the definition of the community cost in this paper. Sustainability 2013, 5 4356 Sustainability 2013, 5 4357

According to this definition, the possible costs that are imposed on the community, based on the community cost. When the waste is not sorted correctly, the packaging waste will not be collected by incorrect waste sorting in terms of countable and uncountable costs are shown in Figure 4. FTI; instead they are collected in white and black bags by BEM. This means that FTI is paid by the packaging fee more than the amount that they actually collect when the waste is not sorted correctly. Figure 4. The countable and uncountable cost of the incorrect waste sorting for Borås community. However, FTI is not the cause of this cost. It is households’ incorrect waste sorting that causes the cost. This cost is identified by the “packaging fee” arrow in Figure 4. For newspapers, this cost is not specific because the paper industries cover the recycling costs. When they in turn sell the raw paper, they include this cost [31,32]. However, other kinds of packaging have fixed fees. Packaging such as plastic (1.71 SEK/kg), paper (1.08 SEK/kg) and metal (steel 2.84 SEK/kg and aluminium 2.32 SEK/kg) are the responsibility of FTI [33]. The fee for glass packaging is about 0.60 SEK/kg, which “SvenskGlasÅtervinning” is responsible for collecting and recycling [34]. The producers and importers have already paid for all the packaging materials to FTI with the packaging fees based on the above-mentioned fees. The total amount for the packaging fee calculated in Table 2 is about 26,000 SEK/year in this pilot area which is around 28% of the total calculated cost. This means that if these materials were sorted at the recycling stations, FTI would use this money to collect them. However, when they were incorrectly sorted in white and black bags, they were not available to be collected by FTI. Therefore, in this case, FTI has been paid even though they did not collect the packaging materials. Therefore, it is counted as a cost for the community. All the fractions that are not sorted correctly in both white and black bags incur a logistics cost, which is estimated to about 28,860 SEK/year. The incorrectly sorted materials in black bags require biological treatment costs (5,590 SEK/year), and the incorrectly sorted materials in the white bags require thermal The countable cost means the ones that are possible to count in currency. The reason for these costs treatment costs (15,210 SEK/year). Based on these assessments, the total countable costs for this area is the fact that the fractions are not sorted correctly in white and black bags. Some of these costs are are calculated to be 93,470 SEK/year (Table 2). Also, the cost for incorrect waste sorting can be the logistics cost from household to MR/TF (1.11 SEK/kg), thermal treatment (0.45 SEK/kg) and estimated to about 1.1 SEK per kg waste generation for this residential area. This was calculated by biological treatment (0.62 SEK/kg) that is covered by the BEM company [30]. For example, when a dividing the yearly cost (93,470 SEK) by total waste generation collected in white and black bags in metal packaging gets into black bags instead of being sorted at recycling stations, it will transfer to a year. MR/TF instead of being collected by FTI, this is identified by “logistics to MR/TF” in Figure 4. Also, Table 2. The countable costs of incorrect waste sorting for the pilot area with 447 inhabitants. when it is in the black bags, it is processed in a biological treatment, but it is not possible to produce biogas from a metal packaging. This means that the tariff which should finance the biological Cost SEK/year EUR/year treatment for the food waste is used wrongly for the fractions which are sorted incorrectly. In Figure 4, Logistics 28,860 3,255 Packaging fee 26,000 2,933 this is shown as the “biological treatment” arrow. Cost for extra bags 17,810 2,009 Also, the white and black bags that are used by the households to separate the combustible and food Preparation for RDF and thermal treatment 15,210 1,715 waste are distributed to the inhabitants free of charge. Incorrect sorting of the waste materials in these Biological treatment 5,590 631 bags therefore contribute to the usage of more bags [30]. The parts that are not sorted correctly occupy Total 93,470 10,543 the main volume of the white and black bags. This was also found in the observations during the pick analysis. Therefore, too many bags are used compared to if the households had sorted correctly. This In order to estimate this cost for the city of Borås, the results of a pick analysis that had been results in an extra cost, which can be estimated to about 17,810 SEK/year in this pilot area (Table 2). conducted by BEM in 2010 were analyzed. This pick analysis was from three residential areas in The other cost is the packaging fees for the part of the packaging that is sorted incorrectly in white Borås representing a variety of socio-economic background as well as apartments and houses [30]. and black bags. As mentioned before, the packaging fee is the one that producers and importers pay to The results of this pick analysis show that the incorrect sorting ratio was about 42 wt%, which can be FTI and to the glass recycling company in advance, in order to collect and recycle these materials. considered a value for the Borås city. By considering the average figures and waste characterization of In practice, the packaging fee is included the price of the goods that the consumers pay. The packaging the pilot study, and generalizing the calculated cost of this area to the Borås municipality, with about fee is therefore paid by the community and if it is not used appropriately, it will be counted as a 105,000 inhabitants, an estimated cost for incorrect sorting in Borås would be over 13 million SEK (1.46 million EUR) that is imposed on the Borås community yearly. Sustainability 2013, 5 4357 community cost. When the waste is not sorted correctly, the packaging waste will not be collected by FTI; instead they are collected in white and black bags by BEM. This means that FTI is paid by the packaging fee more than the amount that they actually collect when the waste is not sorted correctly. However, FTI is not the cause of this cost. It is households’ incorrect waste sorting that causes the cost. This cost is identified by the “packaging fee” arrow in Figure 4. For newspapers, this cost is not specific because the paper industries cover the recycling costs. When they in turn sell the raw paper, they include this cost [31,32]. However, other kinds of packaging have fixed fees. Packaging such as plastic (1.71 SEK/kg), paper (1.08 SEK/kg) and metal (steel 2.84 SEK/kg and aluminium 2.32 SEK/kg) are the responsibility of FTI [33]. The fee for glass packaging is about 0.60 SEK/kg, which “SvenskGlasÅtervinning” is responsible for collecting and recycling [34]. The producers and importers have already paid for all the packaging materials to FTI with the packaging fees based on the above-mentioned fees. The total amount for the packaging fee calculated in Table 2 is about 26,000 SEK/year in this pilot area which is around 28% of the total calculated cost. This means that if these materials were sorted at the recycling stations, FTI would use this money to collect them. However, when they were incorrectly sorted in white and black bags, they were not available to be collected by FTI. Therefore, in this case, FTI has been paid even though they did not collect the packaging materials. Therefore, it is counted as a cost for the community. All the fractions that are not sorted correctly in both white and black bags incur a logistics cost, which is estimated to about 28,860 SEK/year. The incorrectly sorted materials in black bags require biological treatment costs (5,590 SEK/year), and the incorrectly sorted materials in the white bags require thermal treatment costs (15,210 SEK/year). Based on these assessments, the total countable costs for this area are calculated to be 93,470 SEK/year (Table 2). Also, the cost for incorrect waste sorting can be estimated to about 1.1 SEK per kg waste generation for this residential area. This was calculated by dividing the yearly cost (93,470 SEK) by total waste generation collected in white and black bags in a year.

Table 2. The countable costs of incorrect waste sorting for the pilot area with 447 inhabitants. Cost SEK/year EUR/year Logistics 28,860 3,255 Packaging fee 26,000 2,933 Cost for extra bags 17,810 2,009 Preparation for RDF and thermal treatment 15,210 1,715 Biological treatment 5,590 631 Total 93,470 10,543

In order to estimate this cost for the city of Borås, the results of a pick analysis that had been conducted by BEM in 2010 were analyzed. This pick analysis was from three residential areas in Borås representing a variety of socio-economic background as well as apartments and houses [30]. The results of this pick analysis show that the incorrect sorting ratio was about 42 wt%, which can be considered a value for the Borås city. By considering the average figures and waste characterization of the pilot study, and generalizing the calculated cost of this area to the Borås municipality, with about 105,000 inhabitants, an estimated cost for incorrect sorting in Borås would be over 13 million SEK (1.46 million EUR) that is imposed on the Borås community yearly. Sustainability 2013, 5 4358 Sustainability 2013, 5 4359

As it is illustrated in Figure 4, there are some other costs for the incorrect sorting that remain 5. Conclusions uncountable. For example, the incorrect materials in black bags, which are not biodegradable and that Although waste separation at the source is known to be an effective method for sustainable waste are used for the biogas process also causes high maintenance cost if the machines malfunction [30]. management, it is based on the assumption that the inhabitants sort the waste correctly. The results in The internal logistics cost related to transfer of the huge amount of the incorrect materials in MR/TF is this study show that incorrect waste sorting can impose a significant cost on society. This cost for a yet another cost, which is not possible to calculate because there is no specific data in the system to medium-sized Swedish city like Borås, can be estimated to about 13 million SEK yearly. Therefore, it determine it [30]. Also, decisions about the policy and regulation in waste management are usually is necessary to investigate how to improve the available systems in order to make it more convenient made as part of a political process that consumes time and resources. However, when a system does for the consumers to participate in sorting the waste correctly. One way to do this could be to place not work properly, this process still imposes a cost on the society. If the waste is not sorted correctly, recycling stations closer to the residential area (property-close collection system). Furthermore, it is these above-mentioned uncountable costs are still costs for the community. not sufficient to merely inform, but necessary to educate people about the societal benefits of correct 4. Discussion waste sorting. In order to do this, different communication campaigns needs to be developed targeting different groups of consumers. These result-oriented efforts require effective cooperation between According to the results, the community cost for the city of Borås would be 13 million SEK for actors such as the municipality and FTI, involved in the MSWM. This requires also a review of current incorrect sorting per year. This is a significant amount of money, which can be used in more effective regulation and policy in this field. Besides this, further research needs to study consumer behavior and and better ways for the city. To get a perspective on how much money this is, it is important to note to investigate different factors which can influence inhabitants’ behavior in separation at the source that the identified budget for information and education of the inhabitants on how to sort the waste is and reducing the amount of waste. If the amount of waste is reduced, there will be less waste for the about 350,000 SEK/year [30] which is less than 3% of the calculated cost for the incorrect sorting. community to handle and these community costs will be saved. This means that it is worthwhile for Borås to re-think the existing MSWM system in order to make it more convenient and understandable for the inhabitants to participate in source separation. Even though Acknowledgments sorting all the waste fractions correctly may not be possible to achieve, it needs to be acknowledged The authors wish to acknowledge “Borås Energi and Miljö” for supporting and “Sjuhärads that a reduction of the incorrect sorting ratio saves the community’s resources. Kommunalförbund” for funding of this project. Included in this cost is the cost for the packaging fee, which is about 28% of the total cost. Although the reason for this cost is incorrect sorting by the inhabitants, there is a need to review the Conflicts of Interest policy and regulation in this field in order to achieve a more effective cooperation between FTI and the municipalities. In Borås, from the residents’ perspective, BEM is responsible for the entire waste The authors declare no conflict of interest. management system, but the collecting and recycling of the packaging are the responsibility for FTI. This raises the question regarding who is responsible for encouraging the residents to sort the waste References correctly. Even though both actors are putting a lot of efforts into this, the result of this study shows 1. Barr, S. Household Waste in Social Perspective; Ashgate: Aldershot, UK, 2002. that more effective and result-oriented actions are required in order to inform, educate and encourage 2. El-Fadel, M.; Findikakis, A.N.; Leckie, J.O. Environmental impacts of solid waste landfilling. the inhabitants to participate in the source separation schemes. J. Environ. Manage. 1997, 50, 1–25. Besides economic aspects, the incorrect sorting shown in this study is a hindrance for the MSWM 3. Tchobanoglous, G.; Theisen, H.; Vigil, S.A. Integrated Solid Waste Management: Engineering in Borås to move towards its sustainable strategy. The MSWM in Borås is based on the waste Principles and Management Issues; McGraw-Hill: New York, NY, USA, 1993; pp. 3–538. hierarchy top down; prevention, reuse, recycling, energy recovery, and landfilling. In the new waste 4. EU. On the Thematic Strategy on the Prevention and Recycling of Waste. In Proceedings of the management plan for 2020, it is stated that the main goal is to go towards the top of the waste Report from the Commission to the European Parliament, The Council, The European Economic hierarchy [12]; in other words, to prevent waste, reuse and recycling. In order to achieve this, the and Social Committee and the Committee of the Regions, Brussels, Belgium, 19 January 2011. inhabitants have to participate more actively in the waste separation schemes and to sort the waste 5. EU. On Waste and Repealing Certain Directives. In Proceedings of the Directive 2008/98/EC of correctly. As it is now, the incorrectly sorted materials in white and black bags are transferred to power the European Parliament and of the Council, Brussels, Belgium, 19 November 2008. plants as fuel to produce electricity and heat in an energy recovery process; however, this is not a 6. Swedish Waste Management. Statistics of Households’ Waste (Hushållsavfall i Siffror) (in Swedish); sustainable alternative. As mentioned above, based on the waste hierarchy, energy recovery is ranked Rapport U2012:18; Swedish Waste Management: Malmö, Sweden, 2012. lower compared to recycling. Therefore, this incorrect sorting actually prevents the achievement of the 7. Bernstad, A.; la Cour Jansen, J.; Aspegren, H. Life cycle assessment of a household solid waste environmental goals which are stated in the waste management plan in Borås for 2020. source separation programme: A Swedish case study. Waste Manag. Res. 2011, 29, 1027–1042.

Sustainability 2013, 5 4359

5. Conclusions

Although waste separation at the source is known to be an effective method for sustainable waste management, it is based on the assumption that the inhabitants sort the waste correctly. The results in this study show that incorrect waste sorting can impose a significant cost on society. This cost for a medium-sized Swedish city like Borås, can be estimated to about 13 million SEK yearly. Therefore, it is necessary to investigate how to improve the available systems in order to make it more convenient for the consumers to participate in sorting the waste correctly. One way to do this could be to place recycling stations closer to the residential area (property-close collection system). Furthermore, it is not sufficient to merely inform, but necessary to educate people about the societal benefits of correct waste sorting. In order to do this, different communication campaigns needs to be developed targeting different groups of consumers. These result-oriented efforts require effective cooperation between actors such as the municipality and FTI, involved in the MSWM. This requires also a review of current regulation and policy in this field. Besides this, further research needs to study consumer behavior and to investigate different factors which can influence inhabitants’ behavior in separation at the source and reducing the amount of waste. If the amount of waste is reduced, there will be less waste for the community to handle and these community costs will be saved.

Acknowledgments

The authors wish to acknowledge “Borås Energi and Miljö” for supporting and “Sjuhärads Kommunalförbund” for funding of this project.

Conflicts of Interest

The authors declare no conflict of interest.

References

1. Barr, S. Household Waste in Social Perspective; Ashgate: Aldershot, UK, 2002. 2. El-Fadel, M.; Findikakis, A.N.; Leckie, J.O. Environmental impacts of solid waste landfilling. J. Environ. Manage. 1997, 50, 1–25. 3. Tchobanoglous, G.; Theisen, H.; Vigil, S.A. Integrated Solid Waste Management: Engineering Principles and Management Issues; McGraw-Hill: New York, NY, USA, 1993; pp. 3–538. 4. EU. On the Thematic Strategy on the Prevention and Recycling of Waste. In Proceedings of the Report from the Commission to the European Parliament, The Council, The European Economic and Social Committee and the Committee of the Regions, Brussels, Belgium, 19 January 2011. 5. EU. On Waste and Repealing Certain Directives. In Proceedings of the Directive 2008/98/EC of the European Parliament and of the Council, Brussels, Belgium, 19 November 2008. 6. Swedish Waste Management. Statistics of Households’ Waste (Hushållsavfall i Siffror) (in Swedish); Rapport U2012:18; Swedish Waste Management: Malmö, Sweden, 2012. 7. Bernstad, A.; la Cour Jansen, J.; Aspegren, H. Life cycle assessment of a household solid waste source separation programme: A Swedish case study. Waste Manag. Res. 2011, 29, 1027–1042. Sustainability 2013, 5 4360 Sustainability 2013, 5 4361

8. Eriksson, O.; Carlsson Reich, M.; Frostell, B.; Björklund, A.; Assefa, G.; Sundqvist, J.O.; 27. Swedish Statistisc. Socio-Economic Statistics; SCB: Stockholm, Sweden, 2012. Granath, J.; Baky, A.; Thyselius, L. Municipal solid waste management from a systems 28. Berg, P.E. Source Sorting. Theory, Method and Implementation/Källsortering. Teori, Metod och perspective. J. Clean. Prod. 2005, 13, 241–252. Implementering/. Doctoral Thesis (in Swedish); Institutionen för Vattenförsörjnings-och 9. Henriksson, G.; Åkesson, L.; Ewert, S. Uncertainty regarding waste handling in everyday life. Avloppsteknik, Chalmers Tekniska Högskola: Göteborg, Sweden, 1993. Sustainability 2010, 2, 2799–2813. 29. Swedish Waste Management. Manual for Pick Analysis of Houshold Waste/Manual för 10. Ewert, S.; Greger, H.; Åkesson, L. Osäker eller Nöjd: Kulturella Aspekter på Vardagens Plockanalys av Hushållsavfall (in Swedish); RVF Utveckling 2005:19; Avfall Sverige: Malmö, Avfallspraktik (in Swedish); KTH Royal Institute of Technology: Stockholm, Sweden, 2008. Sweden, 2005. 11. Borås statistic. Available online: http://www.boras.se/forvaltningar/stadskansliet/stadskansliet/ 30. Hans Skoglund. Borås Energi och Miljö, Borås, Sweden. Personal communication, 2 March 2012. samhallsplanering/samhallsplanering/borasisiffror/statistik.4.82f0a312665003f0d800014558.html 31. Göran Nilsson. Pressretur, Stockholm, Sweden. Personal communication, 4 April 2013. (accessed on 24 June 2013). 32. Pressretur. About us. Available online: http://www.pressretur.se/omoss.4.328c401048bb 12. Borås Energi och Miljö. Borås Waste Management Plan 2012–2020; Borås Energi och Miljö: 6f3da80001245.html (accessed on 24 June 2013). Borås, Sweden, 2011. 33. REPA. Packaging fee. Available online: http://www.repa.se/avgifterregler/avgifter.4.3bd4d 13. SFS1994a, Förordningen om producentansvar för returpapper (The Ordinance of Producers’ bee12c726a177a8000215.html (accessed on 24 June 2013). Responsibility for Waste Paper). SFS 1994:1205, 1994. 34. Per Johansson. SvenskGlasÅtervinning, Hammar, Sweden. Personla communication, 4 April 14. SFS1994b, Förordningen om producentansvar för förpackningar (The Ordinance of Producers’ 2013. Responsibility for Packaging Materials). SFS 1994:1235, 1994. 15. Förpackning- och tidningsinsamling (FTI). About FTI. Available online: http://www.ftiab.se/forforetag/ © 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article inenglish.4.53781b2513d681fd9d51ff8.html (accessed on 24 June 2013), (in English). distributed under the terms and conditions of the Creative Commons Attribution license 16. SvenskGlasÅtervinning. Glass recycling. Available online: http://www.glasatervinning.se/ (http://creativecommons.org/licenses/by/3.0/). index.php/om-sga (accessed on 24 June 2013), (in Swedish). 17. Förpackning- och tidningsinsamling (FTI). Statistic. Available online: http://www.ftiab.se/180.html (accessed on 24 June 2013), (in Swedish). 18. Borås Energi och Miljö. Sorteringsguide (in Norwegian); Borås Energi och Miljö: Borås, Sweden, 2008. 19. SAKAB. Treatment the hazardous waste. Available online: http://www.sakab.se/en/ treatment-hazardous-waste (accessed on 24 June 2013). 20. Morrissey, A.J.; Browne, J. Waste management models and their application to sustainable waste management. Waste Manage. 2004, 24, 297–308. 21. Tsiliyannis, C.A. Report: Comparison of environmental impacts from solid waste treatment and disposal facilities. Waste Manage. Res. 1999, 17, 231–241. 22. Rabl, A.; Spadaro, J.V.; Zoughaib, A. Environmental impacts and costs of solid waste: A comparison of landfill and incineration. Waste Manage. Res. 2008, 26, 147–162. 23. Emery, A.; Davies, A.; Griffiths, A.; Williams, K. Environmental and economic modelling: A case study of municipal solid waste management scenarios in Wales. Resour. Conserv. Recy. 2007, 49, 244–263. 24. Dahlén, L. Household Waste Collection: Factors and Variations. Ph.D. Thesis, Luleå University of Technology, Luleå, Sweden, 2008. 25. Dahlén, L.; Vukicevic, S.; Meijer, J.-E.; Lagerkvist, A. Comparison of different collection systems for sorted household waste in Sweden. Waste Manage. 2007, 27, 1298–1305. 26. Larsen, A.W.; Merrild, H.; Møller, J.; Christensen, T.H. Waste collection systems for recyclables: An environmental and economic assessment for the municipality of Aarhus (Denmark). Waste Manage. 2010, 30, 744–754. Sustainability 2013, 5 4361

