CPR Working Papers Social Development Department Environmentally and Socially Sustainable Development Network Paper No. 9 (Revised version) March 2004 Do Participatory Development Projects Help Villagers Manage Local Conflicts? A Mixed Methods Approach to Assessing the Kecamatan Development Project, Indonesia ____________________ Patrick Barron Rachael Diprose David Madden Claire Q. Smith Michael Woolcock Acronyms and Indonesian Terms Acronyms AN Austronesian (language group) CDD Community-Driven Development FGD Focus Group Discussion NTT Nusa Tenggara Timur SERP Social and Economic Revitalization Program UNDP United Nations Development Program Indonesian Jurisdictional Levels Kabupaten District, or regency Kecamatan Sub-district Desa Village Dusun Sub-village RT/RW Neighborhood/collection of neighborhoods Institutions Bappeda District Agency of the National Planning and Development Board (Badan Perencanaan Pembangunan Daerah) BAPPENAS National Planning and Development Board BPD Village Representative Council (Badan Perwakilan Desa) BPN National Land Agency (Badan Pertanahan Nasional) BPS Bureau of Statistics (Badan Pusat Statistic) Dinas Social Office of Social Affairs Dina Kesehatan Office/Department of Health Kodim District Military (Komando Daerah Militer) Koramil Sub-district military (Komando Rayon Militer) LKMD Village Development Planning Board (Lembaga Ketahanan Masyarakat Desa) LMD Village Development Board (Lembaga Masyarakat Desa) LSM Non-governmental organization (Lembaga Swadaya Masyarakat) PKK Kecamatan Development Program (Program Pembangunan Kecamatan) PMD Community Development Agency under the Ministry of Home Affairs (Pemberdayaan Masyarakat Desa) Polres District police (Kepolisian Resort) Polsek Sub-district police (Kepolisian Sektor) Puskesmas Community health center (Pusat Kesehatan Masyarakat) Government Positions Bupati District, or regency, head Camat Sub-district head Kades Village head (Kepala Desa) Kapolres Head of district police (Kepala Polisi Resort (Kabupaten)) Kapolsek Head of sub-district police (Kepala Kepolisian Sektor (Kecamatan)) Sekdes Village secretary (Sekretaris Desa) KDP positions/institutions FD Village facilitator (of KDP) (Fasilitator Desa) FK Sub-district facilitator (of KDP) (Facilitator Kecamatan) KDP Kecamatan Development Program KMKab District Management Consultant (Konsultan Managemen Kabupaten) Tim Kordinasi Coordination Team (District Level) PjOK KDP Project Manager (sub-district),Development Activities Officer (Penanggung Jawab Operasional Kegiatan) TPK Activities Management Team (Tim Pelaksanah Kegiatan) TTD Technical Assistance Team (Tengaga Teknis Desa) UDKP Kecamatan Development Forum, composed of village representatives and key local officials (Unit Daerah Kerja Pembangunan) UDKP3 Third Kecamatan Development Forum UPK KDP Financial Unit (sub-district) (Unit Pelaksanaan Kegiatan) Other Indonesian terms used Adat Traditional or customary (as in hukum adat, traditional law) Bahasa Language (also refers to Indonesian language) Hakim Judge Ibu Mother (respectful term of address for elder woman) Jaksir Advocates Kepala Head (as in head of, or chief) Kyai Islamic leader Rp. Indonesian currency (Rupiah) Surat Letter Tim Team Toko Masyarakat Community Elder Foreword Conflict is a normal feature in any society, and can surface for a number of reasons, but it can have creative or destructive consequences. Failure to manage conflict and prevent it from turning violent can be devastating--causing deaths and destruction, reversing hard-won development gains, and often leaving a trail of grievances that can reignite conflict in a never-ending cycle of violence. In turn, anticipating and effectively managing conflict can provide a solid foundation for more inclusive, sustainable and equitable development processes. As the Bank increasingly emphasizes community-driven approaches in its assistance strategies, it would be useful to know whether and how development projects contribute to local conflict resolution and, more specifically, whether the social skills learned through group-based decision-making--as is characteristic of community-driven development (CDD) projects--are in fact "transferable" to the management of local conflict. Can these new groups become local forums for mediating everyday forms of village conflict? Do these project-based skills generate social externalities that manifest themselves in something we might recognize as "empowerment" or "social capital", in which previously marginalized groups more confidently and competently engage the formal institutions of their society? Do these formal institutions, in turn, become more receptive to the concerns, interests, and aspirations of the poor as a result of the presence of a CDD project? Can more empowered communities, with a higher stock of positive social capital, more ably channel conflict constructively? The methodology outlined in this paper seeks to document one attempt to answer these questions, using as its empirical referent the Kecamatan Development Program (KDP) in Indonesia. It spells out how qualitative and quantitative approaches were combined to elicit in two provinces detailed ethnographic accounts of the pathways by which conflict emerges and then is resolved or inflamed. A particular feature of the design is that it collects information that will allow comparisons between villages in which KDP was present for three years, and statistically comparable villages in which KDP was not present. As such this paper is the first in a longer line that will emerge from this study and it is hoped that both the methodology and the results will be useful to others engaged in or contemplating similar studies or interventions in this area. The methodology presented here and the analysis that will flow from its application represent an important contribution to ongoing efforts by the Conflict Prevention and Reconstruction Unit and the Social Development Department to better understand, evaluate and document more systematically the contribution that CDD approaches can make to enhancing social capital in conflict-affected societies and make communities more resilient to conflict. It complements recent analysis on the experience with CDD approaches in post-conflict-reconstruction* and ongoing work on CDD and conflict. The authors wish to thank a number of people who were instrumental in the development of this methodology: Scott Guggenheim, Leni Dharmawan, Sri Kuntari, Junko Onishi, Menno Pradhan, Vivi Alatas, Yatrin Kaniu, Sentot Satria, Ben Olken and Kai Kaiser from the World Bank Office Jakarta; Susan Wong and Juana Brachet from the World Bank in DC; Peter Uvin and Pauline Peters, from Tufts University and Harvard University, respectively; and, of course, the researchers, especially Adam Satu, NTT provincial coordinator. Generous financial assistance for the project as a whole came from DfID, AusAID, the Norwegian Trust Fund (Measuring Empowerment Study), and the World Bank's Conflict Prevention and Reconstruction Unit, and Development Economics Vice Presidency (Research Support Budget). The authors also thank Katherine Whiteside (World Bank in DC) for editorial assistance. Ian Bannon Manager Conflict Prevention and Reconstruction Unit *Sarah Cliffe, Scott Guggenheim and Markus Kostner. 2003. "Community-Driven Reconstruction as an Instrument in War-to- Peace Transitions". Social Development Papers, CPR Working Papers, No. 7, August. Washington, DC: World Bank. Executive Summary This paper outlines a methodology that seeks to determine whether and how development projects contribute to conflict resolution and whether the social skills learned through group-based decision- making--a key feature of community-driven development (CDD) approaches--are transferable to the successful management of local conflicts. The methodology uses as its empirical reference point Indonesia's Kecamatan Development Program (KDP)--KDP is a massive community development project, the largest in Southeast Asia, operating since 1998 and covering over 20,000 villages across Indonesia. The paper spells out how qualitative and quantitative approaches are being combined in two provinces--Nusa Tenggara Timur (NTT) and Jawa Timur (East Java)--to produce detailed ethnographic accounts of the pathways by which conflict is generated and then resolved or exacerbated. The research first seeks to understand and map conflicts and tensions that exist in the research areas, as a preliminary step to understanding the capacity of communities to cope with conflict. The research then looks at the conflict resolution mechanisms that exist, whether they are successful or not, and which mechanisms people turn to in times of trouble. The study seeks to integrate qualitative and quantitative tools, in a way that takes advantage of the strengths of each approach, while controlling for the weaknesses. A particular feature of the methodology is that it elicits information enabling comparisons to be made between villages in which KDP was present for three years, and statistically comparable villages in which KDP was not present. The research seeks to answer five questions: (i) what are the main factors that affect local level capacity to manage conflict; (ii) how important is the nature and extent of interaction between different groups, and between those groups and the state, to local conflict mediation, and how boundaries between different groups are constructed and sustained; (iii) does KDP help communities manage conflict more constructively and can external agents help establish more inclusive, transparent, and accountable local- level institutions for mediating conflict; (iv) if so, for what types of conflict and under what conditions; and (v) if so, which elements of KDP appear to be most influential. As far as the authors can ascertain, this is the first systematic and comprehensive effort to establish whether and how a CDD project can help to improve local conflict mediation processes. Also noteworthy are its reliance on integrated--quantitative and qualitative--data sources, and the relatively long periods of time spent by field staff in villages collecting data. Outputs from the research will take a number of forms. Local field staff will have the opportunity to disseminate locally the material from their case study narratives, and importantly, to provide feedback to the communities that originated the data. Too often data collection is a one-way process, with researchers gathering data but rarely returning to share the results. Eight workshops are planned at the sub-district level to disseminate and discuss results. The researchers will be responsible for running these workshops. The material collected in the research will also be publicly available once it has been cleaned up to protect confidentiality. Other researchers, academics, NGOs, and the government will be able to use the data collected by the study. In addition, the results of this research are expected to feed into the design and implementation of follow-on KDP projects (KDP2 and KDP3) but would also be of interest to researchers and practitioners working on participatory mechanisms more broadly. Table of Contents Foreword 4 A. Background, Aims and Concepts 10 1. BACKGROUND .....................................................................................................................................10 2. RESEARCH QUESTIONS........................................................................................................................11 3. CONCEPTS, ASSUMPTIONS AND DEFINITIONS.....................................................................................11 3.1 Conflict, Social Tensions and Violence ........................................................................................11 3.2 Factors Affecting Levels of Conflict and Local Capacity.............................................................12 3.3 Groups, Ethnicity and Civic Interaction.......................................................................................13 3.4 KDP, CDD, and Conflict Resolution............................................................................................13 B. Research Topics and Hypotheses 14 1. THE FACTORS AFFECTING LOCAL LEVEL CAPACITY TO MANAGE CONFLICT.......................................14 1.1 Economic/Structural Factors........................................................................................................15 1.2 Psychological/Cultural Factors....................................................................................................15 1.3 Institutional Factors .....................................................................................................................15 2. FORMS OF CIVIC INTERACTION WHICH HELP RESOLVE OR MANAGE CONFLICT ..................................16 3. THE RELATIONSHIP BETWEEN KDP AND COMMUNITY CONFLICT MANAGEMENT..............................17 3.1 Types of Conflict which KDP Is Most Effective in Addressing.....................................................17 3.2 Cases in which KDP Has a Negative Effect .................................................................................17 3.3 External Factors that Affect the Ability of KDP to Successfully Manage Conflict.......................18 3.4 Mechanisms by which KDP Might Effect Change........................................................................18 C. Understanding Conflict and Conflict Management: Overview of Methodology 19 1. MIXED METHODS: QUALITATIVE AND QUANTITATIVE TOOLS ..........................................................19 2. PHASES OF RESEARCH.........................................................................................................................21 3. PROVINCIAL AND DISTRICT SAMPLING...............................................................................................22 3.1 Provincial Research Site Selection...............................................................................................22 3.2. District Research Site Selection...................................................................................................23 4. PREPARATION AND PRE-FIELDWORK..................................................................................................23 4.1 Recruitment...................................................................................................................................23 4.2 Preliminary Research ...................................................................................................................25 5. TRAINING ............................................................................................................................................25 5.1 Training 1 .....................................................................................................................................26 5.2 Training 2 .....................................................................................................................................26 5.3 Training 3 .....................................................................................................................................26 5.4 Training 4 .....................................................................................................................................26 6. QUALITATIVE FIELDWORK..................................................................................................................27 6.1 Qualitative Fieldwork Phase 1 .....................................................................................................27 6.2 Qualitative Fieldwork Phases 2A and 2B.....................................................................................28 6.3 Qualitative Fieldwork Phases 3A and 3B.....................................................................................28 6.4 Qualitative Writing Workshops ....................................................................................................29 7. QUANTITATIVE RESEARCH..................................................................................................................30 7.1 PODES Survey..............................................................................................................................30 7.2 Governance and Decentralization Survey (GDS).........................................................................30 7.3 Key Informant Survey ...................................................................................................................31 7.4 Aims of Quantitative Research .....................................................................................................31 8. TYPES OF DATA...................................................................................................................................32 8.1 Case Data .....................................................................................................................................32 8.2 General Data ................................................................................................................................33 D. Sampling of Local Level Research Sites 37 1. KECAMATAN (SUB-DISTRICT) SELECTION ...........................................................................................37 8 1.1 Kecamatan Selection: Phases 1, 2A and 2B.................................................................................37 1.2 Kecamatan Selection: Phases 3A and 3B.....................................................................................38 2. VILLAGE AND PRIMARY CASE SELECTION (PHASES 2A AND 2B) ......................................................39 2.1 Selecting Villages..........................................................................................................................39 2.2 Selection of Cases One and Two: Similar cases, different outcomes (within same kecamatan) ..40 2.3 Selection of Case Three: Similar cases in two kecamatan............................................................40 2.4 Selection of Case Four: Peaceful resolution in a violent area.....................................................41 2.5 Selection of Case Five: Violent case in a peaceful area...............................................................41 2.6 Selection of Case Six: KDP-related case......................................................................................41 2.7 Primary Case Studies: Phase 2A..................................................................................................42 3. `NON-PRIMARY' CASES (QUALITATIVE WORK).................................................................................44 4. VILLAGE SELECTION (PHASES 3A AND 3B)........................................................................................44 E. Sampling of Respondents 45 1. QUALITATIVE RESEARCH (PHASE 1)...................................................................................................45 2. QUALITATIVE RESEARCH (PHASES 2A AND 2B).................................................................................45 2.1 Cross-Section of Population.........................................................................................................45 2.2 Locating Additional Respondents Involved in a Conflict Case.....................................................46 2.3 Focus Group Discussions (FGDs)................................................................................................47 3. KEY INFORMANT SURVEY...................................................................................................................47 F. Qualitative Research Instrument #1: In-depth Interviews 47 1. OVERVIEW OF IN-DEPTH INTERVIEWS................................................................................................47 2. OBJECTIVES AND TARGET GROUPS.....................................................................................................48 3. METHODS AND PROCEDURES..............................................................................................................48 3.1 Qualitative Research Phase 1.......................................................................................................48 3.2 Qualitative Research Phases 2 and 3 ...........................................................................................48 4. RECORDING AND WRITE-UP................................................................................................................49 G. Qualitative Research Instrument # 2: Focus Group Discussions (FGDs) 49 1. OVERVIEW OF FGDS...........................................................................................................................49 2. OBJECTIVES AND TARGET GROUPS.....................................................................................................50 3. METHOD AND PROCEDURE..................................................................................................................50 3.1 Focus Groups with Women...........................................................................................................50 3.2 Focus Group Discussions with Men .............................................................................................51 4. RECORDING AND WRITE-UP................................................................................................................51 H. Qualitative Research Instrument # 3: Informal Interviews and Participant Observation 51 1. OVERVIEW OF INFORMAL INTERVIEWS AND PARTICIPANT OBSERVATION........................................51 2. OBJECTIVES AND TARGET GROUPS.....................................................................................................52 3. RECORDING AND WRITE-UP................................................................................................................52 3.1 Informal Interviews.......................................................................................................................52 3.2 Participant Observation ...............................................................................................................52 I. Qualitative Research: Recording the Data 53 1. RAW DATA FORMATS..........................................................................................................................53 1.1 Phase 1..........................................................................................................................................54 1.2 Phase 2..........................................................................................................................................54 1.3 Phase 3..........................................................................................................................................55 2.ANALYSIS FORMATS AND PIECES ........................................................................................................55 2.1 Phase 1..........................................................................................................................................55 2.2 Phase 2..........................................................................................................................................55 2.3 Phase 3..........................................................................................................................................56 J. Writing Case Studies 56 1. OVERVIEW OF CASE STUDY WRITING ................................................................................................56 2. CONTENT OF CASES ............................................................................................................................57 9 3. SOURCING AND GLOSSARIES...............................................................................................................58 K. Data Entry and Analysis 58 1. DATA ENTRY AND TRANSLATION PROCEDURES.................................................................................58 1.1 Qualitative Research Phase 1.......................................................................................................58 1.2 Qualitative Fieldwork Phases 2 and 3: Primary Data.................................................................58 1.3 Qualitative Fieldwork Phases 2 and 3: Analyzed Data................................................................59 2. AGGREGATION AND ANALYSIS OF DATA............................................................................................59 3. REVIEW