27. Swedish Statistisc. Socio-Economic Statistics; SCB: Stockholm, Sweden, 2012. 28. Berg, P.E. Source Sorting. Theory, Method and Implementation/Källsortering. Teori, Metod och Implementering/. Doctoral Thesis (in Swedish); Institutionen för Vattenförsörjnings-och Avloppsteknik, Chalmers Tekniska Högskola: Göteborg, Sweden, 1993. 29. Swedish Waste Management. Manual for Pick Analysis of Houshold Waste/Manual för Plockanalys av Hushållsavfall (in Swedish); RVF Utveckling 2005:19; Avfall Sverige: Malmö, Sweden, 2005. 30. Hans Skoglund. Borås Energi och Miljö, Borås, Sweden. Personal communication, 2 March 2012. 31. Göran Nilsson. Pressretur, Stockholm, Sweden. Personal communication, 4 April 2013. 32. Pressretur. About us. Available online: http://www.pressretur.se/omoss.4.328c401048bb 6f3da80001245.html (accessed on 24 June 2013). 33. REPA. Packaging fee. Available online: http://www.repa.se/avgifterregler/avgifter.4.3bd4d bee12c726a177a8000215.html (accessed on 24 June 2013). 34. Per Johansson. SvenskGlasÅtervinning, Hammar, Sweden. Personla communication, 4 April 2013.

© 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

Paper III

I I Rousta, K., Bolton, K., Lundin, M., Dahlén, L., (2015) Quantitative Assessment of Distance to Collection Point and Improved Sorting Information on Source Separation of Household Waste PAPER I PAPER Waste Management 40: 22-30

Paper III

Waste Management 40 (2015) 22–30

Contents lists available at ScienceDirect

Waste Management

journal homepage: www.elsevier.com/locate/wasman

Quantitative assessment of distance to collection point and improved sorting information on source separation of household waste

a, a a b Kamran Rousta ⇑, Kim Bolton , Magnus Lundin , Lisa Dahlén a Swedish Centre for Resource Recovery, University of Borås, 501 90 Borås, Sweden b Waste Science and Technology, Luleå University of Technology, 971 87 Luleå, Sweden article info abstract

Article history: The present study measures the participation of households in a source separation scheme and, in Received 3 November 2014 particular, if the household’s application of the scheme improved after two interventions: (a) shorter Accepted 1 March 2015 distance to the drop-off point and (b) easy access to correct sorting information. The effect of these inter- Available online 24 March 2015 ventions was quantified and, as far as possible, isolated from other factors that can influence the recycling behaviour. The study was based on households located in an urban residential area in Sweden, where Keywords: waste composition studies were performed before and after the interventions by manual sorting (pick Source separation analysis). Statistical analyses of the results indicated a significant decrease (28%) of packaging and news- Household waste print in the residual waste after establishing a property close collection system (intervention (a)), as well Recycling behaviour Information as significant decrease (70%) of the miss-sorted fraction in bags intended for food waste after new infor- Distance mation stickers were introduced (intervention (b)). Providing a property close collection system to collect more waste fractions as well as finding new communication channels for information about sorting can be used as tools to increase the source separation ratio. This contribution also highlights the need to evaluate the effects of different types of information and communication concerning sorting instructions in a property close collection system. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Source separation has become mainstream in Swedish waste management, increasing material recycling and decreasing land- Source separation of waste refers to the collection of different filling of household waste, since the ordinance of producer waste fractions where the waste is sorted at the place where it is responsibility was applied in 1994 (SFS, 1994a,b) and the ordi- generated and where the fractions are collected in separate con- nance about landfilling in 2001 (SFS, 2001). According to the tainers. Several different technical systems for source separation Swedish Waste Management Association (2014), 33% of household of household waste are available, all of which rely on active, effi- waste was recycled in 2013. This means that a substantial amount cient and correct participation at the household level (Baltes of recyclable materials was still not recycled, but recovered in et al., 2009; Leitol, 2014; Rada et al., 2013). For example, two meth- other ways such as incineration with energy recovery. The ods that are used in Sweden to collect separated waste are the Swedish Environmental Protection Agency has suggested some bring/drop-off system and the property close collection system. objectives to increase the recycling rate of household waste and In the bring/drop-off system inhabitants bring their sorted, dry biological treatment of food waste by 2020 (SEPA, 2013). recyclable materials to a drop-off station which may be located Increased collection rates of correctly sorted recyclables and a in another part of the city. In the property close collection system, reduced amount of miss-sorted materials in the residual waste is a bin-room, which contains different bins for collection of some required to achieve this goal. Therefore, identification and quan- recyclable waste, is provided in close proximity to multi-family tification of the factors that influence source separation of waste apartment buildings for collection of sorted, recyclable materials. (called influential factors) are important. These factors can subse- In some residential areas, single family houses have multi-com- quently be used as a tool for decision makers and engineers to partment bins for kerbside collection of sorted recyclables. improve waste management systems. Several researchers have investigated influential factors on inhabitants’ recycling behaviour. A synthesis of research results

Corresponding author. Tel.: +46 33 435 4644. identified consumer knowledge and commitment to recycling as ⇑ E-mail address: [email protected] (K. Rousta). intrinsic factors, economic rewards and social influence as extrinsic http://dx.doi.org/10.1016/j.wasman.2015.03.005 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved. K. Rousta et al. / Waste Management 40 (2015) 22–30 23 factors and frequency of the collection as an external facilitator aspects of the determinants, while the effects have not been quan- (Castagna et al., 2013; Grodzin´ ska-Jurczak et al., 2006; Hornik tified. That is, none of the studies quantifies the role of distance et al., 1995). Seacat and Northrup (2010) named recycling informa- and information, separately and in isolation from other factors that tion, motivation and behavioural skills as determinants for engag- can influence recycling. The overall aim of this paper is to con- ing in recycling. The influence of socio-demographic variation on tribute to increased and correct participation in source separation recycling is still under debate. In situations where recycling has programs for household waste. The specific objective is to assess become a habit and is widely spread throughout population, the and quantify the effect of: (1) shorter distance to a collection point demographic variable is weakly correlated to waste sorting beha- for packaging and (2) easy access to information about sorting of viour (Del Cimmuto et al., 2014; Hornik et al., 1995). Similarly, a food waste. study in Kiruna, Sweden, found that socio-demographic factors have no effects on waste sorting behaviour (Vencatasawmy et al., 2. Background of household waste management in Borås 2000). This was confirmed in a recent meta-analysis study which revealed that the socio-demographic factor is a poor predictor of This study was conducted in the city of Borås, Sweden, which waste sorting behaviour, while information, environmental con- has applied a source separation system for more than 20 years. cern and convenience are the strongest predictors (Miafodzyeva The municipality provides inhabitants with free plastic bags in and Brandt, 2013). In contrast, Sidique et al. (2010) found that two different colours so that they can sort their residual waste at demographic factors influence the use of drop-off centres. home into two separate fractions; food waste (black bag) and com- Similarly, several studies show that convenience in sorting, storage bustible waste (white bag). The black and white bags are not sepa- space at home, availability of sorting facilities, access to a kerbside rated but collected in a common container, and subsequently collection system and distance to collection points are important mechanically separated in an optical sorting plant. Here, com- influential factors that can increase the recycling rate (Ando and bustible waste refers to the non-recyclable, non-packaging and Gosselin, 2005; Barr and Gilg, 2005; Bernstad, 2014; Derksen and non-hazardous waste such as diapers and tissue. The food waste Gartrell, 1993; Folz, 1991; Garces et al., 2002; González-Torre is processed for biogas production, which can be used to fuel cars and Adenso-Díaz, 2005; McCarty and Shrum, 1994; Sidique et al., and busses for public transportation, whereas combustible waste is 2010; Thøgersen, 1997). For example, quantitative analysis of a incinerated in the city’s power plant with heat and electricity questionnaire study by Sidique et al. (2010) showed that increased recovery. Other materials, such as packaging, newsprint, batteries, distance to drop-off points is likely to decrease the number of vis- hazardous waste, and bulky waste, should be collected at desig- its. In another questionnaire study, Ando and Gosselin (2005) con- nated drop-off points. The waste management system in Borås, cluded that recycling rates would rise by 66% if doorstep collection based on black and white bags, is a convenient system for the was used instead of only using drop-off points. Similar results were inhabitants to separate the food waste. Bernstad (2014) investi- obtained in a study in Sweden, which used the survey method with gated food waste sorting in another city in Sweden where sorting self-reports, where the property close collection system was found equipment for collection of food waste was installed in the kitchen. to increase collection of packaging materials compared to the The aim was to increase convenience of food waste sorting and the drop-off system (Hage et al., 2009). Dahlén et al. (2007) used waste result was increased collected amount of food waste. The waste composition studies by manual sorting (pick analysis) to analyse management system in Borås was implemented in the early collection systems by comparing the waste flow composition in 1990s and is continuously being developed. The food waste was six Swedish municipalities, and concluded that the availability of initially collected and used for composting, and was therefore property close collection instead of a drop-off system increased denoted the compostable fraction. It was also recommended that the collection rate of sorted recyclables. They also stressed that other waste, such as diapers and pads (all of which are called dia- reliable waste flow data, required for definitive conclusions, are pers in this contribution), were included in the compostable frac- scarce and difficult to access. tion. In addition, trash cans for the black bags were distributed In addition to facilities close to the property, knowledge about to each household in an effort to encourage the inhabitants to sort what and where to sort the waste for recycling is an important deter- the food waste fraction. Information about which waste should be minant for recycling behaviour (Barr and Gilg, 2005; Hornik et al., sorted in the black bag, including diapers, was visible on the trash 1995). Schultz (2002) noted that lack of knowledge is a barrier to can. Ten years later (early 2000s) the technical system had been recycling. Further, lack of information on how to implement the developed and the food waste treatment was changed from com- sorting scheme was listed as a predictor for non-recyclers posting to biogas production. Diapers therefore had to be sorted (De Young, 1988; Vining and Ebreo, 1990), and the availability of into the white bags instead of the black bags, but the old informa- information was shown to be a long-term factor that should be taken tion (recommending sorting of diapers into the black bags) into account (Iyer and Kashyap, 2007). It is also worth noting that the remained on the trash cans. This probably caused miss-sorting of complexity of sorting waste, and how information is communicated, fractions in the black bags. A recent waste composition study con- have been named as key issues in distributing information (Schultz, ducted by the city of Borås revealed that the percentage of miss- 2002). For example, in a recent study in Sweden Bernstad (2014) sorted waste in both black and white bags was, on average, 42% found that written information did not increase separation of food for the city (Rousta and Ekström, 2013). The goals in the waste waste significantly, but that information campaigns need to be cor- management plan for the city require that this should decrease. rectly designed and proper methods must be used to spread this The research presented in this paper was initiated to measure information. It should be noted, however, that studies which inves- the effect of specific interventions aimed at decreasing the miss- tigate the use of knowledge and information to find new ways of sorting ratio for a limited area of the city of Borås. waste sorting differ from those that assess how these factors change existing behaviour in source separation system. The literature review presented above shows that, among the 3. Methods and materials different variables, there is consistency regarding two crucial factors for increasing participation in source separation programmes: 3.1. The study area and participants (1) convenience such as short distance to the collection point and (2) relevant information about the recycling program. The main The study was performed in an urban residential area consisting part of the studies discussed above has focused on behavioural of 208 apartments in nine eight floor buildings. The area is part of 24 K. Rousta et al. / Waste Management 40 (2015) 22–30 the city of Borås, located in the southwest of Sweden. The area had renovated a room that was behind the apartment buildings that 447 inhabitants with diverse socio-economic backgrounds. Two were included in this study. The room was adjacent to existing bins thirds of the population were born outside of Sweden, approxi- that were used for collection of newsprints and glass packaging, mately 37% had medium or high income, approximately 58% had and is about 20–100 m from the entrance of the buildings. The upper secondary school or higher education and 33% of the house- room was called an environmental room and a number of bins were holders owned a car (Swedish Statistics, 2012). According to the provided to collect sorted paper packaging, plastic packaging, housing corporation, most of the inhabitants had been living in this metal packaging, textiles, batteries, light-bulbs, small electronics area for more than seven years, and there was a change of tenants and cooking oils. These bins were located inside the room, and in 10% of the apartments yearly during the period of the study the existing bins for glass packaging and newsprints were posi- (2011–2013). There were several reasons to choose this particular tioned outside the room. In other words, a complete drop-off point area for this study. First, according to the results of a pick analysis was set up for the inhabitants near to their homes, and this prop- in 2010 (Rousta et al., 2011), this area had a high miss-sorted ratio erty close collection point (including the environmental room) was compared to the other districts in the city, and any possible only available for the inhabitants from this particular residential improvements in sorting due to the interventions are expected to area. The room was inaugurated in the middle of June 2013, five be larger compared to areas that already have good sorting. months before the second pick analysis. Since the aim of this study Second, the diversity of the socio-economic background in this was to identify the effect of the distance to drop-off point, informa- area is typical for multi-family apartment buildings in the city. tion that was given to the inhabitants was limited to the availabil- Hence, upcoming results from this study may be used to improve ity of the environmental room and nothing more. A brochure about recycling behaviour in other districts of the city. Third, sampling the environmental room, which was available in different lan- of the white and black bags for pick analysis was possible in this guages, was distributed to all apartments. During these five area since they were easily accessible. This allowed for successful months there were no extra information campaigns so that the fac- implementation of the planned sampling scheme. tor of information was as constant as possible. However, this does In this area black and white bags are normally collected each not eliminate the possibility that inhabitants took the initiative to week in 28 wheeled bins (660 l and simply called bins below). find new information. These bins are placed in designated areas close to the entrance of each building. The closest drop-off point, 20–100 m away from 3.4. Intervention (b): new information sticker on the food waste trash the apartments, had four 600 l bins, two for newsprints, one for cans coloured and one for non-coloured glass packaging. Other sorted recyclables, i.e. paper packaging, metal packaging and plastic New trash cans for food waste were distributed to each apart- packaging, must be transported by the inhabitants to another ment in the area of study. All trash cans had visible stickers with drop-off point that is located one and a half to 2 km away from new sorting information concerning food waste. Compared to the the residential area. information on the old trash cans, the new stickers were designed In a previous study of the same area, it was proved that the with less text and clear figures with correct sorting information for miss-sorted fractions in black and white bags impose a con- the black bags (Fig. 1). All apartments in the building received a let- siderable community cost. Combustible waste in black bags ter one week prior to the distribution date, informing the inhabi- (intended for food waste) and packaging and newsprint in white tants of the intended distribution. On the day of the distribution bags (intended for combustibles) were the most miss-sorted one of the authors, with help of three assistants, knocked on all materials (Rousta and Ekström, 2013). of the (208) apartment doors to distribute the new trash cans or exchange them for the old ones. A short introduction and explana- 3.2. Design of the study tion for changing the trash cans was given. No other information about sorting the waste was given, since we wanted to isolate Two interventions were implemented and evaluated with the effect of the new information on the trash cans on any possible regard to change in the ratio of miss-sorted waste in each of the change in sorting behaviour. The inhabitants that did not answer black and white bags: (a) setting up a property close collection the door on the distribution day were approached (using the same point for sorted waste and (b) new, visible information stickers method) on two other occasions. In total 80% of the 208 apart- on the food waste trash cans in the apartment kitchens. Repeated ments received these new trash cans. measurements of waste composition, by sampling and manual There were no other information campaigns, and the inhabi- sorting (pick analysis), was completed before and after the inter- tants obtained no additional types of information, during this time ventions to analyse their effect. The first pick analysis was con- in order to examine the effect of the new stickers on the inhabi- ducted in November 2011. In June 2012 the trash cans with new tants’ recycling in the black bags when the factor of the conveni- information stickers (intervention (b)) were distributed to the ence was constant. apartments and a property close collection point was established one year later, in June 2013. The second pick analysis was con- 3.5. Waste composition study/pick analysis ducted in November 2013 in order to compare the waste composition with the first pick analysis. The two interventions are Waste composition study by manual sorting, called pick analy- discussed in more detail below. sis, is a practical way for characterizing household waste. The pick analysis method was based on the instructions from Swedish 3.3. Intervention (a): property close collection, establishing an Waste Management Association (2005) which is described and dis- environmental room cussed by Dahlén and Lagerkvist (2008). The pick analyses were performed to measure the ratio of miss-sorted materials in the Some studies (e.g. Dahlén et al., 2007) show that participation black bags (intended for food waste) and white bags (intended in a recycling scheme can increase when the distance to the recy- for combustibles). The pick analyses were performed before and cling station decreased. However, quantifying the effect that this after the interventions, which may have influenced the inhabitants’ distance has on recycling behaviour was not assessed. Therefore recycling behaviour. intervention (a) has been designed to assess this effect. To do this, The study area had twenty-eight 660 l bins for waste collection the housing corporation, in collaboration with the municipality, from all of the 208 apartments, and sampling was done from the K. Rousta et al. / Waste Management 40 (2015) 22–30 25