PLANS....................................................................................................................................59 L. Final Outputs 60 1. VILLAGE FEEDBACK AND CAPACITY BUILDING.................................................................................60 2. OPERATIONAL SUPPORT......................................................................................................................60 3. POLICY RESEARCH ARTICLES .............................................................................................................60 4. METHODOLOGICAL TOOLS..................................................................................................................61 M. References 62 Annex A: Research Topics, Data Codes and Field Guides 63 A1: GENERAL DATA TOPICS AND CODES ...............................................................................................63 A2: GENERAL DATA FIELD QUESTION GUIDE........................................................................................64 A3: CASE DATE TOPICS, CODES AND FIELD QUESTION GUIDE..............................................................73 Annex B: Interview Guidelines 76 B1: GENERAL INTERVIEW GUIDELINES ..................................................................................................76 B2: GUIDELINES FOR ACCESSING AND INTERVIEWING GOVERNMENT OFFICIALS ................................77 B3: GUIDELINES FOR ACCESSING AND INTERVIEWING MILITARY AND POLICE.....................................77 B4: GUIDELINES FOR ACCESSING AND INTERVIEWING HEALTH OFFICIALS ..........................................78 Annex C: Data Recording Formats--Qualitative Research Phase 1 80 C1: STANDARD DATA FORMAT ..............................................................................................................80 C2: STANDARD INTERVIEW RECORDING FORMAT .................................................................................81 C3: CASE STUDY RECORDING FORMAT..................................................................................................82 C4: HEALTH OFFICIALS RECORDING FORMAT .......................................................................................83 Annex D: Data Recording Formats--Qualitative Research Phases 2A and 2B 88 D1: STANDARD DATA FORMAT ..............................................................................................................88 D2: STANDARD FGD DATA FORMAT .....................................................................................................89 D3: STANDARD INTERVIEW TRANSCRIPT RECORDING FORMAT............................................................90 D4: FOCUS GROUP DISCUSSION TRANSCRIPT RECORDING FORMAT .....................................................90 D5: CASE STUDY RECORDING FORMAT..................................................................................................91 Annex E: Data Recording Formats--Qualitative Research Phase 3 92 E1: STANDARD DATA FORMAT...............................................................................................................92 E2: STANDARD INTERVIEW RECORDING FORMAT..................................................................................92 E3: CASE STUDY RECORDING FORMAT..................................................................................................92 E4: KDP DATA FORMAT: KECAMATAN LEVEL........................................................................................93 E5: KDP DATA FORMAT: VILLAGE LEVEL.............................................................................................96 E6: KECAMATAN DEMOGRAPHIC FORM...................................................................................................99 E7: VILLAGE DEMOGRAPHIC FORM......................................................................................................103 E8: INFORMANT LIST ............................................................................................................................110 10 A. Background, Aims and Concepts 1. Background Since the fall of Soeharto in 1998, violent conflict has become more widespread in Indonesia. This destructive conflict has not only played out in the high profile conflict regions (Aceh, West Papua, the Malukus, Central Sulawesi, and Central and West Kalimantan), but across the archipelago. The dynamics of transition--from an authoritarian state to the foundations of a democratic one--help explain the background to many of these conflicts. During the thirty years of New Order government, a `vertical' institutional state structure regulated conflict but at the tragic expense of the human rights of the Indonesian people. Recent decentralization and regional autonomy laws, accompanied by ongoing democratization, have set Indonesia on a path that has the potential to dramatically change the nature of Indonesian society (Friend, 2003). Yet a fully functioning democracy has still to take its place, and a conflict management vacuum is still to be filled. This macro picture is increasingly observed. However, what is less clear is how this plays out at the local level. Given that many of the external or structural factors caused by transition are similar across Indonesia, why do some communities experience violence and others not? And within regions, why are some communities violent while others are peaceful? What localized factors matter in determining whether or not conflict takes violent form? This research uses a mixed methodology (with qualitative and quantitative components) to identify the factors that affect local level capacity to manage conflict. By local level we refer to the sub-district level and below--the kecamatan (sub-district), desa (village), dusun (sub-village or hamlet), and the household. The research has taken place in two Indonesia provinces: Nusa Tenggara Timur and Jawa Timur.1 In particular, the research will evaluate the extent to which local level organizational and civic skills-- particularly civic interaction and the existence and capacity of local institutions--are important (relative to other factors) in determining whether communities suffer from violent conflict. The research will also examine the extent to which the Kecamatan Development Project (KDP), the World Bank's primary community-driven development project in Indonesia, helps villagers build such skills, and will seek to test empirically whether this accounts for higher conflict management capacity. If organizational and civic skills are acquired by villagers in a participatory development project such as KDP, and if these skills carry over to the successful management of local conflicts, then such projects may have a broader development impact than alternative delivery mechanisms. If they do not, the study will still yield important insights regarding the capacity or limits of external interventions seeking to instill basic "grassroots" democratic procedures and conflict resolution skills in the citizens of post- autocratic developing countries. The research will seek first to understand (and map) conflicts and tensions that exist in the research areas. Understanding the range of conflicts is the first step toward understanding community capacity to cope with conflicts. The research will look, second, at the conflict resolution mechanisms that exist, whether they are successful or not, and which mechanisms people turn to in times of trouble. These activities are a prerequisite to answering the following five key research questions. 1Hereafter, referred to by their English names: NTT and East Java. How and why these particular provinces were selected is outlined below (see Section C3). The research may extend into further Indonesian provinces in the future. 11 2. Research Questions The research is seeking to answer five questions: 1. What are the main factors that affect local level capacity to manage conflict? 2. How important is the nature and extent of interaction between different groups, and between those groups and the state, to local conflict mediation? How are boundaries between different groups constructed and sustained? 3. Does KDP help communities manage conflict more constructively? More generally, can external agents help establish more inclusive, transparent, and accountable local level institutions for mediating conflict? 4. If so, for what types of cases of conflict, and under what conditions? 5. If so, which elements of the KDP program appear to be most influential? 3. Concepts, Assumptions and Definitions 3.1 Conflict, Social Tensions and Violence Conflict exists wherever people live together. In all human societies people disagree about the distribution of scarce resources, about values, and about the access to power that allows influence over the two. Conflict is characterized by disagreements and often animosity between individuals or groups. Conflict in itself is not a negative phenomenon. Development is premised on the notion of change and progression--social, political, and economic. Conflict inevitably accompanies such change because it involves realignments of power and resources as well as challenges to existing interests and values. Conflict is thus both a necessary catalyst to, and an inevitable by-product of, development (Bates, 2000). Social tensions are the substantive areas over which individuals or groups disagree and compete. They may take the form of disagreements over specific resources (e.g., land, water, or political power), over different values (e.g., moral and religious rules and codes of living), or over a combination of both. Conflict plays out through social tensions; social tensions are the forces that underlie situations of conflict. Of concern is not so much social tensions or conflict, but whether such tensions lead to violence. While social tensions and conflict are to some extent inevitable, violent conflict is not. This study assumes that violent conflict is a negative. Violent conflict is defined in this study as conflict that is destructive. Indicators of violent conflict include injuries/deaths and destruction of property or goods. Violent conflict, rather than conflict per se, has a negative impact on attempts at poverty reduction and development. This study is concerned with why some conflicts become violent and others not.2 We have studied this by looking at comparable cases of violent and nonviolent conflict. That is, we have studied similar social tensions where the outcomes were different--violent or peaceful. A dispute is a particular incident of conflict. By an incident, we mean that we can (or at least could if we had adequate information) distinguish the actors, the location, and the time it took place. A broader conflict may be made up of a number of disputes; a social tension may manifest itself in a number of disputes. We have used the dispute as our main level of analysis as we followed `conflict cases' in the qualitative research. A dispute may, or may not, be violent. Conflicts and disputes can take a range of forms and intensities. Previous research has shown that forms of conflict tend to be related, that small disputes act as triggers for bigger conflicts (Esman and Herring, 2001). In order to understand these larger conflicts we need to study the smaller ones that led to them. 2Violent conflict is a major barrier to development and, through its impact on human security, constitutes both a cause and consequence of `undevelopment'. We argue that the vast majority of violence (at least in Indonesia) is unconstructive and does not act as a mechanism for progressive social change. 12 We are interested in studying a wide range of conflicts, from seemingly unimportant disputes such as arguments between neighbors and incidences of robbery to large-scale conflicts such as inter-village burnings. For the purposes of this study, we thus define the terms `conflict' and `dispute' broadly. 