Fig. 1. Old sticker (left) and new sticker (right) on the trash cans for black bags (intended for food waste). Translation of stickers from top to bottom: left: compostable, leftover food, eggshell, meat and fish, teabags, leftover tea and coffee, tea bags and coffee filters, peels and other remains of fruit and vegetables, wet towel paper, diapers, pads and similar, leftover bread, flower soil and waste, cat litter; right: In the black bag you throw away food waste. These are counted as food waste: leftover meat, fish, dairy product, bread, rice, pasta, fruit and vegetables. entire waste of this area. According to the Swedish Waste 4. Results Management Association (2005), the minimum amount of total sampled waste which is sufficient for statistically meaningful The proportion of analysed samples to total waste flow in the results is approximately 500 kg. The total waste sampled in this area was about 28.0% in 2011 and 28.6% in 2013, which is sufficient study was more than 500 kg for each of the white and black bags. for statistically meaningful results (Swedish Waste Management Sampling was done at the collection point. One bin (660 l) was one Association, 2005). The total weight that was sampled from waste sample. Seven out of 28 bins containing the mixed white and black in the black and white bags is shown in Table 1. bags were randomly selected every Friday in November 2011 and The data from the two pick analyses was statistically analysed November 2013. November has no special holidays in Sweden, for the two interventions, and are presented in the following and was therefore chosen in order to measure the ordinary waste sections. flow and sorting behaviour in the area. At the sampling occasions, the remaining garbage, i.e. in the other 21 bins, was weighed on a 4.1. Results for intervention (a): property close collection weighbridge and the total amount of waste per week was recorded. The number of black and white bags was counted and documented The discussion presented below assumes that intervention (b), before the waste was sorted in the stipulated 271 pick analysis frac- which was new information on the trash cans for food waste, does tions, weighed and registered for each bin (i.e. each sample). During not have a large effect on the sorting of packaging and newsprint in the process numerous photographs were taken and anything unu- the black and white bags. This is because neither the old (2011) nor sual that was found in the waste was documented. The amount of the new (2013) stickers had information on how to discard packa- packaging (plastic, paper and metal) that was collected in the ging or newsprint. Hence intervention (b) is not expected to environmental room during one week in the second pick analysis change sorting of these fractions. (November 2013) was weighed. This was done to estimate the col- The total miss-sorted ratio was 39% in 2013 compared to 55% in lection rate of these fractions after intervention (a). The pick analysis 2011. Over half (56%) of this 16% point decrease came from a data were compiled in kg per household per week (kg/hh/w) as well reduced amount of packaging and newsprint in the total waste as weight percentage (wt.%) of the ratio of miss-sorted fractions. from both bags (see footnote 1 for the different packaging frac- Through this study a two sample t-test and 95% confidence tions). As shown in Fig. 2, the amount of packaging and newsprint interval (p < 0.05) was used to ascertain the statistical relevance was much less in the black bags (intended for food waste) than in of any differences obtained from the sampling campaigns con- white bags (intended for combustible waste) in both years. A two ducted in 2011 and 2013. sample t-test with a 95% confidence interval (p < 0.05) showed that the decrease in miss-sorted packaging and newsprint in white bags 3.6. Reference area is statistically significant. The miss-sorted packaging and newsprint in the black and The city of Borås was willing to share results from pick analyses, white bags is displayed in kg per household per week (kg/hh/w) based on the same method used here, performed in 2010 and 2013 in Fig. 3. in other city districts. A reference area in the western part of the The data for the packaging is shown for each type of packaging city was chosen, since there had been no changes or interventions fraction, and is for before and after introducing the environmental with regards to waste sorting in this area during the relevant years. room to the area, i.e. intervention (a). The two sample t-test and The reference area had 299 apartments with a similar waste man- 95% confidence interval (p < 0.05) showed that the only sta- agement system compared to that used in the area of study. Two tistically significant decrease was paper packaging, glass packaging drop-off points for dry recyclables were available within a distance and newsprints in the white bags. of 400 m from the apartment buildings. Table 1 Weight of the sampled waste from black and white bags. 1 The sample was sorted in the following 27 fractions: newsprint, cardboard, plastic film packaging, foamed plastic packaging, dense plastic packaging, paper packaging, Year Waste in black bags Waste in white bags Proportion glass packaging, metal packaging, glass packaging with deposit, metal packaging with (intended for food (intended for of total (%) deposit, plastic packaging with deposit, food waste, garden waste, diapers and pads, waste) (kg) combustibles) (kg) non-packaging paper, non-packaging glass, non-packaging metal, non-packaging 2011 662.2 1070.7 28.0 plastic, textile fabric, wood, other combustibles, ceramics, other non-combustibles, 2013 712.9 986.1 28.6 medicines, batteries, light-bulbs, electronic equipment. 26 K. Rousta et al. / Waste Management 40 (2015) 22–30

packaging and newsprint decreased. None of these changes are statistically signiﬁcant.

4.2. Results from the collected fractions in the environmental room

As discussed above, all sorted recyclables that were collected in the designated bins in the environmental room were weighed and calculated in units of kg per household per week (kg/hh/w). The weighing was done at the same time as the second pick analysis (November 2013). The results are shown in Table 4, where they are also compared to the total reduction in miss-sorted wastes obtained from both black and white bags (see the last row in Table 3). These results are discussed in more detail below.

4.3. Results for intervention (b): new information sticker on the trash cans for food waste

The discussion presented below assumes that intervention (a), Fig. 2. Comparison of the proportion of miss-sorted packaging and newsprint in which was the introduction of property close collection, does not black bags (intended for food waste) and white bags (intended for combustibles). Results are obtained from the pick analyses in 2011 and 2013. have a large effect on the sorting of food waste in the black and white bags. This is because it is expected that the introduction of As shown in Table 2, the largest percentage reduction was for property close collection will only lead to better sorting of the frac- newsprints in the white bags (approximately 50%) and the lowest tions relevant to the environmental room (packaging, newsprint, was for plastic packaging in the black bags (approximately 2%). In etc.) and not food waste. Improved sorting of the fractions relevant general, the reduction of packaging and newsprint was larger in to the environmental room will affect the percentage of food waste the white bags, except for paper packaging where the reduction in the black and white bags, but not the weight per household per was approximately the same in black and white bags (although it week, which is the unit used in the discussion presented below. was far larger in the white bags when measured in units of kg/ The old trash cans used for food waste were replaced by new hh/w). cans with updated, visible information stickers for correct sorting Table 3 shows the decrease in miss-sorted packaging and news- of food waste. It was not easy to make contact with all households, print per household per week summed over both the black and but after three attempts more than 80% of the apartments had a white bags. That is, the table shows the total miss-sorted packa- new trash can. The results of the pick analyses showed that the ging and newsprint in the residential area (per household and correctly sorted food waste (in black bags) was about 81% in week). It is evident that the total reduction of miss-sorted packa- 2013, which was about 10% more than before the intervention ging and newsprint in the black and white bags after intervention (2011). Fig. 5 shows the average amount of food waste per house- (a) was about 0.7 kg per household per week, which is 31%. hold per week in the black bags (correctly sorted) and white bags Fig. 4 shows the same analysis as that in Fig. 3 but for the miss- (miss-sorted). Data are shown for the area of study and the refer- sorted packaging and newsprint in the reference area. It is evident ence area. In the area of study the amount of correctly sorted food that there was no trend in decreasing or increasing miss-sorting waste after intervention (a) was 2.4 kg/hh/w compared to 2.0 kg/ from 2010 to 2013. The amount of miss-sorted plastic packaging hh/w before the intervention. During the same period the miss- was unchanged in the white bags and decreased slightly in the sorted food waste in white bags (intended for combustibles) black bags. Miss-sorted paper packaging and metal packaging decreased from 1.2 to 0.7 kg/hh/w, which is statistically signiﬁcant. increased in both black and white bags, whereas miss-sorted glass Analysis of the data from the reference area (shown in Fig. 5)

Fig. 3. Comparison of the average amount of miss-sorted packaging and newsprint in black bags (intended for food waste) and white bags (intended for combustibles) [kg per household per week]. The results are from the pick analyses from 2011 and 2013.The two sample t-test and 95% conﬁdence interval (p < 0.05) showed that there was a signiﬁcant decrease in paper packaging, glass packaging and newsprint in the white bags. K. Rousta et al. / Waste Management 40 (2015) 22–30 27

Table 2 Reduction rate of packaging and newsprint in the bags after intervention (a).

Plastic packaging (%) Paper packaging (%) Metal packaging (%) Glass packaging (%) Newsprint (%) Reduction in black bags 1.6 25.0 14.3 19.7 35.7 Reduction in white bags 14.5 23.3 21.0 48.1 49.5

Table 3 The sum of miss-sorted packaging and newsprint found in both black and white bags before (2011) and after (2013) intervention (a) [kg per household per week]. The third row of data shows the reduction in each waste fraction.

Plastic packaging Paper packaging Metal packaging Glass packaging Newsprint Total (kg/hh/w) (kg/hh/w) (kg/hh/w) (kg/hh/w) (kg/hh/w) (kg/hh/w) 2011 0.548 0.562 0.130 0.231 0.762 2.233 2013 0.482 0.430 0.105 0.133 0.391 1.541 Reduction 0.066 0.132 0.025 0.098 0.371 0.692

Fig. 4. Same as Fig. 3 but for the reference area. (Data obtained from the city of Borås).

Table 4 Packaging collected in the environmental room [kg per household per week].

Plastic Paper Metal packaging packaging packaging (kg/hh/w) (kg/hh/w) (kg/hh/w) Collected packaging in 0.075 0.197 0.026 environmental room Reduction miss-sorted in 0.066 0.132 0.025 white and black bags

revealed that the amount of correctly sorted food waste did not change, but miss-sorted food waste in the white bags increased. The improved sorting of food waste in black bags (in kg per household per week) increases the collection rate of this fraction. As mentioned above, the original trash cans for the food waste showed that diapers should be included in this fraction, and this information was changed by intervention (b). Table 5 shows that Fig. 5. The amount of correctly sorted food waste (black bags) and miss-sorted food the miss-sorted diapers in the black bags were reduced about waste (white bags) in the area of study before and after intervention (b), compared 70% from 2011 to 2013. to the reference area [kg per household per week]. Fig. 6 compares the sorting of diapers after intervention (b) with the same data from the reference area. A two sample t-test and 95% conﬁdence interval (p < 0.05) in the reference area. The decrease of miss-sorted diapers in black showed that there was a signiﬁcant decrease of miss-sorted dia- bags (0.16 kg/hh/w) was less than the increase in white bags pers in the black bags in the area of study, which was not found (0.24 kg/hh/w). 28 K. Rousta et al. / Waste Management 40 (2015) 22–30