3.2 Factors Affecting Levels of Conflict and Local Capacity This study looks at conflict as it plays out at the local level. We define the local level as being the sub- district level and below: the village, the sub-village, and the household. We use these units of analysis because these are the levels at which KDP operates and hence, we assume, the levels at which it potentially has an impact. In some places, we use the term community. A community is a group, usually spatially concentrated, that shares certain values, beliefs, or demographic attributes. The characteristics and importance of these attributes, however, (a) are heavily context specific (that is, can change according to time, place, and circumstance), and often (b) serve to create and sustain the perceived importance of differences between groups. The extent to which violent conflict occurs at the local level is dependent on exogenous and endogenous factors. Exogenous factors may include national economic policy, population flows, and centralized security institutions. Such forces can overwhelm local communities. Large shocks can lead to both high levels of social tension and violent conflict. However, external macro forces only partially explain why some conflicts take violent form. Social tensions exist in many places, yet only lead to violent conflict in some. External forces of the same type and scale impact in different ways in different places. This study is particularly interested in factors endogenous to the local level that influence the nature and extent of violent conflict. We make the assumption, based on previous research we have conducted (Barron and Madden, 2003a; Smith, 2004), that levels of violent conflict at the local level are determined, in significant part, by localized factors and by the lack of local capacity to constructively manage conflict. By constructive conflict management we mean the ability to manage and resolve problems in ways that prevent the conflict from taking violent form. We use a framework that disaggregates localized factors that influence levels of violent conflict at the local level into three sub-sets: (1) economic and structural factors; (2) psychological and cultural factors; and (3) social and political institutional factors. (1) Economic and structural factors are descriptors of social structure that can be observed objectively and usually measured quantitatively. They include economic factors (such as inequality--between individuals and groups--and social mobility) and demographic factors (for example, cleavages along the lines of age, gender and ethnicity). Prior research we have conducted tends to indicate that these factors often determine the parties involved in conflict. (2) Psychological and cultural factors concern dimensions of cognition. Conflict cannot solely be understood by the cleavages that dictate and describe social structure at the local level. Conflict is not only determined by these structural factors, but by people's interpretations of such structures and their meanings, and the values people place in and upon them. Our prior research has found that factors such as trust and cultural practice tend to explain the intensity of conflict. (3) Social and political institutional factors concern the formal and informal institutions at the local level. Constructive conflict management involves distributing resources, balancing interests and values, and encouraging development in ways that prevent conflict from becoming violent. Inclusive, accountable and just institutions, which people believe to be trustworthy and credible, are necessary to do this. We define institutions broadly to include groups ranging from formal state-dictated government bodies to informal functional collectives (such as saving and loans groups) to social forums (such as prayer groups). 13 Institutions help dictate the extent to which the first two categories of factors (economic/structural and psychological/cultural) develop in ways that either promote or prevent conflict. All three subsets of factors thus help determine levels of conflict at the local level. At the same time, the three subsets also help determine the extent to which conflict becomes violent. When effective institutions are in place, problems are managed or resolved so that they do not become violent. Thus institutional factors are key. At the same time, the first two categories may influence the likelihood that effective conflict managing institutions will come into being. Thus all three categories are related and together determine local conflict management capacity. We use the term local capacity to mean the extent to which communities have norms, processes, and institutions that allow conflict to be addressed constructively--i.e., in nonviolent ways. 3.3 Groups, Ethnicity and Civic Interaction Another assumption of this study is that a large proportion of conflict in Indonesia is group-based. Even where conflicts are between individuals (e.g., a fight over a woman at a party), group-based identities tend to determine who the conflicting parties are. Thus much of the violence in Indonesia is, to some extent, communal (i.e., between groups and based on group identity). By group-based identity we mean forms of meaning and self-worth that adhere to membership in a group. These groups can be either ascriptive (race, language, clan, caste, language, etc) or prescriptive (village, political party, economic group where mobility is possible, etc.). In many cases the two may overlap. Following the lead of Donald Horowitz (1985), we term all identities based on ascriptive factors `ethnic'. Thus ethnic conflict includes not only `racial' or `linguistic' conflicts but also conflicts between clans or between religious groups. Our prior research has found that most group-based conflict has an ascriptive group basis. Thus, in some places we use the terms group-based conflict and ethnic conflict interchangeably. A key hypothesis of this study is that the nature and extent of social interaction between individuals and groups is a key variable in determining levels of conflict and local conflict management capacity. We use the term civic interaction as short-hand to refer to these interactions. Civic interaction can take two forms: `associational' and `non-associational'. Associational interaction refers to interaction that takes place in an associational setting (formal or informal). Non-associational interaction refers to everyday forms of engagement, including meeting in the market, attending celebrations and parties, and children playing together. Sometimes social scientists and economists aggregate these interactions and term them social capital. For the purposes of this study, however, we choose to use a range of terms that more precisely capture the particular features of organizational life we are studying. In addition to dichotomizing interactions into associational and non-associational, we also differentiate between intra-group and inter-group interactions. We hypothesize that there is a qualitative difference in the two in terms of their impact on levels of violent conflict. We hypothesize that when the former type, intra-group interactions, alone take place, the likelihood of violent conflict increases. Comparatively, the presence of the latter form, inter-group interactions, can lead to reductions in violent conflict. We will thus explore these two sets of social interactions. Further, we use the term bridging interactions to refer to interactions that take place between people of different identity groups. 3.4 KDP, CDD, and Conflict Resolution This study seeks to evaluate the impact of the Kecamatan Development Project (KDP) on communities' ability to manage conflict. Initiated in 1998, KDP is a massive community development project, the largest in Southeast Asia, operating in over 20,000 villages across Indonesia. Through block grants given directly to kecamatan (sub-districts) and villages for small-scale infrastructure, social and economic activities, it aims to alleviate poverty and improve local governance. 14 KDP is perhaps the original large-scale community-driven development project. Community-Driven Development (CDD) can be defined as development projects that are designed, implemented, and maintained by the community. This implies communities having control over both decisions and resources, with support organizations and service-providers intended to be demand-responsive (Alkire, et. al., 2001). As such, KDP focuses on processes (`means') as much as outputs (`ends'). A key hypothesis that we are seeking to test is that KDP is having `democratization' effects--by creating new channels for participation, and supporting existing ones--and that this, in turn, is improving communities' capacity to manage conflict.3 While KDP was by no means designed as a conflict resolution program, anecdotal evidence suggests this may be one possible unexpected spillover from the program. Research undertaken by the authors of this paper, cited above, as well as by the Government of Indonesia (2002), has found cases where KDP forums have been used to help resolve conflicts. A principle aim of this study is thus to test, systematically and empirically, the extent to which KDP is having an effect on local conflict management capacity. By studying the case of KDP, we hope to be able to make some more general claims about the extent to which CDD mechanisms can positively impact upon situations which might otherwise lead to violent conflict. B. Research Topics and Hypotheses From our five key research questions, we have three main areas for research. For each research topic we explored hypotheses about factors that may enhance community capacity to negotiate problems and manage conflict, thus reducing levels of violence. The first two areas were explored in both parts of Phase 2 of the qualitative research, in the KDP and non-KDP sub-districts (see Section D on sub-district sampling); the last area--the relationship between KDP and community conflict management--was explored only in Phase 2B of the qualitative research, in the KDP sub-districts, and in both parts of Phase 3.4 Data gathered during the quantitative survey(s) will allow us to test the generality of some of the hypotheses, both within and across research sites. Listed below are the hypotheses we want to test and the measures by which we will seek to test them.5 1. The factors affecting local level capacity to manage conflict Here we are interested in disentangling the different factors that help dictate levels of conflict and communities' abilities to manage it. Using the framework outlined above (Section A3.2), there are three key areas we will review: · Economic and structural factors · Psychological and cultural factors · Social and political institutional factors 3We are, in effect, testing whether the causal chain outlined in the latest UN Human Development Report (2002), which sees the promotion of democracy as a key conflict reducing tool, holds out in a specific democratization intervention. 4The main part of the qualitative research ran in eight sub-districts (four with KDP) in four districts in the two provinces. (Phases 3A and 3B expanded into an additional eight sub-districts in the same districts, all KDP locations). The qualitative research was divided into five phases: Phase 1, Phase 2A, Phase 2B, Phase 3A and Phase 3B. Phase 1 provided a broad overview of the types, locations and levels of tension and conflict in all the sub-districts. Phase 2A was primarily village level research in the non-KDP sub-districts; phase 2B was research in the KDP sub-districts. Phases 3A and 3B examined in closer detail the relationship between KDP and conflict (and its resolution), and involved simultaneous qualitative and quantitative research (see Section C6 for a description of the aims of each phase of the qualitative fieldwork). 5In some cases, these `measures' correspond to quantitatively testable indicators; in others, scales will need to be constructed from the qualitative research. 