Table 5 environmental room, which proves that the environmental room Amount and proportion of food waste and diapers measured in black bags before had an effect on better sorting behaviour. (2011) and after (2013) intervention (b). A second point which further supports the fact that the Food waste Diapers Total Food waste Diapers improved sorting was not due to changes in consumption is that (kg/hh/w) (kg/hh/w) (kg/hh/w) (wt.%) (wt.%) the results from the reference area – where there was no interven- 2011 1.976 0.221 2.761 71.6 8.0 tion – showed no trend in decreased or increased miss-sorted 2013 2.368 0.060 2.918 81.1 2.1 material. It can be concluded that intervention (a), shorter distance to the drop-off point, was the main driver of the decreased miss- sorting of packaging and newsprint found in the pick analysis after the intervention. Another important observation is that reduction in miss-sorted plastic packaging is less than the other packaging materials. Further evaluation of the pick analyses showed that more than 80% of the miss-sorted plastic packaging was plastic film such as freezer bags, and shopping bags. This may indicate that the inhabitants regard these materials as combustibles. An unexpected result was the large (50%) reduction of newsprint in the white bags. The reason for this was investigated, and it was found that the collection of newsprints from the entire city had reduced over this period. According to statistics from the Swedish Waste Management Association, collected newsprints decreased by approximately 17% from 2011 to 2013 in this city. This was also confirmed by a reduction of collected newsprints of 22% in the reference area from 2010 to 2013 (Fig. 4). Hence, the 50% decrease of collected newsprint was partially due to changes in consumption patterns, for example using e-news instead of paper news. However, the difference between the reduction in the study area (50%) and that in the rest of the city (17%) or refer- Fig. 6. The amount of miss-sorted diapers in black bags (intended for food waste) ence area (22%) indicates that the primary reason for the large and correctly sorted diapers in white bags (intended for combustibles) in the area of study before and after intervention (b), compared to the reference area [kg per reduction was introducing the environmental room. If one takes household per week]. the change in consumption into account then the total reduction of miss-sorted newsprint due to intervention (a) is 0.308 kg/hh/w 5. Discussion (17% less than the 0.371 kg/hh/w shown in Table 3). Consequently, the total reduction of packaging and newsprint 5.1. Intervention (a): property close collection due to intervention (a) is 0.629 kg/hh/w (instead of 0.692 kg/hh/ w shown in Table 3). As a result, the recalculated reduction of The amount of miss-sorted packaging and newsprint in white miss-sorted packaging and newsprint that was achieved by intro- and black bags decreased significantly when there was a substan- ducing the environmental room was 28%, which still is a con- tial decrease in the distance to drop-off point (Fig. 3). This finding siderable improvement of the source separation system. The supports the results of other studies which concluded that a recalculated data are shown in Table 6. shorter distance to the recycling point results in increased recy- The present study identifies an additional factor that can be cling rate (Derksen and Gartrell, 1993; González-Torre and used to improve source separation systems. Bins for collection of Adenso-Díaz, 2005). The question might be raised: Was the glass packaging and newsprint had been available near to the decrease in distance the only reason for decreasing the miss-sorted apartments in the area of study for many years. Since these bins packaging and newsprint in the bags? A change of consumption were not indoors, citizens from other residential areas also had choices may have affected the results of the second pick analysis. access to them, which increased the uncertainty when measuring The inhabitants may have consumed fewer goods or changed theirs the amount of recyclables per household. However, according to consumption to other types of goods with different types of packa- statistics from the municipal waste management company, who ging, which could have affected the results of the second pick collects the glass packaging when the bins are full, the collection analysis. There was no specific measurement of consumption pat- frequency for glass packaging had increased by 40% after the terns in the study area. However, the material flow of packaging environmental room was introduced. The corresponding statistics and newsprint – which had decreased in white and black bags – for newsprint could not be obtained since newsprint bins are emp- was followed since the total amount of packaging material col- tied each week regardless of how full they are, and the weight was lected in the environmental room was weighed and compared to the decreased amount of miss-sorted packages in the white and black bags (Table 4). For example, 0.075 kg plastic packaging per Table 6 Recalculated amounts of miss-sorted newsprint and total miss-sorted packaging and household per week was collected in the environmental room newsprint [kg per household per week]. The data are taken from Table 3 and take into whereas miss-sorted plastic packaging in the bags decreased account consumption changes in newsprint from 2011 to 2013. 0.066 kg per household. Hence the amount of sorted recyclables Newsprint Total packaging and collected in the environmental room exceeded the decrease of (kg/hh/w) newsprint (kg/hh/w) miss-sorted amounts in the bags. A probable scenario is that the 2011 0.762 2.233 overall packaging waste generation rate did not decrease in the 2013 0.391 1.541 studied area from 2011 to 2013, which confutes the argument Reduction (primary) 0.371 0.692 (31%)a about decreased consumption as a possible explanation of Reduction (considering 0.308 0.629 (28%)a decreased amount of miss-sorted materials. The packaging which consumption changing) was not miss-sorted in the bags had been sorted correctly in the a Percent decrease. K. Rousta et al. / Waste Management 40 (2015) 22–30 29 30 K. Rousta et al. / Waste Management 40 (2015) 22–30 not registered for the area of study. In spite of the fact that these to a 30% decrease in miss-sorted packaging and newsprint and that are uncertaintiesand/or resistance among householders regarding Hornik, J., Cherian, J., Madansky, M., Narayana, C., 1995. Determinants of recycling bins were available before the environmental room was introduced intervention (b) led to a 70% decrease in miss-sorted diapers. If one the correct sorting of this fraction. This indicates that inhabitants behavior: a synthesis of research results. J. Socio-Econ. 24 (1), 105–127. Iyer, E.S., Kashyap, R.K., 2007. Consumer recycling: role of incentives, information, in the area of study, the pick analyses showed that the amount of assumes that these types of interventions have a similar effect need to be informed that these wastes should not be sorted as and social class. J. Consum. Behav. 6 (1), 32–47. both miss-sorted glass packaging and newsprints decreased sig- when applied to other waste fractions (e.g., packaging and news- combustibles. Third, in a well-established source separation sys- Leitol, C., 2014. Resource and cost efficient selective collection. Pollack Periodica 9 nificantly due to this intervention (Table 2). This indicates that a print), then it can be reasonably assumed that these interventions tem, which was implemented in the city of Borås more than (Suppl. 1), 43–54. McCarty, J.A., Shrum, L.J., 1994. The recycling of solid wastes: personal values, value property close collection facility accepting several waste fractions can decrease miss-sorting of these fractions by up to 80% when 20 years ago, improved convenience and information were shown orientations, and attitudes about recycling as antecedents of recycling behavior. motivates inhabitants to sort waste more than a facility that col- they are applied together. Similarly to the discussion given above to be significant influential factors. Finally, the study showed there J. Bus. Res. 30 (1), 53–62. lects only a few fractions. This correlation has been examined in with respect to intervention (a), the possible effects of informal is a need for new ways of communication with inhabitants about Miafodzyeva, S., Brandt, N., 2013. Recycling behaviour among householders: synthesizing determinants via a meta-analysis. Waste Biomass Valorization 4 a study in the UK, which supports the result of the present study information sharing between neighbours which may have risen source separation schemes. Further research is required to mea- (2), 221–235. (Woodard et al., 2006). during intervention (b) was not considered. sure the effects of information and different types of communica- Rada, E.C., Ragazzi, M., Fedrizzi, P., 2013. Web-GIS oriented systems viability for This study focused on the effect of distance to the drop-off point tion channels on recycling behaviour in a convenient system municipal solid waste selective collection optimization in developed and transient economies. Waste Manage. 33 (4), 785–792. when other factors, such as knowledge about why and how to sort such as a property close collection system. 5.3. Sources of errors Rousta, K., Ekström, K.M., 2013. Assessing incorrect household waste sorting in a the waste, were almost constant. The significant decrease of miss- medium-sized Swedish City. Sustainability 5 (10), 4349–4361. sorted packaging showed that inhabitants, who knew why and Acknowledgments Rousta, K., Kaeni Moghadam, M., Haji Karimkhani, P., Richards, T., 2011. Evaluation It is not possible to control all noise factors when a study con- how to sort the waste, will act if the facility is near to their homes. the inhabitants’ participation in separation at source by waste characterization. cerns a large number of people over a long period of time. For Wastes: Solutions Treat. Opportunities. This supports the motivation–ability–opportunity–behaviour The authors acknowledge ‘‘Borås Energi and Miljö’’ for example, one factor which might have affected the results is that Schultz, P.W., 2002. Knowledge, information, and household recycling: examining model where motivation alone is not enough for a behavioural supporting of this study and ‘‘Boråsregionen-Sjuhärads the knowledge-deficit model of behavior change. In: Dietz, T., Stern, P.C. (Eds.), there was a change of tenants in 10% of the apartments yearly dur- change. It requires opportunity, in this case the nearby facilities, Kommunalförbund’’ for funding. New Tools for Environmental Protection: Education, Information, and Voluntary ing the period of the study (2011–2013). It was not possible to Measures. National Academy Press, Washington DC, pp. 67–82. which can influence environmental behaviour (Thøgersen, 1994). study the behaviour of the individual inhabitants. Seacat, J.D., Northrup, D., 2010. An information–motivation–behavioral skills The environmental room may also have awakened interest to look References assessment of curbside recycling behavior. J. Environ. Psychol. 30 (4), 393–401. The aim of this study was to see the effect of distance and infor- for information and discuss with neighbours about how to sort the SEPA, 2013. Swedish Environmental Protection Agency. Förslag till etappmål för mation isolated from other factors. Although the interventions ökad förberedelse för återanvändning och materialåtervinning av avfall. waste correctly. This effect was not analysed in the present study. Ando, A.W., Gosselin, A.Y., 2005. Recycling in multifamily dwellings: does were implemented at different times, the measurement of their convenience matter? Econ. Inq. 43 (2), 426–438. Stockholm, Sweden, (in Swedish). SFS, 1994a, Förordningen om producentansvar för returpapper (The Ordinance of effects was conducted at the same time, i.e., with the second pick Baltes, L., Draghici, C., Manea, C., Ceausescu, D., Tierean, M., 2009. Trends in 5.2. Intervention (b): easy access to correct information selective collection of the household waste. Environ. Eng. Manage. J. 8 (4), 985– Producers’ Responsibility for Waste Paper). Swedish legislation., SFS 1994:1205. analysis. There might have been some interaction effects of these 991. (accessed 10.10.14). The type of information that is conveyed to inhabitants, as well SFS, 1994b, Förordningen om producentansvar för förpackningar (The fraction may support improved sorting of other fractions. Such actions to a growing environmental problem: development and application of a as when and how it is conveyed (accessibility), probably affects framework to guide local waste policy. Appl. Geogr. 25 (3), 226–247. Ordinance of Producers’ Responsibility for Packaging Materials). Swedish indirect effects were not considered. legislation., SFS 1994:1235. (accessed 10.10.14). the internet but, in spite of this, they sorted diapers into the black SFS, 2001, Förordning om deponering av avfall (The Ordinance of Landfilling). in the data collection, one of the authors has conducted all sam- 2013. 3R’S From citizens point of view and their proposal from a case-study. bags (intended for food waste). The most likely reason for this is UPB Sci. Bull., Ser. D 75 (4), 253–264. Swedish legislation., (SFS2001:512). (accessed 10.10.14). that the information on the trash cans for the black bags had old, Dahlén, L., Lagerkvist, A., 2008. Methods for household waste composition studies. The same measurement equipment and routines were used for Waste Manage. 28 (7), 1100–1112. Sidique, S.F., Lupi, F., Joshi, S.V., 2010. The effects of behavior and attitudes on drop- incorrect information, i.e., that diapers should be sorted into this off recycling activities. Resour. Conser. Recycl. 54 (3), 163–170. both pick analyses. Dahlén, L., Vukicevic, S., Meijer, J.-E., Lagerkvist, A., 2007. Comparison of different trash can. The 70% reduction of miss-sorted diapers in the black collection systems for sorted household waste in Sweden. Waste Manage. 27 Swedish Statistics, 2012. Socio-economic statistics, Stockholm, Sweden, (in Swedish). bags, when the new trash cans with correct information were dis- (10), 1298–1305. De Young, R., 1988. Exploring the difference between recyclers and non-recyclers: Swedish Waste Management Association, 2005. Manual for Pick Analysis of tributed, confirmed that this type of information is important. 6. Conclusion the role of information. J. Environ. Sys. 18 (4), 341–351. Houshold Waste/Manual för plockanalys av hushållsavfall, Malmö, Sweden (in Comparing the reduction of diapers in the study area with the Del Cimmuto, A., Mannocci, A., Ribatti, D., Boccia, A., La Torre, G., 2014. Impact on Swedish). (accessed 10.10.14). reference area (Fig. 6) shows that intervention (b) is the main rea- Two interventions: (a) shorter distance to the drop-off point and knowledge and behaviour of the general population of two different methods of solid waste management: an explorative cross-sectional study. Waste Manage. Swedish Waste Management Association, 2014. Swedish waste management 2014 son for change, i.e., there was a decrease of miss-sorted diapers in (b) easy access to correct information, significantly improved source Res. 32 (6), 556–561. (Svensk avfallshantering 2014). Malmö, Sweden, (in Swedish). (accessed 10.10.2014). pers in the white bags in the area of study whereas there was a point from 2 km to 50 m (intervention (a)) reduced the amount of 434–442. Folz, D.H., 1991. Recycling program design, management, and participation: a Thøgersen, J., 1994. A model of recycling behaviour, with evidence from Danish decrease of diapers in both bags in the reference area. The reason miss-sorted packaging and newsprint from 2.23 to 1.61 kg per national survey of municipal experience. Public Admin. Rev. 51 (3), 222–231. source separation programmes. Int. J. Res. Mark. 11 (2), 145–163. for the decrease (in both bags) in the reference area may be due household per week. Thus property close location of the drop-off Garces, C., Lafuente, A., Pedraja, M., Rivera, P., 2002. Urban waste recycling Thøgersen, J., 1997. Facilitating recycling: reverse-distribution channel design for behavior: antecedents of participation in a selective collection program. participation and support. Soc. Mark. Q. 4 (1), 42–55. to decreased need for diapers in a population with unchanged point reduced the miss-sorted packaging by approximately 30%. Vencatasawmy, C.P., Öhman, M., Brännström, T., 2000. A survey of recycling waste sorting behaviour. In addition to improving sorting of dia- Environ. Manage. 30 (3), 378–390. This decrease is significant, and it can therefore be concluded that González-Torre, P.L., Adenso-Díaz, B., 2005. Influence of distance on the motivation behaviour in households in Kiruna, Sweden. Waste Manage. Res. 18 (6), 545– pers in the area of study, the miss-sorted food waste in white bags the shorter distance to drop-off point is necessary. Combining this and frequency of household recycling. Waste Manage. 25 (1), 15–23. 556. decreased and was sorted correctly into the black bags (Fig. 5). with interventions that increase the inhabitants’ knowledge of Grodzin´ ska-Jurczak, M., Tomal, P., Tarabuła-Fiertak, M., Nieszporek, K., Read, A.D., Vining, J., Ebreo, A., 1990. What makes a recycler? A comparison of recyclers and nonrecyclers. Environ. Behav. 22 (1), 55–73. Both of these improvements show the effect of intervention (b), how to use these facilities may further decrease miss-sorting. 2006. Effects of an educational campaign on public environmental attitudes and behaviour in Poland. Resour. Conser. Recycl. 46 (2), 182–197. Woodard, R., Harder, M.K., Bench, M., 2006. Participation in curbside recycling i.e. introduction of the new sticker with correct information on Information stickers about food waste sorting (intervention (b)) Hage, O., Söderholm, P., Berglund, C., 2009. Norms and economic motivation in schemes and its variation with material types. Waste Manage. 26 (8), 914–919. the trash cans for black bags. A sticker on the trash can is a type reduced the miss-sorted fraction, such as diapers, by more than household recycling: empirical evidence from Sweden. Resour. Conser. Recycl. of information which is always accessible. It does not require time 70%. It was concluded that easily accessible and correct informa- 53 (3), 155–165. to seek in a brochure or on the internet and it is visible every day. tion about how to sort the waste can lead to a large reduction in Such simple and visible information will continuously be seen by the miss-sorted fraction when the system is convenient. the inhabitants, and will be seen at the moment when the informa- Therefore, it is expected that combining a property close collection tion is needed (i.e., when they sort the waste). It is concluded that system (to have a drop-off point near to the inhabitants’ homes) visible and easily understood information, which is seen every day and proper information channels can reduce the miss-sorted and at the time of sorting, improves a convenient source separation packaging and newsprints much more than 30%, and as much as system (i.e. having two bags of different colours for sorting food 80%. Although it is difficult to generalise these results for other waste and combustible waste in the kitchen). This conclusion can source separation schemes, they support the possibility of open a new way of thinking about the most effective channels quantifying the effects of interventions in recycling behaviour. for information in the future. However, further research is required There are several practical applications of the conclusions. First, to reveal the effect of different information channels in this field. a property close facility that collects several separated fractions Results from both interventions show that a property close col- provides a larger motivation for inhabitants to participate in source lection system can be a very effective way to increase the recycling separation schemes than facilities that collect fewer fractions. ratio when correct information is communicated to inhabitants in Second, the small reduction that was seen for miss-sorted plastic appropriate ways. In fact, this study shows that intervention (a) led packaging (compared to other recyclables) revealed that there 30 K. Rousta et al. / Waste Management 40 (2015) 22–30 are uncertaintiesand/or resistance among householders regarding Hornik, J., Cherian, J., Madansky, M., Narayana, C., 1995. Determinants of recycling the correct sorting of this fraction. This indicates that inhabitants behavior: a synthesis of research results. J. Socio-Econ. 24 (1), 105–127. Iyer, E.S., Kashyap, R.K., 2007. Consumer recycling: role of incentives, information, need to be informed that these wastes should not be sorted as and social class. J. Consum. Behav. 6 (1), 32–47. combustibles. Third, in a well-established source separation sys- Leitol, C., 2014. Resource and cost efficient selective collection. Pollack Periodica 9 tem, which was implemented in the city of Borås more than (Suppl. 1), 43–54. McCarty, J.A., Shrum, L.J., 1994. The recycling of solid wastes: personal values, value 20 years ago, improved convenience and information were shown orientations, and attitudes about recycling as antecedents of recycling behavior. to be significant influential factors. Finally, the study showed there J. Bus. Res. 30 (1), 53–62. is a need for new ways of communication with inhabitants about Miafodzyeva, S., Brandt, N., 2013. Recycling behaviour among householders: synthesizing determinants via a meta-analysis. Waste Biomass Valorization 4 source separation schemes. Further research is required to mea- (2), 221–235. sure the effects of information and different types of communica- Rada, E.C., Ragazzi, M., Fedrizzi, P., 2013. Web-GIS oriented systems viability for tion channels on recycling behaviour in a convenient system municipal solid waste selective collection optimization in developed and transient economies. Waste Manage. 33 (4), 785–792. such as a property close collection system. Rousta, K., Ekström, K.M., 2013. Assessing incorrect household waste sorting in a medium-sized Swedish City. Sustainability 5 (10), 4349–4361. Acknowledgments Rousta, K., Kaeni Moghadam, M., Haji Karimkhani, P., Richards, T., 2011. Evaluation the inhabitants’ participation in separation at source by waste characterization. Wastes: Solutions Treat. Opportunities. The authors acknowledge ‘‘Borås Energi and Miljö’’ for Schultz, P.W., 2002. Knowledge, information, and household recycling: examining supporting of this study and ‘‘Boråsregionen-Sjuhärads the knowledge-deficit model of behavior change. In: Dietz, T., Stern, P.C. (Eds.), Kommunalförbund’’ for funding. New Tools for Environmental Protection: Education, Information, and Voluntary Measures. National Academy Press, Washington DC, pp. 67–82. Seacat, J.D., Northrup, D., 2010. An information–motivation–behavioral skills References assessment of curbside recycling behavior. J. Environ. Psychol. 30 (4), 393–401. SEPA, 2013. Swedish Environmental Protection Agency. Förslag till etappmål för Ando, A.W., Gosselin, A.Y., 2005. Recycling in multifamily dwellings: does ökad förberedelse för återanvändning och materialåtervinning av avfall. convenience matter? Econ. Inq. 43 (2), 426–438. Stockholm, Sweden, (in Swedish). Baltes, L., Draghici, C., Manea, C., Ceausescu, D., Tierean, M., 2009. Trends in SFS, 1994a, Förordningen om producentansvar för returpapper (The Ordinance of selective collection of the household waste. Environ. Eng. Manage. J. 8 (4), 985– Producers’ Responsibility for Waste Paper). Swedish legislation., SFS 1994:1205. 991. (accessed 10.10.14). actions to a growing environmental problem: development and application of a SFS, 1994b, Förordningen om producentansvar för förpackningar (The framework to guide local waste policy. Appl. Geogr. 25 (3), 226–247. Ordinance of Producers’ Responsibility for Packaging Materials). Swedish Bernstad, A., 2014. Household food waste separation behavior and the importance legislation., SFS 1994:1235. (accessed 10.10.14). 2013. 3R’S From citizens point of view and their proposal from a case-study. SFS, 2001, Förordning om deponering av avfall (The Ordinance of Landfilling). UPB Sci. Bull., Ser. D 75 (4), 253–264. Swedish legislation., (SFS2001:512). (accessed 10.10.14). Waste Manage. 28 (7), 1100–1112. Sidique, S.F., Lupi, F., Joshi, S.V., 2010. The effects of behavior and attitudes on drop- Dahlén, L., Vukicevic, S., Meijer, J.-E., Lagerkvist, A., 2007. Comparison of different off recycling activities. Resour. Conser. Recycl. 54 (3), 163–170. collection systems for sorted household waste in Sweden. Waste Manage. 27 Swedish Statistics, 2012. Socio-economic statistics, Stockholm, Sweden, (in (10), 1298–1305. Swedish). De Young, R., 1988. Exploring the difference between recyclers and non-recyclers: Swedish Waste Management Association, 2005. Manual for Pick Analysis of the role of information. J. Environ. Sys. 18 (4), 341–351. Houshold Waste/Manual för plockanalys av hushållsavfall, Malmö, Sweden (in Del Cimmuto, A., Mannocci, A., Ribatti, D., Boccia, A., La Torre, G., 2014. Impact on Swedish). (accessed 10.10.14). solid waste management: an explorative cross-sectional study. Waste Manage. Swedish Waste Management Association, 2014. Swedish waste management 2014 Res. 32 (6), 556–561. (Svensk avfallshantering 2014). Malmö, Sweden, (in Swedish). (accessed 434–442. 10.10.2014). Folz, D.H., 1991. Recycling program design, management, and participation: a Thøgersen, J., 1994. A model of recycling behaviour, with evidence from Danish national survey of municipal experience. Public Admin. Rev. 51 (3), 222–231. source separation programmes. Int. J. Res. Mark. 11 (2), 145–163. Garces, C., Lafuente, A., Pedraja, M., Rivera, P., 2002. Urban waste recycling Thøgersen, J., 1997. Facilitating recycling: reverse-distribution channel design for behavior: antecedents of participation in a selective collection program. participation and support. Soc. Mark. Q. 4 (1), 42–55. Environ. Manage. 30 (3), 378–390. Vencatasawmy, C.P., Öhman, M., Brännström, T., 2000. A survey of recycling González-Torre, P.L., Adenso-Díaz, B., 2005. Influence of distance on the motivation behaviour in households in Kiruna, Sweden. Waste Manage. Res. 18 (6), 545– and frequency of household recycling. Waste Manage. 25 (1), 15–23. 556. Grodzin´ ska-Jurczak, M., Tomal, P., Tarabuła-Fiertak, M., Nieszporek, K., Read, A.D., Vining, J., Ebreo, A., 1990. What makes a recycler? A comparison of recyclers and 2006. Effects of an educational campaign on public environmental attitudes and nonrecyclers. Environ. Behav. 22 (1), 55–73. behaviour in Poland. Resour. Conser. Recycl. 46 (2), 182–197. Woodard, R., Harder, M.K., Bench, M., 2006. Participation in curbside recycling Hage, O., Söderholm, P., Berglund, C., 2009. Norms and economic motivation in schemes and its variation with material types. Waste Manage. 26 (8), 914–919. household recycling: empirical evidence from Sweden. Resour. Conser. Recycl. 53 (3), 155–165.

Paper IV

V Rousta, K., Bolton, K., Dahlén, L., (2016) A Procedure to Transform Recycling Behavior for Source Separation of Household Waste PAPER I PAPER Recycling 1(1): 147-165

Paper IV

V Rousta, K., Bolton, K., Dahlén, L., (2016) A Procedure to Transform Recycling Behavior for Source Separation of Household Waste PAPER I PAPER Recycling 1(1): 147-165 recycling

Article A Procedure to Transform Recycling Behavior for Source Separation of Household Waste

Kamran Rousta 1,*, Kim Bolton 1 and Lisa Dahlén 2 1 Swedish Centre for Resource Recovery, University of Borås, Borås 501 90, Sweden; [email protected] 2 Waste Science and Technology, Luleå University of Technology, Luleå 971 87, Sweden; [email protected] * Correspondence: [email protected]; Tel.: +46-33-4354-4644 Academic Editor: Michele Rosano Received: 18 March 2016; Accepted: 14 April 2016; Published: 16 April 2016

Abstract: Household waste separation at the source is a central part of waste management systems in Sweden. Resource recovery of materials and energy increased substantially after separate collection was implemented in the 1990s. A procedure to transform recycling behavior for the sorting of household waste—called the recycling behavior transition (RBT) procedure—was designed and implemented in a waste management system in Sweden. Repeated use of this procedure, which will assist in the continual improvement of household sorting, consists of the following four consecutive steps: (i) evaluating the current sorting behavior; (ii) identifying appropriate interventions; (iii) implementing the interventions, and; (iv) assessing the quantitative effect of the interventions. This procedure follows action research methodology and it is the ﬁrst time that such a procedure has been developed and implemented for the sorting of household waste. The procedure can easily be adapted to any source separation system (which may have different local situations) and, by improving the source separation, will increase the resource recovery in the waste management system. The RBT procedure, together with its strengths and weaknesses, is discussed in this paper, and its implementation is exempliﬁed by a pilot study done in Sweden.

Keywords: recycling behavior; resource recovery; source separation; household waste

1. Introduction Household waste separation at the source was introduced in Sweden in the early 1990s. Here household waste refers to all waste that is collected in curbside collection systems, the packaging materials in bring systems, and hazardous waste, electronics, bulky waste, etc. which are collected at recycling centers. There is not a single system for the entire country, but different regions provide different collection systems for the inhabitants. Source separation increases both the rate of recycling of recyclable materials and of biological treatment of food waste. Material recycling has more than doubled, and biological treatment has more than quadrupled, since source separation was implemented in Sweden. In 2014 the country generated about 4.55 million tons of household waste, of which 36% was treated by material recycling, 16% by biological methods, 47% was used for energy recovery, and 1% was landﬁlled [1]. This illustrates that an effective source separation scheme can increase resource recovery from household waste. Although waste separation at the source is a common and economical way for separating the recyclable fractions, effective participation by inhabitants is required to increase the correct sorting of the recyclables and food waste [2]. Many previous studies have investigated factors that determine the sorting behavior of inhabitants to better understand this phenomenon. A wide range of factors have been studied, and the factors that are chosen for a particular study often depend on the researcher’s interest and discipline, the case and situation, the scope of the study, and the method used for the study. Some studies found that intrinsic factors, such as attitudes towards recycling and environmental

Recycling 2016, 1, 147–165; doi:10.3390/recycling1010147 www.mdpi.com/journal/recycling recycling

Article A Procedure to Transform Recycling Behavior for Source Separation of Household Waste

Keywords: recycling behavior; resource recovery; source separation; household waste

Recycling 2016, 1, 147–165; doi:10.3390/recycling1010147 www.mdpi.com/journal/recycling Recycling 2016, 1, 147–165 148 Recycling 2016, 1, 147–165 149 concern, affect sorting behavior (e.g., [3–8]). Other studies showed that convenient and easy access to and design interventions for improvement of source separation systems and to assess their effect in an recycling facilities are decisive factors (e.g., [9–12]). objective and quantifiable manner. Other studies developed a model and framework to explain how different factors affect recycling behavior. For example, the motivation-ability-opportunity-behavior model explains that motivation is Objectives necessary but not sufficient for environmentally friendly behavior, but that ability and opportunity The aim was to create, test, and evaluate a procedure for continuous improvement of any source to behave in the correct manner are also required [13]. For source separation of household waste, separation system for household waste. Possible changes in sorting behavior do not depend on the situational factors such as convenient and easy access to recycling stations are opportunities, and procedure itself, but on the intervention that is used and how it is implemented. The procedure, called knowledge of why and how to sort the waste, as well as past experiences and habits of sorting, recycling behavior transition (RBT) in this paper, identifies the extent of the missorting of household are the examples of ability. In a similar study, Barr (2002) [14] presented a conceptual framework waste and interventions that can improve the sorting; it develops and implements the intervention of environmental behavior. Barr (2002) [14] explained that environmental values of inhabitants do and then assesses the effect of the intervention. The objectives of the study are summarized in three not have a direct impact on behavior. Instead, behavioral intention is needed as a bridge between research questions: environmental values and behavior, and this bridge has a central role in his framework. Situational factors and psychological variables can influence both intention and behavior, and are included in 1. How should the current sorting behavior be evaluated? the framework, but are not as central as behavioral intention. Context, knowledge, and experience 2. How should the appropriate interventions to improve this behavior be identified? are examples of situational factors, whereas intrinsic motivation, subjective norms, and altruism are 3. How should the effect of interventions be assessed? examples of psychological variables. The studies done by Thøgersen (1994) [13] and Barr (2002) [14] (which were previously discussed 2. Methods and Materials above), used recycling behavior (i.e., waste sorting at the source) to exemplify how their model and Although the procedure is novel, it is based on established scientific methods (such as pick framework can explain which factors are most important for environmental behavior. Both studies analysis and interview techniques). The scientific merits of these methods have been discussed in found that factors such as information about the waste sorting scheme and accessibility to the collection detail in the literature (e.g., [20–22]) and hence this is not a focus of the present contribution. Instead, facilities played a crucial role in determining the behavior. we discuss the ways that these methods are combined in the RBT procedure, and exemplify the Pieters (1991) [15] examined recycling behavior by combining a survey study with measurements methods and their combination by presenting results when the procedure was implemented in a of the recyclable materials in the waste stream. The aim was to identify which factors determine residential area in Sweden. inhabitants’ participation in waste separation schemes. In contrast, Tucker et al. (1999) [16] used a mathematical model to simulate the household waste management behavior. In another study, Tucker 2.1. Design of the Study (2001) [17] developed a hypothesized cause-effect model of recycling. Intention to recycle plays the The RBT procedure was tested in a pilot residential area in Sweden with the intention to test the central role in this model, and intention can be influenced by pro-recycling attitudes, social norms, and method and to improve waste sorting behavior. The first step was to identify the prevailing sorting specific barriers such as perceived obstacles and perceived effectiveness. This model also emphasizes behavior, which was examined by measurement of waste composition by sampling and manual sorting the fact that intention to recycle is not sufficient to cause recycling. Similar to Barr (2002) [14] and (pick analysis). In the second step, factors that hindered correct sorting were identified using two types Thøgersen (1994) [13], this model also considers the factors, called personal difficulties, which hinder of interviews with the inhabitants: semi-structured and structured interviews. Although it is possible the transformation of intention into actual recycling behavior. The studies mentioned above describe that these hindrances can be identified by other means, such as literature surveys, interviews have the the various factors and determinants that influence recycling behavior, and the importance of these advantage in that they include the inhabitants in the identification of appropriate interventions which different factors depends on the concept of the study, the method used in the study, and where the increases the probability that the chosen intervention will be effective. Next, based on the interview study was conducted. results, interventions in the waste collection system were designed and implemented. Finally, another Almost all of the above studies have been performed under specific circumstances and in a pick analysis was conducted to investigate the effect of the interventions. The pilot area and methods limited geographical area. It is difficult to generalize the findings to source separation schemes in used are presented in more detail below. different places in the world. Wilson et al. (2012) [18] compared waste management in 20 cities and clearly showed that there is not a single solution that is suitable for all management systems 2.2. Description of the Pilot Area Used to Test the Procedure and cities. Depending on the available facilities, the length of time that the recycling scheme has been implemented, the structure of the waste management system, relevant policies, etc. different The pilot area that was selected for testing the procedure is not relevant to the procedure and, as factors could influence the recycling scheme and its improvement in different ways. Results from mentioned above, the procedure can be implemented in any area that has waste sorting at the source. previous studies provide a wide range of factors that can improve recycling behavior (i.e., to create In spite of this, a description of the pilot area is provided since this is needed if the results presented effective interventions). However, no study has introduced a procedure for designing relevant here are to be compared with results from other areas. interventions and assessing their effect on recycling behavior. The procedure described in this work The pilot area that was selected is a residential area in the city of Borås, which lies in the southwest is based on the action research routine including look (gathering data and describing situation), think of Sweden. This area was chosen since the waste separation scheme was not functioning as well as (analyzing and theorizing), and act (implementation action and evaluate its effectiveness) [19]. This in other parts of the city [23]. According to Statistics Sweden [24], the pilot area had 447 inhabitants is the first procedure that combines interventions with their assessment. In addition, the procedure with a diverse socio-demographic background who lived in nine apartment buildings (a total of can be used in any source separation scheme anywhere in the world. No matter which kind of 208 apartments). Approximately 67% of the inhabitants were born outside of Sweden and immigrated waste collection system is implemented, procedures for continuous improvement of the system to Sweden more than three years before the procedure was implemented [24]. 31% of the inhabitants (quality management tools) are needed. Therefore, it is relevant to develop a procedure to investigate were aged from 25 to 44 years and 20% were 45 to 64 years old. It was mainly inhabitants from these age groups that were involved in the interviews in Step 2 of the procedure. Approximately 63% of the Recycling 2016, 1, 147–165 149 and design interventions for improvement of source separation systems and to assess their effect in an objective and quantifiable manner.

Objectives The aim was to create, test, and evaluate a procedure for continuous improvement of any source separation system for household waste. Possible changes in sorting behavior do not depend on the procedure itself, but on the intervention that is used and how it is implemented. The procedure, called recycling behavior transition (RBT) in this paper, identiﬁes the extent of the missorting of household waste and interventions that can improve the sorting; it develops and implements the intervention and then assesses the effect of the intervention. The objectives of the study are summarized in three research questions:

1. How should the current sorting behavior be evaluated? 2. How should the appropriate interventions to improve this behavior be identiﬁed? 3. How should the effect of interventions be assessed?

2. Methods and Materials Although the procedure is novel, it is based on established scientiﬁc methods (such as pick analysis and interview techniques). The scientiﬁc merits of these methods have been discussed in detail in the literature (e.g., [20–22]) and hence this is not a focus of the present contribution. Instead, we discuss the ways that these methods are combined in the RBT procedure, and exemplify the methods and their combination by presenting results when the procedure was implemented in a residential area in Sweden.

2.1. Design of the Study The RBT procedure was tested in a pilot residential area in Sweden with the intention to test the method and to improve waste sorting behavior. The first step was to identify the prevailing sorting behavior, which was examined by measurement of waste composition by sampling and manual sorting (pick analysis). In the second step, factors that hindered correct sorting were identified using two types of interviews with the inhabitants: semi-structured and structured interviews. Although it is possible that these hindrances can be identified by other means, such as literature surveys, interviews have the advantage in that they include the inhabitants in the identification of appropriate interventions which increases the probability that the chosen intervention will be effective. Next, based on the interview results, interventions in the waste collection system were designed and implemented. Finally, another pick analysis was conducted to investigate the effect of the interventions. The pilot area and methods used are presented in more detail below.

2.2. Description of the Pilot Area Used to Test the Procedure The pilot area that was selected for testing the procedure is not relevant to the procedure and, as mentioned above, the procedure can be implemented in any area that has waste sorting at the source. In spite of this, a description of the pilot area is provided since this is needed if the results presented here are to be compared with results from other areas. The pilot area that was selected is a residential area in the city of Borås, which lies in the southwest of Sweden. This area was chosen since the waste separation scheme was not functioning as well as in other parts of the city [23]. According to Statistics Sweden [24], the pilot area had 447 inhabitants with a diverse socio-demographic background who lived in nine apartment buildings (a total of 208 apartments). Approximately 67% of the inhabitants were born outside of Sweden and immigrated to Sweden more than three years before the procedure was implemented [24]. 31% of the inhabitants were aged from 25 to 44 years and 20% were 45 to 64 years old. It was mainly inhabitants from these age groups that were involved in the interviews in Step 2 of the procedure. Approximately 63% of the Recycling 2016, 1, 147–165 150 Recycling 2016, 1, 147–165 151 citizens in this area had low income (less than 15,600 Euro per year). In addition, approximately 58% subcategories, and fractions of the waste that was weighed during the pick analysis, as well as where it had upper secondary school or higher education, and 33% of the residents owned a car [24]. should be sorted, are shown in Table 2. The categories used in the pick analysis were chosen according The source separation system used in the city was established in 1991. It is based on sorting food to the municipality’s instructions for waste sorting, which are similar to those used in other Swedish waste in black bags and combustible waste in white bags (combustible waste is all household waste cities. Any other material than food waste in the black bags and combustible waste in white bags was which is neither food, recyclable, nor hazardous). The waste sorted in the black bags is intended for regarded as missorted fractions. production of biofuel and that in the white bags for combustion. The black and white plastic bags are distributed free of charge to all households. Both bags are supposed to be located in the kitchen Table 2. Waste fractions and categories used in the pick analyses. so that they are easily accessible. To maintain low costs for the logistics, both bags are collected at Sorting the same time and placed in a single container. They are subsequently separated in an optical sorting Category Subcategory Fractions Instruction machine at material recovery facilities. This type of optical sorting is not commonly used in Sweden. Food Food Leftover food, fruits, vegetables, etc. Black bag Recyclable materials, such as paper packaging, plastic packaging, metal packaging, glass packaging, Paper packaging Cardboard, paper packaging Recycling station and newsprints, should be sorted in a bring system (i.e., they should be taken to recycling stations). Plastic film packaging, foam plastic Packaging, Plastic packaging The nearest complete recycling station to the pilot area was one and a half kilometers away. However, packaging, dense plastic packaging newsprints newsprints and glass packaging could be sorted in separate containers that were located immediately Glass packaging Colored and clear glass packaging behind the apartment buildings. Other materials, such as bulky and hazardous wastes, should be Metal packaging Metal packaging delivered to one of the five recycling centers; the nearest one being located about two kilometers Newsprint, advertisements, paperbacks, Newsprint from the apartment buildings. Collection of the packaging materials at recycling stations, and bulky writing/drawing papers PET 2, Aluminum cans and waste and hazardous waste at recycling centers, is a common part of Swedish waste management Deposit bottles 1 system. In these systems, inhabitants should bring their waste to these places and sort them in the glass bottles with deposit appropriate containers. Diapers Diapers, pads White bag Combustible Textile fabrics Clothes, shoes, different textiles 2.3. Methods Wood, small non-packaging plastic, garden waste, wood, non-packaging paper, cat A combination of quantitative and qualitative methods is needed in the RBT procedure. These are Combustible waste sand, tissues, envelopes, Christmas cards, small baby tools, pens, cigarette butts, described below. vacuum cleaner bags, etc. Non-packaging plastic Big parts Recycling center 2.3.1. Waste Composition Study (Pick Analysis) Other Non-packaging metal Any metal parts Measuring the amount of different fractions in the waste streams (e.g., in the black and white Non-packaging glass Broken glass bags) and characterizing them is called pick analysis. The method used was developed by Dahlén et al. Other non-combustibles Ceramics, broken mug, bricks, etc. (2008) [20] and has been published as a manual by the Swedish Waste Management Association [25,26]. Medicines 3 Recycling center Pick analyses were conducted in November 2011, November 2013, and November 2015 (i.e., before and Batteries after interventions in the waste collection system). November was chosen since there are no special Hazardous Small electronics Mobile phones, clocks, battery chargers, etc. Light-bulbs days (such as holidays) in Sweden during this month. It can therefore be assumed that the waste Other hazardous Glues, chemicals, full sprays collected in this month was representative of ordinary waste generation during the year. The pilot 1 Deposit bottles should be collected in large supermarkets. 2 PET (polyethylene terephthalate) bottles are area had twenty-eight 660 liter wheeled bins for collection of the white bags (intended for combustible usually used for beverages. 3 Medicines should be collected in pharmacies. waste) and black bags (intended for food waste). The white and black bags are collected in the same bins (i.e., the bins are not separated into those dedicated for black or white bags). Seven of the bins The amounts of correctly and incorrectly sorted waste were analyzed using the MINITAB software were randomly selected each week in November, giving a total of 28 samples. Table 1 shows the details (Minitab Ltd., Coventry, UK) to determine the average amounts of missorted materials in kg per of sampling for these two pick analyses. The size and number of samples used in this study ensure the household per week, as well as the statistical relevance of any changes in sorting behavior due to the statistical certainty of the results according to the description of the method [20]. interventions. The results from the first pick analysis revealed the prevailing waste sorting behavior in the pilot area. The results from the second and third pick analyses, which were conducted after each of Table 1. Details of sampling for pick analysis. the two interventions, were compared with each other and with the results from the first pick analysis using the two sample t-test with 95% confidence interval. Hence, the second and third pick analyses Date Unit of Sample No. of Samples per Week Total Samples Total Weight (kg) quantitatively assessed the effect of the interventions. November 2011 Waste in a 660 liter bin 7 28 1732.9 November 2013 Waste in a 660 liter bin 7 28 1699.0 November 2015 Waste in a 660 liter bin 7 28 1592.5 2.3.2. Interview The results from the first pick analysis were used to design interview templates. The goals of the The samples were collected for the pick analysis, and the rest of the waste in the area was weighed interviews were to identify hindrances for correct waste sorting and the changes (interventions) in the each week in order to estimate the total waste that was generated. Waste characterization and weighing waste collection system that would enable and motivate improved sorting. To do this, two types of of waste fractions in the samples were conducted separately for the white and black bags. The weight interviews were designed: semi-structured interviews and structured interviews. They are described of the plastic bags themselves (white and black) was excluded from the pick analysis. Categories, in detail below. Recycling 2016, 1, 147–165 151 subcategories, and fractions of the waste that was weighed during the pick analysis, as well as where it should be sorted, are shown in Table 2. The categories used in the pick analysis were chosen according to the municipality’s instructions for waste sorting, which are similar to those used in other Swedish cities. Any other material than food waste in the black bags and combustible waste in white bags was regarded as missorted fractions.

Table 2. Waste fractions and categories used in the pick analyses.

Sorting Category Subcategory Fractions Instruction Food Food Leftover food, fruits, vegetables, etc. Black bag Paper packaging Cardboard, paper packaging Recycling station Plastic ﬁlm packaging, foam plastic Packaging, Plastic packaging packaging, dense plastic packaging newsprints Glass packaging Colored and clear glass packaging Metal packaging Metal packaging Newsprint, advertisements, paperbacks, Newsprint writing/drawing papers PET 2, Aluminum cans and Deposit bottles 1 glass bottles with deposit Diapers Diapers, pads White bag Combustible Textile fabrics Clothes, shoes, different textiles Wood, small non-packaging plastic, garden waste, wood, non-packaging paper, cat Combustible waste sand, tissues, envelopes, Christmas cards, small baby tools, pens, cigarette butts, vacuum cleaner bags, etc. Non-packaging plastic Big parts Recycling center Other Non-packaging metal Any metal parts Non-packaging glass Broken glass Other non-combustibles Ceramics, broken mug, bricks, etc. Medicines 3 Recycling center Batteries Hazardous Small electronics Mobile phones, clocks, battery chargers, etc. Light-bulbs Other hazardous Glues, chemicals, full sprays 1 Deposit bottles should be collected in large supermarkets. 2 PET (polyethylene terephthalate) bottles are usually used for beverages. 3 Medicines should be collected in pharmacies.

The amounts of correctly and incorrectly sorted waste were analyzed using the MINITAB software (Minitab Ltd., Coventry, UK) to determine the average amounts of missorted materials in kg per household per week, as well as the statistical relevance of any changes in sorting behavior due to the interventions. The results from the first pick analysis revealed the prevailing waste sorting behavior in the pilot area. The results from the second and third pick analyses, which were conducted after each of the two interventions, were compared with each other and with the results from the first pick analysis using the two sample t-test with 95% confidence interval. Hence, the second and third pick analyses quantitatively assessed the effect of the interventions.