15 1.1 Economic/Structural Factors (a) High levels of economic inequality increase envy and resentment, and reduce the extent to which residents of a given community regularly interact with one another, thereby increasing the likelihood of conflict. Measures: - Evidence of wealth extremes - `Categorical' inequality (e.g., by occupation, age, ethnicity) - Differences in stability/predictability of income over one year - Levels of interaction across groups (b) The likelihood of conflict leading to violence increases where the possibility of economic mobility through law-abiding behavior and hard work is low, especially for young men. (Conversely, the more young men must resort to illegal and/or illegitimate means to secure their livelihood, the more likely conflict leads to violence.) Measures: - Unemployment rate (especially for young men)--current and trend - Extent of illegal economic activity (e.g., drugs) - Income differentials between legal and illegal sectors - Social mobility measures (c) Ethnic diversity reduces the potential for conflict when it is either very low (single group) or very high (many groups), but increases it when diversity is moderate (e.g., three or four dominant groups). Measures: - Number of identifiable (by respondents) ethnic, religious, and clan groups - Range of ways in which "social difference" is constructed and maintained 1.2 Psychological/Cultural Factors (a) Conflict will be less likely to lead to violence in communities where a culture of negotiation and inclusive consensus-seeking (as opposed to retribution, vengeance, and exclusively elite male decision-making) prevails. Measures: - Ethnographies of everyday local level conflict resolution mechanisms, and the extent to which they are regarded as fair and effective - Expectations/norms regarding appropriate procedures for seeking redress in the event of wrongdoing (intentional and unintentional) (b) The trauma and grief associated with extended intense conflict (of whatever origin) will itself be a source of instability, making violence more likely. Measures: - Intensity, frequency, and duration of violent conflict - Nature and extent of individual coping strategies (or lack thereof) 1.3 Institutional Factors (a) When government and community leaders are trusted, they play a key role in helping communities negotiate conflicts without violence. Measures: - Trust in village government - Trust in community leaders - Involvement of leaders in conflict management (b) Open and inclusive village governments with predictable and transparent decision making processes ensure lower levels of violent conflict through assisting communities in resolving conflicts. Measures: - Participation in decision-making 16 - Knowledge of government decisions and processes - Access to information on government decision-making - Predictability of decision making - Levels of violent conflict (c) Where decision-making mechanisms in local (non-government) associations are transparent, open, and accountable, communities are more likely to be able to negotiate conflicts and prevent them from becoming violent. Measures: - Extent to which community leaders are accountable to their constituents - Extent to which decision-making is transparent (and examples of how it is ensured, e.g., community bulletin boards) - Extent to which a local association's rules permit open participation in decision-making forums (d) Communities that allow women to occupy key leadership roles will be less prone to violence Measures: - Number of women in leadership positions - Range of mechanisms for including/excluding women in/from leadership positions 2. Forms of civic interaction which help resolve or manage conflict We are interested here in the role and presence of inclusive inter-community mechanisms and whether they help communities to manage conflict, thus reducing levels of violent conflict. We will explore the following hypotheses: (a) Communities with institutions that involve or represent different ethnic groups will have lower levels of conflict. Conversely, communities are less able to manage conflict when institutions only operate intra-ethnically. Measures: - Involvement by ethnic group in different institutions (Government, religious, adat (traditional), and other community groups) (b) Formal associational inter-ethnic institutions have a greater impact on lowering levels of conflict than informal everyday forms of inter-ethnic engagement.6 Measures: - Involvement by ethnic group in different institutions (Government, religious, adat, and other community groups) - Measurements of `non-associational quotidian' interaction (c) Where multi-ethnic communities do not have bridging forums, inter-group conflict tends to trigger violent conflict. Measures: - Involvement by ethnic group in different decision-making forums (Government, religious, adat, and other community groups) (d) Traditional or customary (adat) institutions may restrict the capacity of communities to resolve conflict when they do not include other ethnic or marginalized groups in their decision-making mechanisms. Measures: - Involvement by ethnic/clan/socio-economic group in adat institutions - Extent to which different groups participate in adat decision-making 6This will test the theory Varshney (2002) generates from the Indian experience regarding the importance of overlapping civic associations for preventing and mediating communal conflict. 17 3. The relationship between KDP and community conflict management KDP is itself multi-faceted, and under different conditions is likely to have had a differential impact on different types of conflict. Here we are interested in four primary questions: · For what types of conflict can KDP have a positive impact? · In what cases does KDP lead to conflict, or worsen existing conflict? · What external conditions need to be met for KDP to have a positive impact on reducing levels of conflict? · Which specific elements of KDP affect communities' ability to manage conflict? For each of these questions, we explore KDP's impact through examining communities over time, comparatively before and after the project. To do this, where it is not explicitly specified in each of the measures below, we attempted to identify each measure at points before, during and after KDP was present in the research areas. 3.1 Types of Conflict which KDP Is Most Effective in Addressing (a) KDP is most effective at helping resolve conflicts generated at the level at which KDP itself operates (which is to say, at the local level, rather than more widespread communal, separatist, or military conflict). Measures: - Pre-existing levels and types of conflict (at different levels) - Different types of `problems' addressed at KDP forums (b) KDP is better at addressing conflicts that involve parties included in KDP decision-making processes than those involving other external parties. Measures: - Participation in KDP (by different groups) - Pre-existing levels and types of conflict (at different levels) 3.2 Cases in which KDP Has a Negative Effect (a) Because it generates its own forms of micro-conflict (by initiating competition to allocate finite resources, etc.), KDP has potentially negative effects in those circumstances in which: § Ethnic and economic tensions are already high, and where (for whatever reason) KDP staff fail to deliver program resources as promised § EntrenchedleadersareabletocircumventKDPrules § Local level institutions have been (or are) ill-prepared to cope with the specific forms of conflict (and/or decision-making) KDP initiates § State law enforcement organizations (e.g., the police) fail to uphold a minimally enabling environment Measures:7 - Facilitator's capacity to effectively engage communities and convey (and enforce) KDP rules - Quality of technical assistance, and institutional support given to facilitators - Pre-existing levels and types of conflict - Nature and extent of engagement with police/law enforcement officials 7Given the small sample size, we were unable to "test" these more specific hypotheses in a deductive manner using qualitative methods alone. The issues to which they are directed are probably best addressed inductively by the field researchers. Field staff thus explored specific ways and means by which KDP has both positive and negative impacts, and the various combinations of "structural", "programmatic", and idiosyncratic factors that have shaped both local organizational effectiveness and the capacity to manage conflict. The quantitative research may, however, be able to identify some more revealing patterns. 18 3.3 External Factors that Affect the Ability of KDP to Successfully Manage Conflict (a) In communities where there is significant elite control/capture of KDP processes, community capacity will not be enhanced. Measures: - Levels and forms of elite control/capture of KDP decision-making forums - Levels and forms of elite control/capture of KDP implementation teams and mechanisms - Levels and forms of elite control of other (non-KDP) development projects - Types of elite control (e.g., where relatives of government officials hold all key decision- making roles inside/outside KDP; where traditional leaders dominate decision-making) - Forms of elite control over local government (e.g., relationships between village and sub- district government officials; where the same group dominates both village parliament and village government apparatus). 3.4 Mechanisms by which KDP Might Effect Change One of our main research questions is to test whether KDP helps communities manage conflict more effectively. This could happen for any, or all, of the following five hypothesized reasons: (a) Cognitive change: The improvement in civic skills and trust between ethnic groups through involvement in the KDP socialization, planning, decision-making and implementation stages. Measures: - Levels of trust between groups over time (pre- and post-KDP) - Participation in KDP (b) Institutional change I: The creation of inter-group forums where problems can be discussed provides new avenues for addressing problems. Measures: - Levels of interaction across ethnic groups in other forums before KDP - Changed levels of interaction post-KDP - Extent to which KDP forums resolved non-KDP-related problems (c) Institutional change II: The increased participation of marginalized groups enhances community capacity to resolve conflict (and that this works in both homogenous and heterogeneous communities). Measures: - Levels of interaction across socio-economic groups (disaggregated by gender) before KDP - Levels of interaction post-KDP (d) Cultural change: KDP, in encouraging democratic decision-making at the local level, changes people's values, and, in doing so, helps create a culture of peace. Measures: - Attitudes toward peace, violence, and decision-making and differences pre- and post- project (e) Outside actor: The more effective the kecamatan facilitator (FK) is at enforcing KDP's rules pertaining to inclusive group decision-making, the more effective the new decision-making mechanisms put in place, and the lower the probability of conflict. Measures: - Role of FK and other extra-village facilitators - Levels of participation by different groups 19 C. Understanding Conflict and Conflict Management: Overview of Methodology 1. Mixed Methods: Qualitative and Quantitative Tools This study seeks to integrate qualitative and quantitative tools. Each method has strengths and weaknesses (see Kanbur, 2002). Quantitative approaches allow generalizations to be made on the basis of small samples if they are representative. Statistical approaches, in measuring covariance, are very good at establishing correlations between phenomena. Quantitative approaches are thus effective at predicting events--if X is observed, how likely is it that Y will occur? And the ability to quantify answers allows, at least in theory, easy comparison between research areas, in a way that most qualitative approaches--open to subjective interpretations and biases--do not. However, quantitative approaches are not without weakness. As Bennet and George (1997) correctly note, prediction is not synonymous with explanation. Quantitative approaches are less effective at establishing why phenomena occur--if X happens, why is it likely that Y will occur? Many important issues--such as identities, and experiences of exclusion--cannot be reduced to numbers. And while quantitative approaches may be very effective at answering some questions, they bare an increased risk of answering the wrong questions. While qualitative questioning is normally developed iteratively, with new issues--and thus lines of questioning--emerging as the research proceeds, quantitative tools are less open to ongoing change. With surveys normally designed in locations far from where they are administered, they can tend to reflect preconceptions and biases in researchers as to what questions are important and what the answers to them may be (Rao and Woolcock, 2003). Qualitative approaches can help in some of these respects. By examining processes, by looking at historical sequences of events, and gauging respondents' views on the reason for these events, qualitative tools can help to explain why events occur and phenomena are observed. In remaining relatively open to the issues that may come up, qualitative interviewing allows for the emergence of unexpected findings. The weaknesses of qualitative approaches are well documented: normally those questioned are not randomly sampled, and thus it is more difficult to make general statements about the population as a whole; sample sizes tend also to be smaller, making generalization all the more difficult; and it is easy to unknowingly build in bias, with interviewers leading interviewees to certain answers. Further, controlling for external variables is difficult, relying on subjective disaggregation. The methodology of this study was designed in a way that takes advantages of the strengths of each approach, while controlling for the weaknesses. A number of instruments were used in order to get at both the breadth and depth of understanding necessary to fully answer the research questions (see Figure C1 below). Qualitative approaches are needed to identify the key mechanisms that trigger, sustain, and resolve (or fail to resolve) conflict; to identify the right kinds of questions (and their wording) for inclusion on a more general quantitative survey; and to provide vital context-specific information. Seven months of qualitative quasi-anthropological fieldwork was conducted to develop case studies of how different actors--villagers, facilitators, local leaders--together negotiate (or fail to negotiate) different types of conflicts in different settings. Using a modified version of the process tracing method (Bennett and George, 1997; Varshney, 2002), researchers investigated `conflict pathways'. Researchers sought to understand the discrete stages in the evolution of conflict, which include conflict triggers, factors that sustain conflict, and the factors and mechanisms that escalate or control conflict. By doing this, researchers were better able to identify the factors that transform underlying social tensions into different outcomes (violence or peace). The case 20 studies were selected in a way that controls for some of the traditional weaknesses of qualitative approaches (see Section D on sampling of research sites and cases). Complementing the qualitative work, survey instruments--including both a focused key informant survey, a larger but less focused key informant survey (PODES) and a nationwide household survey (GDS)--will be used to test the generality of hypotheses emerging from the qualitative work, the extent to which findings hold across a broader range of research sites (villages, sub-districts, districts, and provinces), and to elicit numerical data on: (a) the performance of KDP and its links to conflict and its resolution, (b) types of problem and problem solving, and (c) the basic demographic and institutional characteristics of the settings in which these occur. Importantly, the quantitative instruments were designed and developed while the qualitative work was being conducted; as such, their design reflects the ongoing findings of--and methodological lessons learned from--the qualitative fieldwork. Secondary data research about the field sites was conducted before the intensive qualitative fieldwork began; this yielded important data in its own right, and that was also used to inform the design and implementation strategies for the qualitative and quantitative work to follow. In addition, post-qualitative fieldwork analyses of archives of local newspapers will lead to the creation of datasets recording each violent conflict `incident' in the research districts and--with the use of other secondary data sources such as police criminal data--will allow for estimations to be made of aggregate levels of violence, as well as the characteristics of the incidents (conflict type, actors involved, impacts, etc.), in the research areas.8 When combined with other secondary data sources, including police data and information from health care providers, these will be used to estimate levels of violent conflict in our research areas. Figure C1: Methods Used Quantitative (breadth) Decentralization survey (GDS) Qualitative Village Potential Survey (PODES) Newspaper analysis (depth) Key informant survey Case studies 8For a number of reasons survey instruments do not tend to be particularly useful in establishing aggregate levels of conflict. Given the fact that conflict is a non-random and relatively scarce event, and given the reality of limited budgets and hence sample sizes, it is difficult to use household survey instruments in order to accurately capture levels of conflict in a locale. Key informant surveys overcome some of these problems but can encounter others such as massive underreporting of conflict--especially if they are done with government officials who have incentives to make the area under their jurisdiction look good. Further, the need to build trust with respondents before they will talk about sensitive issues such as conflict, can also lead to underreporting in large-scale surveys where interviewer-respondent time together is limited. See Barron, Kaiser and Pradhan (2004) for more on the shortcoming of survey instruments in measuring conflict. 21 The following sections will outline in more depth the different phases and elements of the research. 2. Phases of Research There were a number of phases of the research. In practice these activities overlapped, with a number occurring simultaneously. Figure C2 summarizes (in roughly chronological order) the different periods of research. The following sections outline the qualitative fieldwork schedule from the selection of the provincial and district research sites, to the preparation work, including recruitment and training, to the qualitative fieldwork itself. Following this, the survey preparation phases are then outlined (see Section C7). Information about sub-district and village sampling procedures is given in the next section (Section D). Further details on the qualitative fieldwork and survey preparation are included throughout. Figure C2: Phases of Study Activity When? Period of Time Preparation 1 Sampling of Research Sites for Qualitative Work (phases 1and June 2002­January 2003 7 months 2) 2 Recruitment of qualitative researchers August­November 2002 4 months 3 Pre-fieldwork/ background research October­December 2002 3 months 4 Training of qualitative researchers (phases 1 and 2) January, March and June 2003 5 weeks 5 Sampling of Research Sites for Qualitative Work (phase 3) October­November 2003 2 months 6 Training of qualitative researchers (phase 3) November 2003 2 days Qualitative Fieldwork 7 Qualitative Fieldwork (Phase 1) January­February 2003 4 weeks 8 Qualitative Fieldwork (Phase 2A) March­May 2003 6 weeks 9 Qualitative Fieldwork (Phase 2B) June­August 2003 7 weeks 10 Qualitative Writing Workshops (Phase 2A and 2B)--district April and August 2003 5 weeks and provincial levels 11 Qualitative Fieldwork (Phase 3A) December 2003 3.5 weeks 12 Qualitative Fieldwork (Phase 3B) February 2004 4 weeks 13 Qualitative Writing Workshops/debriefs (Phase 3) January and March 2004 2 weeks Quantitative Work and Newspaper Data Collection 14 Key Informant Survey Preparation September­November 2003 3 months 15 Preparation of modules for household surveys May 2003­March 2004 10 months 16 Collection newspaper data November 2003­April 2004 6 months Analysis 17 Preliminary qualitative analysis: provincial reports September­March 2003 7 months 18 Key Informant Survey Analysis February­April 2004 3 months 19 PODES Survey Analysis October 2003­February 2004 5 months 20 GDS Analysis October 2004­January 2005 4 months 21 Newspaper Data Analysis April­June 2004 3 months Dissemination 22 Conferences, presentations, etc. Ongoing Ongoing 23 District and sub-district feedback workshops May 2004 1 month 24 National workshop July 2004 1 day 22 3. Provincial and District Sampling 3.1 Provincial Research Site Selection The objective of the provincial selection for the qualitative work was to pick two very different provinces in which to work. The rationale for this is that since we are looking for common patterns of project impact, our findings will be strengthened if they hold up in two different settings. We focused on a range of variables to help determine the nature of a province. These included: · Population size and density · Ethnic homogeneity · Religious homogeneity (and dominant religious group) · Overall level of provincial development (roads, etc.) · Access to resources Using these criteria, we selected two provinces as different as possible from one another to compare. In limiting our selection we excluded those provinces dominated by higher levels of conflict. We did this for two reasons: first, the nature of the KDP project; and, second, our interest in local level conflicts and conflict resolution mechanisms. With regards to the first, a driving factor behind our decision to exclude high-conflict sites was our assumption that development projects are most likely to have a measurable effect at the level at which they operate. KDP operates at the sub-district level and below, with decision-making forums held within the sub-village, village and sub-district levels. If the project does produce any positive externalities that help communities constructively manage conflict, it is thus likely to be in aiding the management of conflicts that exist at the sub-district level and below. Given the nature of conflict in many high-conflict provinces, where cleavages exist on a provincial or at least district level, if we had selected such provinces we would have biased our research against observing any project impact. Further, in areas of high-conflict, where violence levels are affected significantly by external actors and exogenous factors, such as military action, it would be much harder to separate out the potential impact of a local level project from all the other causal variables in the research site. Second, we are interested in local level conflict and conflict resolution mechanisms. As we noted earlier (Section A1), the research study is concerned with the way in which common structural changes in Indonesia--economic, political, and cultural--have played out in different ways locally. Examining local level conflicts is a lens through which to examine this. However, we were concerned that by conducting research in sites dominated by large conflicts, we would miss out on finding out about these local level conflicts. It is much harder to get people to talk about lower level forms of conflict when higher level ones dominate. This lesson had been learned in previous research conducted by an author of this report.9 After extensive consultation and a round of discussion in Jakarta, with key staff from UN agencies, the central government's planning board (BAPPENAS), non-governmental organizations working in conflict areas, and World Bank staff familiar with the macro-social and economic indicators listed above, we selected East Java and NTT as the sites for the qualitative research. Of course, it is also necessary to test the limits of the types of conflicts that KDP can potentially help with, and this involves looking at the ways in which the project interacts with conflict and conflict management capacity in higher-level conflict areas. As such, the quantitative research component will also run in higher-conflict provinces. The surveys will be implemented in two stages: in February in the 9On conducting research in Central Kalimantan in 2002, following the massive outbreak of communal violence in 2001, it was more difficult to explore issues related to lower levels of conflict with informants, as local people's lives were so dominated and distorted by the impact of this larger conflict (on this see Smith, 2004). 