2.3.2. Interview The results from the first pick analysis were used to design interview templates. The goals of the interviews were to identify hindrances for correct waste sorting and the changes (interventions) in the waste collection system that would enable and motivate improved sorting. To do this, two types of interviews were designed: semi-structured interviews and structured interviews. They are described in detail below. Recycling 2016, 1, 147–165 152 Recycling 2016, 1, 147–165 153 Recycling 2016, 1, 147-165 153 Recycling 2016, 1, 147-165 153 Semi-Structured Interview Qualitative interviews, such as semi-structured interviews, are usually performed to obtain deep insight into how people understand a phenomena, and therefore the number of interviewees is not important [21,22]. Semi-structured interviews are used to gather information about peoples’ situations, to collect statements about their opinions, and to explore their motivation and experiences [27]. The initial aim of the semi-structured interviews was to understand why inhabitants sort their waste as shown by the pick analysis, if they think that they need to improve their sorting behavior, and, if so, what they think is needed to improve their sorting. To do this, eight interviewees who were comfortable with participating in the research interview were identified. This was done by sending an invitation letter to all inhabitants as well as asking inhabitants via a contact person. The interviews were conducted three to five months after the first pick analysis (over a two month period) at the University of Borås, and lasted between 50 and 65 min. A practical exercise for sorting waste was also conducted Figure 1. Flowchart of the questions in the structured interview. during the interview , asking the interviewees how they sort different kinds of waste fractions in Figure 1. Flowchart of the questions in the structured interview. (i.e. Figure 1. Flowchart of the questions in the structured interview. four alternatives: white bags, black bags, recycling stations, or recycling centers). This was designed to 2.3.3. Interventions in the Pilot Study 2.3.3. Interventions in the Pilot Study get a better understanding of their knowledge regarding the sorting of waste. The interviewees were 2.3.3. Interventions in the Pilot Study As discussed below, three interventions were designed. These interventions were chosen since, asked to describe how they felt about sorting their waste and the waste management system in the city, As discussed below, three interventions were designed. These interventions were chosen since, basedAs on discussed the results below, of the three interviews, interventions it was hypoth were deesizedsigned. that These they interventions would lead to were improvements chosen since, in how they understood the need for sorting and the system, and what is required for better participation basedthebasedon collection theon the results resultssystem. of of the This the interviews, hypothesisinterviews, it wasit was was tested hypothesizedhypoth in theesized pilot that that area. they they The would wouldinterventions lead lead to toimprovements are improvements described inin in in the system. The interviewees were free to discuss their ideas and comments. All the interviews theSection collectionthe collection 3.1.3 system. to system. clarify This Thistheir hypothesis hypothesis relation to was wasthe testedinterview tested inin theresults. pilot area. area. The The interventions interventions are are described described in in were taped and transcribed. They were subsequently analyzed to identify key words, paragraphs, SectionSection3.1.3 3.1.3to clarifyto clarify their their relation relation to to the the interview interview results.results. or themes. The results obtained from these semi-structured interviews were used to design a short 3. Results and Discussion 3. Results3. Results and and Discussion Discussion (less than five minutes), face-to-face structured interview. Implementation of the RBT procedure in the pilot area has been completed. The results obtainedImplementationImplementation from the procedure of theof the RBT RBTare procedure discussed procedure in below. the in pilotthe The pilot area way hasarea that been hasthis completed.procedurebeen completed. can The be results generalizedThe results obtained Structured Interview fromforobtained the any procedure source from separationthe are procedure discussed system are below. isdiscussed also The discussed. way below. that The this way procedure that this can procedure be generalized can be generalized for any source The semi-structured interviews identified the interventions that some of the inhabitants perceived separationfor any source system separation is also discussed. system is also discussed. were needed in the system. In order to ascertain if these perceptions were shared by a broad range of 3.1. Pilot Area inhabitants, a structured interview was designed. The aim of this short interview was not to perform 3.1.3.1. Pilot Pilot Area Area 3.1.1. First Pick Analysis—Evaluation of the Current Sorting Behavior a separate survey (via a questionnaire) but rather to complement the semi-structured interviews. 3.1.1.3.1.1. First First Pick Pick Analysis—Evaluation Analysis—Evaluation of of the the CurrentCurrent Sorting Behavior Behavior The structured interview was done by the same author who conducted the semi-structured interviews. Figure 2 shows the average waste composition per household per week in the black and The interviews were done three months after the semi-structured interviews and over a three day whiteFigureFigure bags.2 shows2 shows the the average average waste waste compositioncomposition per household household per per week week in inthe the black black and and period so that a larger number of people could participate. The time of the interviews was not whitewhite bags. bags. pre-planned, but instead the interviewer met inhabitants (not children) as they were conducting their daily activities. They were asked if it was convenient for them to participate in the interview and, if so, the interviewer introduced himself and asked the age of the respondent. The respondents were then asked to give short answers to the questions provided to them. There were two types of questions. One type had “yes” or “no” alternatives and the other type questioned how and why the inhabitants sorted their waste. If an answer about “how” specific waste should be sorted was incorrect, it was followed with a ”why” question in order to investigate the reason of the incorrect sorting. The questions were limited to sorting of plastic and paper packaging, since the pick analysis showed that these were the types of packaging that were missorted the most. To keep the interview as short as possible and in order to minimize the information on sorting that could be given to the interviewee (which could affect the outcome of the subsequent pick analysis), milk packaging was used as an example of paper packaging. Figure 1 shows the flowchart for the questions used in this structured interview.

Figure 2. Composition of waste (kg per household per week) in black and white bags before the interventions (2011). The striped regions show that waste that is intended for each bag (food waste in FigureFigure 2. Composition2. Composition of of waste waste (kg (kgper per householdhousehold per week) week) in in black black and and white white bags bags before before the the the black bag and combustible waste in the white bag). interventionsinterventions (2011). (2011). The The striped striped regions regions show show thatthat wastewaste that is is intended intended for for each each bag bag (food (food waste waste in in the black bag and combustible waste in the white bag). the black bag and combustible waste in the white bag). Recycling 2016, 1, 147–165 153 Recycling 2016, 1, 147-165 153 Recycling 2016, 1, 147-165 153

Figure 1. Flowchart of the questions in the structured interview. Figure 1. Flowchart of the questions in the structured interview. Figure 1. Flowchart of the questions in the structured interview. 2.3.3. Interventions in the Pilot Study 2.3.3. Interventions in the Pilot Study 2.3.3. Interventions in the Pilot Study As discussed below, three interventions were designed. These interventions were chosen since, As discussed below, three interventions were designed. These interventions were chosen since, basedAs on discussed the results below, of the three interviews, interventions it was hypoth were deesizedsigned. that These they interventions would lead to were improvements chosen since, in basedthebasedon collection theon the results resultssystem. of of the This the interviews, hypothesisinterviews, it wasit was was tested hypothesizedhypoth in theesized pilot that that area. they they The would wouldinterventions lead lead to toimprovements are improvements described inin in theSection collectionthe collection 3.1.3 system. to system. clarify This Thistheir hypothesis hypothesis relation to was wasthe testedinterview tested inin theresults. pilot area. area. The The interventions interventions are are described described in in SectionSection3.1.3 3.1.3to clarifyto clarify their their relation relation to to the the interview interview results.results. 3. Results and Discussion 3. Results3. Results and and Discussion Discussion Implementation of the RBT procedure in the pilot area has been completed. The results obtainedImplementationImplementation from the procedure of theof the RBT RBTare procedure discussed procedure in below. the in pilotthe The pilot area way hasarea that been hasthis completed.procedurebeen completed. can The be results generalizedThe results obtained fromforobtained the any procedure source from separationthe are procedure discussed system are below. isdiscussed also The discussed. way below. that The this way procedure that this can procedure be generalized can be generalized for any source separationfor any source system separation is also discussed. system is also discussed. 3.1. Pilot Area 3.1.3.1. Pilot Pilot Area Area 3.1.1. First Pick Analysis—Evaluation of the Current Sorting Behavior 3.1.1.3.1.1. First First Pick Pick Analysis—Evaluation Analysis—Evaluation of of the the CurrentCurrent Sorting Behavior Behavior Figure 2 shows the average waste composition per household per week in the black and whiteFigureFigure bags.2 shows2 shows the the average average waste waste compositioncomposition per household household per per week week in inthe the black black and and whitewhite bags. bags.

Figure 2. Composition of waste (kg per household per week) in black and white bags before the interventions (2011). The striped regions show that waste that is intended for each bag (food waste in FigureFigure 2. Composition2. Composition of of waste waste (kg (kgper per householdhousehold per week) week) in in black black and and white white bags bags before before the the the black bag and combustible waste in the white bag). interventionsinterventions (2011). (2011). The The striped striped regions regions show show thatthat wastewaste that is is intended intended for for each each bag bag (food (food waste waste in in the black bag and combustible waste in the white bag). the black bag and combustible waste in the white bag). Recycling 20162016,, 11,, 147-165 147–165 154 Recycling 2016, 1, 147–165 155

The total waste in the black bags, intended for food waste, was 2.8 kg per household per week (kg/hh/w)The total whereas waste in in white the black bags bags, (intended intended for comb for foodustible waste, waste), was it 2.8 was kg 4.5 per kg/hh/w. household The per striped week Table 3. Interviewees’ backgrounds. (kg/hh/w)areas in Figure whereas 2 show in white the part bags of (intended the waste for that combustible was sorted waste), correctly. it was Seventy-one 4.5 kg/hh/w. percent The striped of the a areaswaste in in Figure the black2 show bags the and part 29% of the of waste the waste that was in sortedthe white correctly. bags Seventy-onewas sorted percentcorrectly. of theThe waste total Person Sex Age Arrival Date Years Lived in the City Occupation Family Members inmissorted the black ratio, bags obtained and 29% ofby thesummi wasteng in the the missorted white bags fractions was sorted in both correctly. bags and The dividing total missorted the sum ratio, by 1 F 25 2007 2 No 1 obtainedthe total bywaste summing in both the missorted bags, was fractions 55%, inwhich both bagsis an and indicator dividing thethat sum should by the be total decreased. waste in 2 F 39 1990 15 Yes 5 3 F 50 1992 18 Part time 1 bothThe dominating bags, was 55%, waste which fraction is an in indicator the white that bags should was bemissorted decreased. packaging The dominating and newsprints, waste fraction and was in 4 F 38 1997 10 Yes 4 the1.9 kg/hh/w white bags (43%). was missortedThe second packaging largest missorted and newsprints, fraction andin the was white 1.9 kg/hh/wbags was (43%).food waste The second (26%), 5 F 40 1985 7 Yes 1 largestwhich was missorted often the fraction food that in the remained white bagsin the was packaging food waste when (26%), it was which discarded. was often This type the food of waste that 6 F 45 1997 16 Yes 2 7 M 40 2006 5 Part time 1 remainedis illustrated in thein Figure packaging 3. when it was discarded. This type of waste is illustrated in Figure 3. 8 F 76 Born in Sweden 57 Retired 1 a Year of immigration to Sweden.

Table 3. Interviewees’ backgrounds.

Person Sex Age Arrival Date a Years Lived in the City Occupation Family Members 1 F 25 2007 2 No 1 2 F 39 1990 15 Yes 5 3 F 50 1992 18 Part time 1 4 F 38 1997 10 Yes 4 5 F 40 1985 7 Yes 1 6 F 45 1997 16 Yes 2 7 M 40 2006 5 Part time 1 8 F 76 Born in Sweden 57 Retired 1 a Year of immigration to Sweden.

Most of the respondents had moved to the city from different countries, bringing with them different experiences of how to handle household waste. There was no organized source separation of waste in some of the immigrants’ home countries. For example, one interviewee (#6) had used food waste as animal feed or fertilizer. These people had to learn how to participate in a new waste management system. In addition, there was a large variation in the motivation for participating in source separation schemes. Some of the reasons that were given for participating were resource conservation, respect for regulations, they felt that their participation was a valuable part of the waste treatment system, keeping a clean environment, their knowledge of the post-sorting processes, social norms, and that their participation gave them a sense of satisfaction. One interviewee (#1) stated: “This is the regulation for sorting the waste and I should respect it. This is what people usually do in Sweden”. This means that she found waste separation as a norm in Swedish society and that this social norm can be a motivation to participate. Another interviewee (#8) said that “In the beginning, I didn’t think too much about the environment. I found out that we have to sort our waste from some information leaflet. Knowing how useful it is for the environment and that I can contribute to the next generation, my children and grandchildren whom I love, means a lot to me”. She continued, “At the moment, I cannot do much to protect my environment. This little job, sorting my waste, hopefully can be my part for protect environment for my children and grandchildren”. Knowing what happens with the sorted waste was also a motivating factor. As expressed by one person (#6): “I always sort glass since I have learned that glass is the material that you can recycle millions of times and it motivates me to sort it.” Similarly, another interviewee (#1), who had lived in Sweden for just four years, said: “When I see they produce biogas from food waste that I sort in the black bags and drive the buses, I understand my job is valuable”. One respondent (#7) stated: “I do this because I want to keep my surroundings clean”. According to Thøgersen (1994) [13], these motivational factors cannot be the only reasons for correct source separation by the inhabitants. They also need situational factors such as easy access to facilities as well as knowledge of how to separate at the source. One of the aims of the semi-structured interviews was to investigate what the inhabitants perceive is needed to improve participation in source separation. Hence, once the motivational factors that influence the inhabitants sorting behavior were known, the questions focused on what actions were required from the inhabitants. This was the reason for the practical exercise (explained in the Methods section) given during the interviews, which was conducted to reveal if the interviewees knew how to correctly sort different fractions of waste. After sorting each material, whether it was correct or incorrect, they were asked to explain the reason for their choice of sorting. Most of them sorted food waste in the black bags, which was correct. Also, batteries were sorted correctly because the interviewees perceived them as being environmentally harmful. However, fractions such as newspapers, plastic-, paper-, metal-, and glass-packaging, etc. were often discarded in the white bags rather than being sorted and taken to recycling stations. This is in agreement with the results of the pick analysis, where about 43% of the waste in the white bags was packaging. There were some fractions, such as light bulbs, that the interviewees knew should not be in the white bags, but they were not certain where they should sort it. They were also uncertain of how Recycling 2016, 1, 147–165 156 Recycling 2016, 1, 147–165 157 to sort other fractions such as ceramics and non-packaging glass. These findings are in agreement with communication between friends, neighbors, family members, and colleagues. However, there is no the studies of Henriksson et al. (2010) [28] who also found that people are uncertain about how to sort guarantee that this informal information is correct. Some of the incorrect sorting of the respondents certain waste fractions correctly [28]. was due to incorrect informal information. The type of information and the time of communication are The results discussed above indicate that lack of information and knowledge may hinder the also important. For example, some interviewees sorted diapers in the black bags and they claimed that correct sorting of waste. The interviews also revealed that distance to the recycling stations was there was a sticker on the trash cans for food waste which showed that diapers should be sorted in perceived as a hindrance. For example, one interviewee (#5) did not know that plastic bags such as these bags. These claims were, in fact, correct. The system for handling the black bags was changed potato chips packaging should be sorted in recycling stations. Further, even if she had been aware of from composting to biogas production in 2006, but the information (stickers) from the 1990s (for how to correctly sort the waste, the distance that she needed to take the waste to the recycling station composting) was still on the trash cans. This indicates that simple information, such as stickers on trash was important. She believed that improved sorting would be facilitated if there was a recycling station cans, which is accessible at the time of disposal, has a large influence on the sorting behavior. Hence, close to her home, preferably in the basement of the apartment building. This was also expressed by information should be distributed at the appropriate time, and it should be continuously available or another person (#6) who put paper and plastic packaging in the white bags: “I know that I should at least available at the time and place where waste is discarded. separate them, but we don’t have a recycling station close to home”. In this case, the interviewee The main findings from the semi-structured interviews are: had the knowledge but still chose not to go to the recycling station. She continued: “ ... my child asked me why we don’t sort the milk packaging. My answer was that we have no container close The interviewees were willing to sort the waste. ‚ to our home”. Some of the interviewees said that they were aware that they should sort fractions The interviewees had different motivations to sort the waste. ‚ such as metal packaging, but they did not do so since the recycling station was two km away and Decreasing the distance to collection points for recyclables can be a relevant intervention. they did not have access to a car. Even one interviewee (#8), who sorted all fractions correctly, stated: ‚ There was a lack of knowledge of how to sort the waste. “I am able to walk to recycling station two km away to sort my waste now. I am thinking, after some ‚ Designing different types of information that is communicated at the appropriate time and place, years, when walking is hard for me, I may put all packaging in the white bags instead, because of this ‚ via formal and informal channels, may also be a relevant intervention. distance”. Hence, it was concluded that decreasing the distance to the recycling station was a relevant intervention. Other studies have also revealed that distance to sorting facilities is a significant factor These findings are in agreement with the Thøgersen, Barr, and Tucker models [13,14,17] which for participation in recycling systems [9,10,29]. show that inhabitants’ intention to recycle should be accompanied with other factors in order to form After moving to Sweden, the respondents learned about waste management in different ways. the recycling behavior. It was not primarily through information leaflets, but by talking to and observing colleagues, neighbors, and family members. Most of the respondents could not remember whether they had received any Structured Interview written information. Only one interviewee (#8), who was born in Sweden, indicated that she had The semi-structured interviews showed that (1) distance to recycling stations and (2) different received written information 10 years ago. It is possible that, even though information may have been types of information about how to sort waste, including the communication of this information, may distributed, several of the interviewees may not have been able to read it due to language barriers or be relevant interventions for improving waste source separation in this pilot area. To validate the that it was distributed at an inappropriate time (e.g., when they were focused on adjusting to a new relevance of this hypothesis for a larger number of households, a structured and short interview home, city, and country). All interviewees, apart from two respondents (#5, #8), claimed that language template was designed and conducted with 50 people (20 male and 30 female). All of the interviewees was still a barrier when seeking information, in spite of the fact that they had lived in Sweden for more were between 20 and 50 years old. These 50 people constituted 25% of the population in this age range than 15 years. In addition, five of the interviewees were illiterate when arriving in Sweden. If the in this area (Swedish Statistics, 2012) and they were from 50 different households. This population information had been sent to them soon after their arrival it would not have been possible for them therefore covered about one quarter of the total number of households in the pilot area. Figure 4 shows understand it. This is why they gained knowledge through informal communication with friends, how respondents replied to the different questions in this short interview. neighbors, and colleagues. One person (#2) stated: “When I got a job five years ago I saw that they The majority of respondents, 42 out of 50 (84%), claimed that they sort their waste. We will sort every waste fractions in my workplace. I found that I have to do it at home too”. Another person refer to these people as recyclers. Only 8 out of 50 (16%) were non-recyclers. Among recyclers, only (#5) said: “When I moved to my apartment, I saw two bags in my kitchen, one for food waste and the 24% claimed that they sorted the packaging correctly at the recycling stations. The remainder, 76%, other combustible. It is obvious that I throw the food waste in black and all others, even packaging in threw packaging in the white bags, which is incorrect sorting. The main reason that was given for the the white when I don’t have any other information”. Lack of education can be another barrier. Many incorrect sorting was the distance to the recycling station (42%), and the other reason was that they of the interviewees felt that education is essential when trying to find the right information. They thought that their behavior was correct (i.e., that had incorrect information about how to sort this type also thought that the best place for learning is at school. When the information is insufficient it has a of packaging (33%)). It is not clear if the distance to the recycling station would be a barrier for the negative impact. Insufficient information can lead to incorrect sorting habits that can be difficult to latter group once they obtain the correct sorting information. Some respondents had other reasons change. One person (#2) said: “It is very important that one learns the right things from the beginning, such as: ”I don’t have a car” and “I have just a little amount waste”. The recyclers obtained information because it is very hard to change the wrong habits after a while.” Another interviewee emphasized from informal channels (from neighbors, friends, and their workplaces) more often (90%) than from that it is important to have information as early as possible in order to develop a correct habit. As formal channels (from the municipality). This shows that this area has a strong social network to expressed by one interviewee (#5): “Maybe sorting the garbage is hard in the beginning, but after a distribute this type of information. Three out of eight non-recyclers claimed that they did not sort due while you don’t have to think about it, you do it as a norm, as a habit. It’s part of my culture now. The to a lack of information. Difficulty and lack of motivation for sorting were the reasons for the other more information you can understand and make your own the faster it will become a routine for you”. five non-recyclers. Hence, the interviews revealed that information and knowledge about how and why to sort the The result of the short, structured interviews was consistent with the findings of the waste are important factors. If this is not communicated via formal authorities, it can be informal semi-structured interviews. Therefore, to improve source separation in the pilot area the following Recycling 2016, 1, 147–165 157 communication between friends, neighbors, family members, and colleagues. However, there is no guarantee that this informal information is correct. Some of the incorrect sorting of the respondents was due to incorrect informal information. The type of information and the time of communication are also important. For example, some interviewees sorted diapers in the black bags and they claimed that there was a sticker on the trash cans for food waste which showed that diapers should be sorted in these bags. These claims were, in fact, correct. The system for handling the black bags was changed from composting to biogas production in 2006, but the information (stickers) from the 1990s (for composting) was still on the trash cans. This indicates that simple information, such as stickers on trash cans, which is accessible at the time of disposal, has a large influence on the sorting behavior. Hence, information should be distributed at the appropriate time, and it should be continuously available or at least available at the time and place where waste is discarded. The main findings from the semi-structured interviews are:

The interviewees were willing to sort the waste. ‚ The interviewees had different motivations to sort the waste. ‚ Decreasing the distance to collection points for recyclables can be a relevant intervention. ‚ There was a lack of knowledge of how to sort the waste. ‚ Designing different types of information that is communicated at the appropriate time and place, ‚ via formal and informal channels, may also be a relevant intervention.