23 two qualitative research provinces (East Java and NTT), plus in another SERP province (to be decided).10 In July, the instruments will be used--in slightly modified form--as the full baseline survey for SERP and will be implemented in the other SERP sites. As such, the surveys will eventually run in eight provinces. Conducting research in the extra six provinces, which have experienced generally higher levels of conflict than the two provinces where we conducted the qualitative work, offers the opportunity to see whether the findings from the `middle' or `lower' order conflict provinces also hold true in areas with higher levels of violence. (See Section C7 for fuller details of the quantitative arm of the research). 3.2. District Research Site Selection In each province, two districts were selected--one with a `high capacity' to manage conflict, one with a `low capacity' to manage conflict. These were selected after extensive consultation at the provincial level with government, international and local NGOs, regional development experts, universities, and KDP staff. Picking both `high' and `low' capacity districts for each province allows us to defend our claims regarding the nature and extent of KDP's impact on local conflict resolution by showing that they take place irrespective of whether the broader environment is `conducive' to conflict resolution or not. The selected district sites were: Pamekasan (low capacity, East Java), Ponorogo (high capacity, East Java), Manggarai (low capacity, NTT), and Sikka (high capacity, NTT). Our method of selecting districts for the new provinces--the SERP provinces that were not locations for qualitative research--differs. Rather than selecting `high' and `low' capacity provinces, the surveys will be implemented in the SERP districts, which, by definition, experience high levels of violent conflict and hence, most likely, have low capacity to manage it.11 4. Preparation and Pre-Fieldwork 4.1 Recruitment Around four months were spent putting together the research team. Eleven researchers and four provincial coordinators were recruited to make up--and supervise--four research teams, one for each of the districts studied.12 This involved initial over-hiring to account for the possibility that researchers might drop out during the training periods, and to ensure that those who remained were truly suitable for the task. Recruitment was an inevitably time-intensive task. A range of skills and expertise was required from the researchers, and, aside from individual attributes, we also needed to build a balanced team. A first requirement was prior experience conducting qualitative fieldwork and spending extended periods of time in Indonesian villages. We also looked for those who had formal education to a bachelors level. (In fact, a number of candidates had gone further than this, with some having undertaken graduate courses in Australia and Singapore. Three were on the faculty of Indonesian universities.) Understanding English, although not necessary for the researchers' primary task of collecting data in the villages, was a third requirement, necessary for us to facilitate training sessions and supervision. While local language proficiency was not a prerequisite as such, we did seek to build district research teams with members fluent on the local tongue(s). For the Ponorogo (East Java) team this was easy: almost every candidate we interviewed in Java was fluent in Javanese. For the Madura (Pamekasan) team, finding qualified researchers with local language capacity was more challenging. We ended up hiring only one Madurese speaker as a result. 10SERP (Support for Economic Revitalization Program) is a new World Bank-financed project directly addressing conflict issues in Indonesia. 11More broadly, there were a number of factors for districts being selected as SERP project locations, including: (1) degree or impact of conflict; (2) opportunity or likelihood for positive project impact on the conflict; and (3) need for the specific service offered by the project component. 12See Figure C2 for a diagram of the study's staffing structure. 24 For Flores the situation was even more challenging. In NTT, and in Flores, there are two main language groups--the Austronesian (AN) language family, and the non-Austronesian (non-AN or Papuan) languages. The languages of each group are built on fundamentally different structures. Furthermore, there is significant diversity within each group. In Flores there exist 28 main languages--a mix of AN and non-AN--but even within each there are a number of dialects that can make communication very difficult (Grimes et al., 1997). Other people we spoke to put the number of distinct languages on Flores at more than 50. What is more, the languages are often so different that people in neighboring districts and even sub-districts cannot understand each other at all. Complicating things further, the language map of Flores does not directly correspond to the political district map. In other words, there are many languages within each district; and languages are often spoken over the boundaries of two (or more) districts. Thus our Sikka research team needed to have team members who spoke both Bahasa Sikka and Bahasa Lio. In Manggarai, three main language groups ideally would have been covered.13 Given the size of the language population groups--in some cases much less than 100,000 people--and the other criteria researchers had to meet, finding linguistically balanced research teams was a challenge. There are literally only a handful of people in the world with all these qualifications and who were both good and available for the numerous months this study required. In the end, we managed to cover both of the main local languages in Sikka, and two of the three main tongues in Manggarai. Across our four district research teams, only three researchers did not speak the local language of the area they worked in. A word should also be said about the difficulties we had in ensuring our teams were gender balanced. For the East Java teams, we were initially successful in achieving a gender balance. However, after three of our female researchers dropped out or proved unsuitable, we were unable to find other women to take their place. As a result, we ended up with only one woman out of five East Java researchers. In NTT, we experienced even greater difficulties. Few women came forward to be interviewed; the vast majority of those who did were not fully qualified. Even if we had adopted a quasi-affirmative action policy, we still would not have been able to achieve a gender balance in the Flores teams. In the end, only one out of six Flores researchers was a woman. The problems we had in hiring women was indicative of the imbalanced nature of opportunities and freedoms that exist between genders in Indonesia, as in many other places. This posed some serious problems given the difficulty of male researchers accessing female respondents (see the methodology section of Barron and Madden, 2003a). We thus encouraged the all-male research teams to informally hire local women to help them conduct interviews and, especially, focus group discussions. Our two women supervisors were also able to help in this respect. 13While most people in Flores speak good Bahasa Indonesia, conducting interviews in the national tongue was, in most cases, a second best option. Many of the most marginalized elements of the population we are studying--women, the poor, the elderly-- only speak local languages. Using local languages is an important mechanism for building trust. And in addition, from an anthropological perspective, the fewer layers of `interpretation' that need to go on, the better. 25 Figure C2: Core Staffing Structure for Study14 DC-based research advisor Jakarta management team Training Coordinator Michael Woolcock Claire Smith, Patrick Barron Yatrin Kaniu Jakarta-based research advisor Leni Dharmawan Translators Joanne Sharpe East Java supervisors NTT supervisors Kristen Stokes Rachael Diprose Adam Satu Sue Piper Saifullah Barwani Jessica Gillmore Ponorogo Researchers Pamekasan Researchers Sikka Researchers Manggarai Researchers Cici Novia Anggraini Lutfi Ashari Olin Monteiro Peter Manggut Imron Rasyid Mohammed Said Don A dela Santo Agus Mahur Endro Crentantoro Stanis Didakus Yan Ghewa 4.2 Preliminary Research Once the researchers were in place, three months of `pre-fieldwork' activity helped support the design of the formal qualitative fieldwork and the survey. This work--which took place for approximately three months, concluding late December 2002--included the collection of statistics at provincial level, and basic analyses of local newspapers and previous academic research. Ten preliminary reports were produced by the research team. These provided statistical data, preliminary conflict maps, outlines of historical processes, and summaries of some of the most prominent cases in the research sites. Preliminary interviewing of both government and non-government actors at the provincial and district levels took place. The primary aims of this initial work were three-fold: first, to start the process of mapping conflict (its spatial distribution and its characteristics) to help in the sampling of districts and sub-district sites; second, to provide input into the development of the research hypotheses, questions and the data collection instruments; and, third, to gather background secondary research (statistics, academic writings, newspaper articles, etc.). However, a secondary aim of this stage proved, in hindsight, to be perhaps the most important. `Socializing' the researchers into the project, spending time mentoring them, discussing the aims of the research and, more broadly, the reasons for conflict in Indonesia were vital to the success of the research. A mission to Kalimantan, a seminar in Flores, and `get togethers' in the East Java provincial capital (Surabaya) all gave us a head start on the research. For the vast majority of researchers, a knowledge base was already in place by the time the first formal training began in January. 5. Training Four formal training sessions for a total of approximately five weeks were conducted for project field staff.15 These sessions served the dual purposes of preparing the field staff for the qualitative research and of allowing the methodology to be refined iteratively to allow the inclusion of local expertise. 14For the first three months of the study, David Madden was part of the Jakarta management team. Many other people were involved in the development of the study; they are thanked at the beginning of this paper. The personnel here were those that undertook the bulk of the qualitative research (phases 1, 2A and 2B). For phase 3, each district team consisted of only 2 researchers. 26 5.1 Training 1 Before the field research started, two weeks of training introduced the researchers to key issues to prepare them for the preliminary research phase. The training session took place in January 2003, in Jogyakarta, Central Java. Instructors came from both the World Bank and outside agencies. The training included classroom and field practice sessions on: · Case study methodology · Interview techniques for researching conflict · Guidance in interviewing government and health officials · Gender sensitivity in village research methods · Introduction to KDP mechanisms · Background review of research sites · Understanding and mapping local level institutions · Measuring governance--indicators of good governance · The research tools and data recording formats 5.2 Training 2 After the first Phase of research, we held a ten-day training and debrief workshop with the research team. The workshop took place in Kuta, Bali in March 2003. The workshop involved classroom training as well as a one-day field trip. The three main objectives/outputs of the workshop were as follows: · Review and cross-sharing of preliminary research findings: o Building maps of social tensions and conflicts o Pin-pointing interesting cases to follow up in the second phase of research o Providing feedback for the SERP design team in summary form o Revising village selection for Phase 2 of the research · Refined research instruments for Phase 2 · Researchers trained in the following skills: o Understanding the research hypotheses o Understanding the research areas of interest o Conducting focus group discussions (classroom and field practice) o Undertaking participant observation and informal interviewing o How to use the new data formats for research Phase 2 5.3 Training 3 We held a third, seven-day training workshop between the two phases of village-level research (Phases 2A and 2B). The workshop took place in Bandung, West Java in June 2003. There were, again, three main elements of the training workshop: · Debrief of findings from Phase 2A of the research, both substantive and methodological · Input into the design of the survey instrument. · Briefing on the three new areas of questions (see topics 14-16 in Annex A). The workshop primarily took place in the classroom, but also involved a two-day field trip to nearby villages where the researchers practiced the new KDP-related questions. 