These ﬁndings are in agreement with the Thøgersen, Barr, and Tucker models [13,14,17] which show that inhabitants’ intention to recycle should be accompanied with other factors in order to form the recycling behavior.

Structured Interview The semi-structured interviews showed that (1) distance to recycling stations and (2) different types of information about how to sort waste, including the communication of this information, may be relevant interventions for improving waste source separation in this pilot area. To validate the relevance of this hypothesis for a larger number of households, a structured and short interview template was designed and conducted with 50 people (20 male and 30 female). All of the interviewees were between 20 and 50 years old. These 50 people constituted 25% of the population in this age range in this area (Swedish Statistics, 2012) and they were from 50 different households. This population therefore covered about one quarter of the total number of households in the pilot area. Figure 4 shows how respondents replied to the different questions in this short interview. The majority of respondents, 42 out of 50 (84%), claimed that they sort their waste. We will refer to these people as recyclers. Only 8 out of 50 (16%) were non-recyclers. Among recyclers, only 24% claimed that they sorted the packaging correctly at the recycling stations. The remainder, 76%, threw packaging in the white bags, which is incorrect sorting. The main reason that was given for the incorrect sorting was the distance to the recycling station (42%), and the other reason was that they thought that their behavior was correct (i.e., that had incorrect information about how to sort this type of packaging (33%)). It is not clear if the distance to the recycling station would be a barrier for the latter group once they obtain the correct sorting information. Some respondents had other reasons such as: ”I don’t have a car” and “I have just a little amount waste”. The recyclers obtained information from informal channels (from neighbors, friends, and their workplaces) more often (90%) than from formal channels (from the municipality). This shows that this area has a strong social network to distribute this type of information. Three out of eight non-recyclers claimed that they did not sort due to a lack of information. Difficulty and lack of motivation for sorting were the reasons for the other five non-recyclers. The result of the short, structured interviews was consistent with the findings of the semi-structured interviews. Therefore, to improve source separation in the pilot area the following Recycling 2016, 1, 147-165 157 guarantee that this informal information is correct. Some of the incorrect sorting of the respondents was due to incorrect informal information. The type of information and the time of communication are also important. For example, some interviewees sorted diapers in the black bags and they claimed that there was a sticker on the trash cans for food waste which showed that diapers should be sorted in these bags. These claims were, in fact, correct. The system for handling the black bags was changed from composting to biogas production in 2006, but the information (stickers) from the 1990s (for composting) was still on the trash cans. This indicates that simple information, such as stickers on trash cans, which is accessible at the time of disposal, has a large influence on the sorting behavior. Hence, information should be distributed at the appropriate time, and it should be continuously available or at least available at the time and place where waste is discarded. The main findings from the semi-structured interviews are: • The interviewees were willing to sort the waste. • The interviewees had different motivations to sort the waste. • Decreasing the distance to collection points for recyclables can be a relevant intervention. • There was a lack of knowledge of how to sort the waste. • Designing different types of information that is communicated at the appropriate time and place, via formal and informal channels, may also be a relevant intervention. These findings are in agreement with the Thøgersen, Barr, and Tucker models [13,14,17] which show that inhabitants’ intention to recycle should be accompanied with other factors in order to form the recycling behavior.

Structured Interview Recycling 2016, 1, 147-165 159 Recycling 2016, 1, 147-165 159 The semi-structured interviews showed that (1) distance to recycling stations and (2) different (c) in 2015, this decrease was 39% (Figure 5). Although there was an increase in the missorted food types of information about how to sort waste, including the communication of this information, (c) in 2015, this decrease was 39% (Figure 5). Although there was an increase in the missorted food Recycling 2016, 1, 147–165 158 wasteRecycling in2016 the, 1 ,white 147–165 bags between 2013 and 2015, there was an overall decrease between 2011 and159 may be relevant interventions for improving waste source separation in this pilot area. To validate waste in the white bags between 2013 and 2015, there was an overall decrease between 2011 and 2015. The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, the relevance of this hypothesis for a larger number of households, a structured and short interview 2015. The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, as is the decrease in food waste between 2011 and 2013. The increase in food waste between 2013 and templateinterventions was coulddesigned be relevant: and conducted (1) decreasing with the 50 distance people to(20 the male recycling and station30 female). and (2) All combining of the influenceas is the decrease sorting in food the black waste bags, between whereas 2011 interventionand 2013. The (b) increase was expected in food towaste influence between recycling 2013 and in 2015, and the subsequent decrease between 2011 and 2015 are not statistically significant. intervieweescorrect and appropriate were between information 20 and 50 with years effective old. Th communicationese 50 people constituted channels. 25% of the population the2015, environmental and the subsequent room and decrease in both bags.between Since 2011 the whiteand 2015 bags (intendedare not statistically for combustible significant. waste) The interventions did not lead to any statistically meaningful changes in the sorting of hazardous or in thisHence, age range the resultsin this obtainedarea (Swedish from theStatistics interviews, 2012)answered and they were the second from 50 research different question households. given containedThe interventions more recyclables did not lead than to theany black statistically bags (Figure meaningful1) intervention changes in (b) the was sorting expected of hazardous to influence or other waste, but the hazardous waste decreased by 68% in the white bags in 2015 compared to 2011. Thisin Section population “Objectives” therefore(i.e. covered, how about should one the quarter appropriate of the interventionstotal number of to households improve this in behaviorthe pilot theother sorting waste, in but the the white hazardous bags more waste than decreased in the black by 68 bags.% in the white bags in 2015 compared to 2011. area.be identified?). Figure 4 shows how respondents replied to the different questions in this short interview.

Figure 5. Comparison of waste sorting behavior before and after the interventions. The data shows Figure 5.5. Comparison of waste sorting behavior before and after thethe interventions.interventions. The data shows waste fractions in the white bags, which are intended for combustibles (kg/hh/w). waste fractions in the white bags, which areare intendedintended forfor combustiblescombustibles (kg/hh/w).(kg/hh/w). FigureFigure 4. 4. ResultsResults of of the the structured structured interviews interviews with with number number of of respondents respondents indicated indicated in in parentheses. parentheses. NodesNodes A to D representrepresent questions:questions: ( A(A)) How How did did you you get get information? information? (B )(B How) How do do you you sort sort plastic plastic and andmilk milk packaging? packaging? (C) Why?(C) Why? (D) ( DoD) youDo you want want to sort to sort if you if you get information?get information?

3.1.3. Effect of Interventions Based on the interview results, three interventions were implemented from 2012 to 2015. These are listed in Table 4.

Table 4. Interventions from 2012 to 2015.

Date Intervention (a) Placing new stickers for black bags trash cans which clearly show June 2012 that food waste should be sorted in the black bags. (b) Building a property close collection area for recyclables, called an environmental room, behind the apartment building to decrease the June 2013 distance to the recycling station to 50 meters. The environmental room enabled the sorting of packaging, batteries, small electronics, light bulbs, and clothes. (c) Communicating different types of information including: Installation of a picture on the containers for white and black bags showing where and how to sort this waste in the containers (2014). Figure 6. Comparison of waste sorting behavior before and after the interventions.interventions. The data shows 2014–2015 Sending written information, feedback and a “thank-you letter” Figure 6. Comparison of waste sorting behavior before and after the interventions. The data shows three times to all households in the pilot area (2014). waste fractions in the black bags, which are intended for food waste (kg/hh/w). waste fractions in the black bags, which are intended for food waste (kg/hh/w). Having informal dialogues about using the environmental room and on how to improve the sorting in black bags (2015). MissortedThe amount combustible of missorted waste packaging in black in bags the white(intended bags only decreased for food (33%) waste) between decreased 2011 and(by 44%) 2013, Missorted combustible waste in black bags (intended only for food waste) decreased (by 44%) bywhich intervention is when the(a), when property visible close information collection on was th establishede trash cans (interventionfor black bags (b)).was provided, After intervention and by a by intervention (a), when visible information on the trash cans for black bags was provided, and by a The second pick analysis was performed after implementing interventions (a) and (b), and the further(c) in 2015, 3% in this 2015 decrease (Figure was 6). Similarly, 39% (Figure missort5). Althoughed packaging there and was newsprint an increase decreased in the missorted between food2011 further 3% in 2015 (Figure 6). Similarly, missorted packaging and newsprint decreased between 2011 third pick analysis was conducted after intervention (c). The results are compared with those of waste in the white bags between 2013 and 2015, there was an overall decrease between 2011 and 2015. the first pick analysis (2011) in Figures 5 and 6. It should be noted that two interventions were The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, as is the performed between the first and second pick analyses. Intervention (a) was expected to primarily decrease in food waste between 2011 and 2013. The increase in food waste between 2013 and 2015, and Recycling 2016, 1, 147-165 159 Recycling 2016, 1, 147-165 159 (c) in 2015, this decrease was 39% (Figure 5). Although there was an increase in the missorted food (c) in 2015, this decrease was 39% (Figure 5). Although there was an increase in the missorted food wasteRecycling in2016 the, 1 ,white 147–165 bags between 2013 and 2015, there was an overall decrease between 2011 and159 waste in the white bags between 2013 and 2015, there was an overall decrease between 2011 and 2015. The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, 2015. The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, as is the decrease in food waste between 2011 and 2013. The increase in food waste between 2013 and influenceas is the decrease sorting in food the black waste bags, between whereas 2011 interventionand 2013. The (b) increase was expected in food towaste influence between recycling 2013 and in 2015, and the subsequent decrease between 2011 and 2015 are not statistically significant. the2015, environmental and the subsequent room and decrease in both bags.between Since 2011 the whiteand 2015 bags (intendedare not statistically for combustible significant. waste) The interventions did not lead to any statistically meaningful changes in the sorting of hazardous or containedThe interventions more recyclables did not lead than to theany black statistically bags (Figure meaningful1) intervention changes in (b) the was sorting expected of hazardous to influence or other waste, but the hazardous waste decreased by 68% in the white bags in 2015 compared to 2011. theother sorting waste, in but the the white hazardous bags more waste than decreased in the black by 68 bags.% in the white bags in 2015 compared to 2011.

Figure 6. Comparison of waste sorting behavior before and after the interventions.interventions. The data shows Figure 6. Comparison of waste sorting behavior before and after the interventions. The data shows waste fractions in the black bags, which are intended for food waste (kg/hh/w). waste fractions in the black bags, which are intended for food waste (kg/hh/w). MissortedThe amount combustible of missorted waste packaging in black in bags the white(intended bags only decreased for food (33%) waste) between decreased 2011 and(by 44%) 2013, Missorted combustible waste in black bags (intended only for food waste) decreased (by 44%) bywhich intervention is when the(a), when property visible close information collection on was th establishede trash cans (interventionfor black bags (b)).was provided, After intervention and by a by intervention (a), when visible information on the trash cans for black bags was provided, and by a further(c) in 2015, 3% in this 2015 decrease (Figure was 6). Similarly, 39% (Figure missort5). Althoughed packaging there and was newsprint an increase decreased in the missorted between food2011 further 3% in 2015 (Figure 6). Similarly, missorted packaging and newsprint decreased between 2011 waste in the white bags between 2013 and 2015, there was an overall decrease between 2011 and 2015. The decrease in packaging and newsprints between 2011 and 2015 is statistically significant, as is the decrease in food waste between 2011 and 2013. The increase in food waste between 2013 and 2015, and Recycling 2016, 1, 147–165 160 Recycling 2016, 1, 147–165 161 the subsequent decrease between 2011 and 2015 are not statistically significant. The interventions did Hence, the results obtained from the second and third pick analyses answered the third research not lead to any statistically meaningful changes in the sorting of hazardous or other waste, but the question given in Section “Objectives” (i.e., how should the effect of interventions be assessed?). hazardous waste decreased by 68% in the white bags in 2015 compared to 2011. Missorted combustible waste in black bags (intended only for food waste) decreased (by 44%) by 3.2. Recycling Behavior Transition (RBT) Procedure interventionRecycling 2016, 1 (a),, 147-165 when visible information on the trash cans for black bags was provided, and by160 a The aim of this research was to create, test, and evaluate a procedure which can improve recycling further 3% in 2015 (Figure 6). Similarly, missorted packaging and newsprint decreased between 2011 behavior. The procedure consisted of four steps: (1) first pick analysis; (2a) a few semi-structured andand 20152015 (52%). (52%). These These two two changes changes are statistically are statistically significant. significant. Similarly Similarly to the white to the bags, white there werebags, interviews; (2b) a larger number of structured interviews; (3) interventions based on findings from the nothere significant were no changessignificant in thechanges amounts in the of amou hazardousnts of orhazardous other forms or other of waste. forms of waste. interviews, and; (4) a pick analysis to assess the effect of interventions. The four steps, which included InIn summary,summary, thethe aimaim ofof thethe interventionsinterventions waswas toto decreasedecrease the missorting ratios in the pilot area. both quantitative and qualitative methods for data collection, were implemented and tested. FigureFigure 77 showsshows thethe missortedmissorted ratiosratios inin whitewhite bags,bags, blackblack bags,bags, andand inin the total waste generated in the Pick analysis is a quantitative waste characterization method that measures missorted waste in pilotpilot areaarea (i.e.(i.e.,, thethe averageaverage ofof thethe whitewhite andand blackblack bags).bags). a system. Structured interviews are also considered to be a form of quantitative method, whereas semi-structured interviews are a form of qualitative method; both forms of interview processes were used in this procedure. The combination of quantitative and qualitative methods (i.e., using a mixed method) gives the necessary breadth and depth to understand complex phenomena [30] such as recycling behavior. As discussed above, the first pick analysis quantifies the prevailing sorting behavior. In fact, this pick analysis identifies “How” inhabitants sort their waste without asking them. The output from the characterization of the collected waste is more reliable than using self-reported assessments of the participants in a source separation scheme. In a study of the accuracy of three surveys of waste attitude/behavior undertaken in Scotland and north-west England, P. Tucker and D. Speirs (2003) [8] concluded that the results of these surveys usually give an over-optimistic picture of recycling behavior. They also concluded that it is difficult to make actions, promotions, and campaigns for improvement of recycling behavior by using results from surveys [8]. In contrast, pick analysis measures the actual waste sorting behavior of inhabitants regardless of what they answer in surveys. The interventions that should be investigated are identified in Step 2. To do this one needs to Figure 7. Comparison of missorted ratios in weight percent (wt %) before and after the interventions. Figure 7. Comparison of missorted ratios in weight percent (wt %) before and after the interventions. understand “Why” inhabitants sort in the way that they do (as identified in Step 1), and “What” should The total missorted ratio decreased from 55% to 38% after interventions (a) and (b), and be done to motivate and enable them to improve it. To do this, semi-structured interviews were The total missorted ratio decreased from 55% to 38% after interventions (a) and (b), and decreased decreased further to 36% after intervention (c). Figure 7 also shows that the magnitude of the designed. A semi-structured interview can provide answers to questions that cannot be checked or further to 36% after intervention (c). Figure 7 also shows that the magnitude of the reduction in reduction in the black bags between 2013 and 2015 is larger than for the white bags. Although this measured in other ways. It is difficult to get an absolute answer during an interview, but it can provide the black bags between 2013 and 2015 is larger than for the white bags. Although this difference is difference is statistically insignificant, it indicates that intervention (c) may have had the desired clues and indications that guide the subsequent steps. Although the results of the pick analysis give statistically insignificant, it indicates that intervention (c) may have had the desired effect, since the effect, since the focus of the informal communication was to improve sorting in the black bags. relevant data of the sorting behavior, these data require interpretation via the qualitative interview (i.e., focusThe results of the show informal that communication interventions (a) was and to (b) improve decreased sorting the inmissorted the black ratio bags. significantly. The results showAlthough that the semi-structured interview). This interpretation is needed to understand the sorting behavior. interventionsthere was a continued (a) and (b) decrease decreased in thethe missortedmissorted ratioratio significantly.after intervention Although (c), it there cannot was be a concluded continued For example, in the study conducted in the pilot area, the results of the first pick analysis showed decreasethat this is in due the to missorted the intervention ratio after (the intervention decrease may (c), have it cannot continued be concluded even if intervention that this is due(c) had to thenot that packaging and newsprints dominated the missorted material. Hence, the pick analysis revealed interventionbeen implemented). (the decrease may have continued even if intervention (c) had not been implemented). that one should focus on improving the sorting of packaging and newsprints waste. There are many FigureFigure 77shows shows that that intervention intervention (c) (c) had had less less impact impact compared compared to interventions to interventions (a) and (a) (b). and There (b). possible interventions that could have been studied to achieve this goal, but the interviews identified mayThere be may several be several reasons reasons for this. for One this. is One that is 56 that of the56 of households the households hadinhabitants had inhabitants that movedthat moved into that improved information and decreasing the distance to recycling stations were the interventions theinto pilot the pilot area area during during the project,the project, of whichof which 33 33 had had inhabitants inhabitants that that came came in in 2015. 2015. Since Since nono formalformal that the households perceived as being the most appropriate. The interviews also give the users, who informationinformation regardingregarding thethe useuse ofof thethe environmentalenvironmental roomroom waswas directeddirected towardstowards thethe householdshouseholds afterafter are influenced by the interventions, a central role when developing the waste management system. 2013,2013, manymany ofof thethe newnew inhabitantsinhabitants onlyonly hadhad access to information via informal channels. In addition, In a study in Madeline Island, Newenhouse and Schmit (2000) [31] found that qualitative newnew inhabitants—especiallyinhabitants—especially those those that that moved moved in duringin during 2015—may 2015—may not havenot have had timehad totime build to abuild social a approaches can add value to waste characterization studies in order to discover potential solutions networksocial network that is that necessary is necessary for informal for informal communication communication between between households. households. This This means means that that it isit to change recycling behavior. They discussed the fact that the interviews engage inhabitants in the possibleis possible that that 27% 27% (56 of(56 208) of of208) the of households the househol didds not did have not access have toaccess the correct to the information. correct information. Another process which, according to them, has a positive impact. The results of semi-structured interviews reasonAnother may reason be the may insufficient be the capacity insufficient of containers capacity in theof environmentalcontainers in room.the environmental The collection rateroom. for (Step 2a) can identify relevant interventions for improving the system, but it is important to have the theThe collected collection packaging rate for the in thecollected environmental packaging room in the was environmental therefore limited. room Observation was therefore of this limited. room opinion of a larger part of the population. The structured short interview (Step 2b) ascertains if the showedObservation that theof this packaging room showed containers that were the oftenpackaging overflowing containers on thewere collection often overflowing days. This usuallyon the selected interventions are relevant for many of the inhabitants. Through these two steps, (2a) and (2b), leadscollection to poor days. sorting This behaviorusually leads which to results poor insorting a mess behavior in the room, which which results could in a negatively mess in the affect room, the interview results will be both qualitative and quantitative. Another reason for including Step 2b in the inhabitants’which could participation negatively affect in the the sorting inhabitants’ scheme. participation It is also not in clear the howsorting the scheme. households It is perceivedalso not clear the procedure is to involve as many inhabitants as possible. In the pilot study, the result of the structured newhow informationthe households in the perceived form of written the new information, information stickers, in the thank-youform of written letters, information, and feedback. stickers, short interviews was in agreement with those of the semi-structured interviews. If they had not been thank-you letters, and feedback. Hence, the results obtained from the second and third pick analyses answered the third research question given in Section “Objectives” (i.e., how should the effect of interventions be assessed?).

Hence, the results obtained from the second and third pick analyses answered the third research question given in Section “Objectives” (i.e., how should the effect of interventions be assessed?).

3.2. Recycling Behavior Transition (RBT) Procedure The aim of this research was to create, test, and evaluate a procedure which can improve recycling behavior. The procedure consisted of four steps: (1) first pick analysis; (2a) a few semi-structured interviews; (2b) a larger number of structured interviews; (3) interventions based on findings from the interviews, and; (4) a pick analysis to assess the effect of interventions. The four steps, which included both quantitative and qualitative methods for data collection, were implemented and tested. Pick analysis is a quantitative waste characterization method that measures missorted waste in a system. Structured interviews are also considered to be a form of quantitative method, whereas semi-structured interviews are a form of qualitative method; both forms of interview processes were used in this procedure. The combination of quantitative and qualitative methods (i.e., using a mixed method) gives the necessary breadth and depth to understand complex phenomena [30] such as recycling behavior. As discussed above, the first pick analysis quantifies the prevailing sorting behavior. In fact, this pick analysis identifies “How” inhabitants sort their waste without asking them. The output from the characterization of the collected waste is more reliable than using self-reported assessments of the participants in a source separation scheme. In a study of the accuracy of three surveys of waste attitude/behavior undertaken in Scotland and north-west England, P. Tucker and D. Speirs (2003) [8] concluded that the results of these surveys usually give an over-optimistic picture of recycling behavior. They also concluded that it is difficult to make actions, promotions, and campaigns for improvement of recycling behavior by using results from surveys [8]. In contrast, pick analysis measures the actual waste sorting behavior of inhabitants regardless of what they answer in surveys. The interventions that should be investigated are identified in Step 2. To do this one needs to understand “Why” inhabitants sort in the way that they do (as identified in Step 1), and “What” should be done to motivate and enable them to improve it. To do this, semi-structured interviews were designed. A semi-structured interview can provide answers to questions that cannot be checked or measured in other ways. It is difficult to get an absolute answer during an interview, but it can provide clues and indications that guide the subsequent steps. Although the results of the pick analysis give relevant data of the sorting behavior, these data require interpretation via the qualitative interview (i.e., the semi-structured interview). This interpretation is needed to understand the sorting behavior. For example, in the study conducted in the pilot area, the results of the first pick analysis showed that packaging and newsprints dominated the missorted material. Hence, the pick analysis revealed that one should focus on improving the sorting of packaging and newsprints waste. There are many possible interventions that could have been studied to achieve this goal, but the interviews identified that improved information and decreasing the distance to recycling stations were the interventions that the households perceived as being the most appropriate. The interviews also give the users, who are influenced by the interventions, a central role when developing the waste management system. In a study in Madeline Island, Newenhouse and Schmit (2000) [31] found that qualitative approaches can add value to waste characterization studies in order to discover potential solutions to change recycling behavior. They discussed the fact that the interviews engage inhabitants in the process which, according to them, has a positive impact. The results of semi-structured interviews (Step 2a) can identify relevant interventions for improving the system, but it is important to have the opinion of a larger part of the population. The structured short interview (Step 2b) ascertains if the selected interventions are relevant for many of the inhabitants. Through these two steps, (2a) and (2b), interview results will be both qualitative and quantitative. Another reason for including Step 2b in the procedure is to involve as many inhabitants as possible. In the pilot study, the result of the structured short interviews was in agreement with those of the semi-structured interviews. If they had not been Recycling 2016, 1, 147–165 162 Recycling 2016, 1, 147–165 163 in agreement, then it would have been necessary to re-analyze the semi-structured interviews or to The RBT procedure focuses on user involvement in waste management systems. When the ‚ conduct more semi-structured interviews to identify other possible interventions. intervention is identified from the needs of the users, there is a higher probability of success. Step 4, the second and third pick analyses, assesses the effect of the interventions. These analyses The continued improvement of source separation systems results in the collection of recyclable ‚ provide data on the actual recycling behavior after the interventions, which can be compared with the materials in larger quantities and of better quality, and consequently help in the development of a Recycling 2016, 1, 147-165 162 first analysis in order to see if the interventions were effective or not. circular economy and sustainable development. The RBTRBT procedureprocedure is is summarized summarized in in Figure Figure8 together 8 together with with the the methods methods for datafor data collection. collection. The There are also several potential weaknesses of the RBT procedure that need to be recognized and, procedureThe procedure was calledwas called recycling recycling behavior behavior transition transi sincetion a sourcesince a separation source separation system can system be improved can be where possible, removed in future studies. byimproved a transition by a oftransition the behavior of theof behavior the inhabitants of the inha inbitants sorting in their sorting waste. their The wa dashedste. The line dashed in Figure line in8 showsFigure that8 shows the procedure that the procedure can be repeated can be to repeated enable continued to enable development continued development of the waste managementof the waste system. After doing the last step (i.e., the second or third pick analysis), there are two scenarios: (1) The The RBT procedure requires careful design in both time and implementation, and one often needs management system. After doing the last step (i.e., the second or third pick analysis), there are two ‚ interventionsscenarios: (1) wereThe interventions effective: In this were case effective: the result In ofthis second case the pick result analysis of second can be pick the basis analysis for finding can be to wait for a specific time period for implementing interventions so that the pick analyses that are newthe basis interventions for finding and furthernew interventions improvement, and including further aimprovement, new round of interviewsincluding a with new inhabitants; round of performed before and after the interventions can be compared to each other. In addition, since the (2)interviews The interventions with inhabitants; were not (2) effective.The interventions In this casewere the not factors effective. which In this prevented case the thefactors success which of procedure requires a long time, the effectiveness of the interventions is reduced if many residents theprevented interventions the success need of to the be interventions identified (e.g., need by ato new be identified series of interviews).(e.g., by a new It isseries possible of interviews). to choose move into and out of the neighborhood during the time between successive pick analyses. another qualitative approach other than semi-structured personal interviews to communicate with Interviews with the inhabitants to identify proper interventions are good since it identifies It is possible to choose another qualitative approach other than semi-structured personal interviews ‚ inhabitantsto communicate (e.g., focuswith inhabitants group or telephone (e.g., focus interviews group or or telephone combining interviews interviews or with combining observations). interviews An interventions that are relevant to the inhabitants and also increases the involvement the inhabitants importantwith observations). element with An important Step 2 is to element understand with the Step needs 2 is ofto theunderstand inhabitants the for needs effective of the participation inhabitants in the waste management system. However, these interviews may lead to improved sorting infor source effective separation. participation in source separation. behavior (which is good in many perspectives) which means that the effect of the intervention is difficult to isolate. Pick analyses, which are an integral part of the RBT procedure, require large sample sizes to be ‚ statistically meaningful. This increases the cost of the procedure.

4. Conclusions and Outlook The waste sorting behavior of households can be quantitatively measured and evaluated by waste sampling and pick analysis (i.e., waste characterization by manual sorting). Thereafter, appropriate interventions for improvement of the sorting can be identified through interviews in two steps: (a) semi-structured interviews (with a few households) where the questions are based on the pick analysis results; and (b) short, structured interviews (with a large population of households) based on the results of the semi-structured interviews. Once the interventions have been implemented, their effect can be evaluated by a new pick analysis. The missorting ratios before and after the interventions can thereby be quantified and compared. This four step recycling behavior transition procedure enables continued improvement of waste sorting behavior in any household waste collection system that includes waste sorting at the source. Figure 8. The Recycling Behavior Transition (RBT) procedure for continuous improvement of source Figure 8. The Recycling Behavior Transition (RBT) procedure for continuous improvement of source The combination of quantitative and qualitative methods allows the RBT procedure to evaluate separation and waste collection systems. separation and waste collection systems. the actual sorting behavior, to understand the reasons for this behavior, to identify the needs for Recycling behavior is a complex phenomenon. The RBT procedure showed that it is possible to improvement, to design relevant interventions, and to assess the effect of the interventions. The RBT Recycling behavior is a complex phenomenon. The RBT procedure showed that it is possible to improve this behavior with a simple and accurate procedure. There are several advantages with the procedure can be designed and applied in any source separation system irrespective of local factors improve this behavior with a simple and accurate procedure. There are several advantages with the RBT procedure, and these are listed and commented upon below. such as culture and social-economic backgrounds. RBT procedure, and these are listed and commented upon below. Further development of the RBT procedure is needed. Repeated use of the procedure in various • The combination of quantitative and qualitative methods provides reliable and useful results to types of waste separation systems will identify more strengths and weaknesses of the procedure. Theimplement combination the appropriate of quantitative interventions. and qualitative methods provides reliable and useful results to The RBT procedure can also be combined with other methods such as life cycle assessment and cost ‚ • implementThe RBT theprocedure appropriate can interventions.be applied in any source separation system in any location, analysis for a wider system analysis of a waste management system. independent of local circumstances. The procedure can be used to identify the relevant The RBT procedure can be applied in any source separation system in any location, independent ‚ intervention in other cultures, socio-economic backgrounds, and other source separation Author Contributions: Kamran Rousta has taken part in planning the study, conducting the experimental of local circumstances. The procedure can be used to identify the relevant intervention in other work, analyzing the results, and has been the principal writer of the manuscript. Lisa Dahlén and Kim Bolton, schemes. cultures, socio-economic backgrounds, and other source separation schemes. supervisors, have taken part in evaluating the results and writing the manuscript. • The RBT procedure quantifies the effect of the interventions. Actors responsible for planning The RBT procedure quantifies the effect of the interventions. Actors responsible for planning Conflicts of Interest: The authors declare no conflict of interest. ‚ waste management may hesitate to apply interventions because they do not know how to waste management may hesitate to apply interventions because they do not know how to identify identify them and how to measure their effect. This procedure can help them not only to them and how to measure their effect. This procedure can help them not only to improve their improve their system, but also to control the quality of the system. system, but also to control the quality of the system. • The RBT procedure focuses on user involvement in waste management systems. When the intervention is identified from the needs of the users, there is a higher probability of success. • The continued improvement of source separation systems results in the collection of recyclable materials in larger quantities and of better quality, and consequently help in the development of a circular economy and sustainable development. There are also several potential weaknesses of the RBT procedure that need to be recognized and, where possible, removed in future studies. Recycling 2016, 1, 147–165 163

The RBT procedure focuses on user involvement in waste management systems. When the ‚ intervention is identiﬁed from the needs of the users, there is a higher probability of success. The continued improvement of source separation systems results in the collection of recyclable ‚ materials in larger quantities and of better quality, and consequently help in the development of a circular economy and sustainable development.

There are also several potential weaknesses of the RBT procedure that need to be recognized and, where possible, removed in future studies.

The RBT procedure requires careful design in both time and implementation, and one often needs ‚ to wait for a specific time period for implementing interventions so that the pick analyses that are performed before and after the interventions can be compared to each other. In addition, since the procedure requires a long time, the effectiveness of the interventions is reduced if many residents move into and out of the neighborhood during the time between successive pick analyses. Interviews with the inhabitants to identify proper interventions are good since it identifies ‚ interventions that are relevant to the inhabitants and also increases the involvement the inhabitants in the waste management system. However, these interviews may lead to improved sorting behavior (which is good in many perspectives) which means that the effect of the intervention is difficult to isolate. Pick analyses, which are an integral part of the RBT procedure, require large sample sizes to be ‚ statistically meaningful. This increases the cost of the procedure.

Author Contributions: Kamran Rousta has taken part in planning the study, conducting the experimental work, analyzing the results, and has been the principal writer of the manuscript. Lisa Dahlén and Kim Bolton, supervisors, have taken part in evaluating the results and writing the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest. Recycling 2016, 1, 147–165 164 Recycling 2016, 1, 147–165 165

Abbreviations 26. Swedish Waste Management. Manual for Pick Analysis of Houshold Waste/Manual för Plockanalys av Hushållsavfall; RVF Utveckling 2013:11; Swedish Waste Management: Malmö, Sweden, 2013. (In Swedish) The following abbreviations are used in this manuscript: 27. Drever, E. Using Semi-Structured Interviews in Small-Scale Research. A Teacher’s Guide; ERIC: Glasgow, UK, 1995. RBT Recycling Behavior Transition 28. Henriksson, G.; Åkesson, L.; Ewert, S. Uncertainty regarding waste handling in everyday life. Sustainability 2010, 2, 2799–2813. [CrossRef] 29. Thøgersen, J. Facilitating recycling: Reverse—Distribution channel design for participation and support. References Soc. Mark. Q. 1997, 4, 42–55. [CrossRef] 1. Swedish Waste Management. Swedish Waste Management 2015; Rapport U2015:25; Swedish Waste 30. Tashakkori, A.; Teddlie, C. Sage Handbook of Mixed Methods in Social & Behavioral Research; Sage: Thousand Oaks, Management: Malmö, Sweden, 2015. (In Swedish) CA, USA, 2010. 2. Rousta, K.; Dahlén, L. Source separation of household waste; technology and social aspects. In Resource 31. Newenhouse, S.C.; Schmit, J.T. Qualitative methods add value to waste characterization studies. Recovery to Approach Zero Municipal Waste; Taherzadeh, M.J., Richards, T., Eds.; CRC Press: Boca Raton, FL, Waste Manag. Res. 2000, 18, 105–114. [CrossRef] USA, 2015; pp. 61–77. 3. Guerin, D.; Crete, J.; Mercier, J. A multilevel analysis of the determinants of recycling behavior in the © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access european countries. Soc. Sci. Res. 2001, 30, 195–218. [CrossRef] article distributed under the terms and conditions of the Creative Commons Attribution 4. Jesson, J. Household waste recycling behavior: A market segmentation model. Soc. Mark. Q. 2009, 15, 25–38. (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). [CrossRef] 5. Martin, M.; Williams, I.D.; Clark, M. Social, cultural and structural influences on household waste recycling: A case study. Resour. Conserv. Recycl. 2006, 48, 357–395. [CrossRef] 6. Meneses, G.D.; Palacio, A.B. Recycling behavior a multidimensional approach. Environ. Behav. 2005, 37, 837–860. [CrossRef] 7. Oskamp, S.; Harrington, M.J.; Edwards, T.C.; Sherwood, D.L.; Okuda, S.M.; Swanson, D.C. Factors influencing household recycling behavior. Environ. Behav. 1991, 23, 494–519. [CrossRef] 8. Tucker, P.; Speirs, D. Count me as doing my bit: An appraisal of the accuracy of waste attitude/behaviour surveys. CIWM Sci. Tech. Rev. 2003, 4, 2003. 9. Dahlén, L.; Åberg, H.; Lagerkvist, A.; Berg, P.E.O. Inconsistent pathways of household waste. Waste Manag. 2009, 29, 1798–1806. [CrossRef][PubMed] 10. González-Torre, P.L.; Adenso-Díaz, B. Influence of distance on the motivation and frequency of household recycling. Waste Manag. 2005, 25, 15–23. [CrossRef][PubMed] 11. Porter, B.E.; Leeming, F.C.; Dwyer, W.O. Solid waste recovery a review of behavioral programs to increase recycling. Environ. Behav. 1995, 27, 122–152. [CrossRef] 12. Refsgaard, K.; Magnussen, K. Household behaviour and attitudes with respect to recycling food waste—Experiences from focus groups. J. Environ. Manag. 2009, 90, 760–771. [CrossRef][PubMed] 13. Thøgersen, J. A model of recycling behaviour, with evidence from danish source separation programmes. Int. J. Res. Mark. 1994, 11, 145–163. [CrossRef] 14. Barr, S. Household Waste in Social Perspective; Values, Attitudes, Situation and Behaviour; ASHGATE: Burlington, VT, USA, 2002. 15. Pieters, R.; Verplanken, B. Changing our mind about behavior. In The Consumption of Time and the Timing of Consumption—Toward a New Behavioral and Socio-Economics; North-Holland: Amsterdam, The Netherlands, 1991. 16. Tucker, P.; Smith, D. Simulating household waste management behaviour. J. Artif. Soc. Soc. Simul. 1999, 2, 31. 17. Tucker, P. Understanding recycling behaviour. Pap. Technol. 2001, 42, 51–54. 18. Wilson, D.C.; Rodic, L.; Scheinberg, A.; Velis, C.A.; Alabaster, G. Comparative analysis of solid waste management in 20 cities. Waste Manag. Res. 2012, 30, 237–254. [CrossRef][PubMed] 19. Stringer, E.T. Action Research; SAGE: Los Angeles, CA, USA, 2014. 20. Dahlén, L.; Lagerkvist, A. Methods for household waste composition studies. Waste Manag. 2008, 28, 1100–1112. [CrossRef][PubMed] 21. Myers, M.D. Qualitative Research in Business and Management; Sage: Thousand Oaks, CA, USA, 2009. 22. Kvale, S.; Brinkmann, S. Den Kvalitativa Forskningsintervjun; Studentlitteratur AB: Lund, Sweden, 2009. 23. Rousta, K.; Ekström, K.M. Assessing incorrect household waste sorting in a medium-sized Swedish city. Sustainability 2013, 5, 4349–4361. [CrossRef] 24. Swedish Statistisc. Socio-Economic Statistics; Swedish Statistisc: Stockholm, Sweden, 2012. 25. Swedish Waste Management. Manual for Pick Analysis of Houshold Waste/Manual för Plockanalys av Hushållsavfall; RVF Utveckling 2005:19; Swedish Waste Management: Malmö, Sweden, 2005. (In Swedish) Recycling 2016, 1, 147–165 165

26. Swedish Waste Management. Manual for Pick Analysis of Houshold Waste/Manual för Plockanalys av Hushållsavfall; RVF Utveckling 2013:11; Swedish Waste Management: Malmö, Sweden, 2013. (In Swedish) 27. Drever, E. Using Semi-Structured Interviews in Small-Scale Research. A Teacher’s Guide; ERIC: Glasgow, UK, 1995. 28. Henriksson, G.; Åkesson, L.; Ewert, S. Uncertainty regarding waste handling in everyday life. Sustainability 2010, 2, 2799–2813. [CrossRef] 29. Thøgersen, J. Facilitating recycling: Reverse—Distribution channel design for participation and support. Soc. Mark. Q. 1997, 4, 42–55. [CrossRef] 30. Tashakkori, A.; Teddlie, C. Sage Handbook of Mixed Methods in Social & Behavioral Research; Sage: Thousand Oaks, CA, USA, 2010. 31. Newenhouse, S.C.; Schmit, J.T. Qualitative methods add value to waste characterization studies. Waste Manag. Res. 2000, 18, 105–114. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).