5.4 Training 4 We held a fourth, short training session for each of the provincial teams in late November before Phase 3 of the research. The workshops took place in Ende (NTT team) and Surabaya (East Java team). 15See the field research guides produced for each phase of fieldwork for more details on the material used for the trainings. 27 The purposes of the training were: · Give background on the aims of Phase 3 and discuss the hypotheses to be explored · Train the researchers in implementing the key informant survey · Go over the new formats the researchers would use in this phase, and the outputs for Phase 3 · Discuss sampling issues 6. Qualitative Fieldwork In each of the four selected district sites, research took place in two kecamatan (sub-districts). Within each district we selected a KDP sub-district and a non-KDP control sub-district. (See Section D1 for details on how and why this was done). Qualitative fieldwork took place in the eight sub-district sites from February to August 2003. Four teams of 2-3 researchers16 (one team for each district) spent: (1) Research Phase 1: four weeks at the district and sub-district levels; (2) Research Phase 2A: six weeks mainly at the village level in a non-KDP sub-district (plus two weeks of in-field write-up and debrief); (3) Research Phase 2B: six weeks mainly at the village level in a KDP sub-district (plus two weeks of in- field write-up and debrief). In addition, a third phase of research (in two parts: 3A and 3B) took place between November 2003 and March 2004 and expanded the research into an additional two sub-districts in each of the research districts. 6.1 Qualitative Fieldwork Phase 1 The first phase of research ran from 10 February to 5 March 2003. Researchers spent approximately one week at the district level, one week in the KDP sub-district, and one week in the non-KDP sub-district. They spent the last week finishing writing up data formats and analytical pieces, and debriefing with the provincial supervisors. The first phase primarily aimed to get an overview of the situation in the district and sub-district. Conflict mapping is a vital prerequisite for any more in-depth research. We needed to know the basic spatial distribution of conflict, the types that exist and their intensities, the actors involved (including victims), and the broad impacts in order to be able to select sub-district research locations. The information gathered in this phase thus informed sampling and the development of the methodology, as well as questions to be asked, informants to interview, and research topics to explore. We pursued this basic conflict mapping through interviews with, in large part, authority figures at the district and sub-district level. These included representatives from government and the civil service, NGOs, religious and adat leaders, police and military, and health care service providers. (For details, see informant list in Section E1.) By the end of this phase we had collected basic aggregated data on: types, levels and locations of violent conflict; types, levels and locations of social tensions; levels and type of injuries (and deaths) from conflict; main conflict resolution mechanisms (at district and sub-district level); and levels of crime. We used this data in the selection of villages and cases to follow in the second phase of research, in conjunction with data on which villages actually participated in KDP (see Section D2 for details on village selection). Researchers recorded data on a range of formats (see Section I for an outline of the data formats that were used in this phase, and Annex B for the actual data recording formats). This phase of the research produced approximately 120 interview transcripts.. 16See the researcher list above. Some teams hired additional local helpers for some periods of the research. 28 6.2 Qualitative Fieldwork Phases 2A and 2B Whereas the first phase of research took place at the district and sub-district levels, most of the research in Phase 2 took place at the dusun (sub-village) and the desa (village) levels. This was for two main reasons. First, this is the level at which KDP operates and therefore presumably where it is most likely to have a measurable impact; and, second, we are interested in the issues or factors that affect local level capacity to manage conflict, and the village and sub-village levels are the most important units for this kind of analysis.17 Researchers lived in the communities that they were studying. This was important because conflict is by nature both a sensitive and (relatively) infrequent event. As such, to better understand its impacts on people, researchers had to establish a reasonable level of rapport with villagers so that the latter would be comfortable sharing their experiences. To maximize the probability of being present when actual episodes of conflict occur, and to witness it firsthand, researchers needed to maintain a constant presence in the village. Phase 2 of the research took place in two stages, between early April and mid-August. Each stage involved six weeks of research and then two weeks of in-field write-up. After approximately three weeks of research, an in-field debrief was held (see Section C6.3 below). Three research tools were used: · in-depth interviews with a cross-section of population and key actors (see Section F); · focus group discussions with certain population groups (see Section G); · participant observation and informal interviewing (see Section H). The main aims of this phase of research were two-fold. First, we wanted to follow particular cases of conflict and conflict management and to develop detailed `conflict pathway' case studies. As noted earlier, examining the conflict `case' in detail provided a particularly useful basis for generating broader judgments about the nature of conflict and, more broadly, the social relations--between individuals, between groups, and between citizens and the state--in a society. Between the four teams, fifty-six conflict cases were tracked and written-up as conflict pathway case studies. A second purpose of Phase 2 of the qualitative research was to collect general `background' data on a range of factors that relate to the structural, economic, institutional and cultural environments of the research sites, and on levels of social tension and violence in the districts, sub-districts, and the twenty- five villages that were studied. This information complements the case studies, allowing cross-village analysis that can help us better answer questions about the factors that `matter' in dictating levels of conflict and community responses.18 6.3 Qualitative Fieldwork Phases 3A and 3B As we have seen, the first two phases of research (Phases 1, 2A and 2B) gathered detailed and comprehensive data on the first aim of this study: namely, to understand the conditions that make conflict more or less likely, that affect community's abilities to manage it, and that help determine the pathways that individual conflicts take. The third phase of research turned to the second aim: to measure KDP's impacts on those conflict pathways and, specifically, to ascertain whether KDP processes are generating any externalities or spillover effects that help communities solve their problems. 17See discussion above in Section C3.1. 18See Section C8 below for more on the concepts behind, and methods of, following conflict pathways case studies and collecting `general' data. 29 The primary reason for the relative lack of data on KDP and conflict from phases 1, 2A and 2B was the relatively small sample of KDP locations: 12 villages in 4 kecamatan. This was too small a sample because: (a) as we were only working in one KDP kecamatan per district, we did not know if the KDP performance in that district was representative of how it works in other kecamatan in the district; and (b) as KDP was not working as it is meant to in some of the districts (especially Kecamatan Lamba Leda in Manggarai, and Kecamatan Proppo in Pamekasan) it was impossible to find any effects. There were two implications from this: first, we needed to increase the number of KDP kecamatan (and villages) we were working in; and, second, we needed to be sure that the KDP kecamatan had at least a minimum level of KDP performance (i.e. the program needed to be basically working as it ought to).19 As such, Phase 3 extended the research to two more kecamatan in each research district, with an additional two villages per kecamatan (thus four more villages per kabupaten). All of the villages and kecamatan that we visited in Phase 3 had had KDP for at least three years: either 3 years of KDP1, or 2 years of KDP1 and the first year of KDP2.20 The data from these extra villages, combined from the information from Phase 2, allowed us to test more thoroughly some of the hypotheses about how, and why, KDP may impact on communities' capacity to manage conflict. In addition to examining more closely the relationship between KDP and conflict (and conflict management), Phase 3 also involved the collection of some standardized qualitative and quantitative data. As such, in research Phase 3 we conducted a range of standardized interviews primarily relating to KDP and its links to conflict and conflict resolution. We conducted this key informant survey in both the new villages we worked in and in the original KDP research villages. 6.4 Qualitative Writing Workshops A total of six writing workshops were held throughout the implementation of the qualitative research. These were organized by the provincial coordinators and were held at the provincial level in East Java, and at the district level in NTT where geographical distances were extreme between the two districts. These workshops proved to be crucial for reviewing research strategy, key findings and village selection; organizing data; troubleshooting research implementation and administrative problems; and, conducting preliminary compilation and analysis of the data collected. They also proved to be an important team- building exercise, where we shared findings and discussed strategies to overcome difficulties associated with accessing respondents and collecting data. (i) Research Writing Workshop 1: Post Phase 1 The first writing workshop was held to write analysis pieces from the data collected in the interviews with stakeholders conducted at the district level. Key outputs for each district included analysis pieces examining the: · Socio-cultural background of each district; · Main types of conflict in each district/research sub-district; · Conflict map of the spatial locations of each type of conflict; · Overview of the main conflict mechanisms in each district; · Overview of the underlying causes of conflict and social tensions in each district. (ii) Research Writing Workshops 2 and 4: Mid-Phase 2A and 2B These writing workshops constituted a mid-way debrief for the researchers to review their village selection, case strategy, findings from the field to date, and to begin the process of case chronology tracing for each of the conflicts being researched. It also proved useful for meeting administration deadlines and organizing the data from the first three weeks of the phase. 19Sample selection based on KDP performance means that we are testing for KDP impacts where the program is functioning relatively well, and not for KDP impacts more broadly. 20In some cases (in Ponorogo and Sikka) the kecamatan have had KDP for 4 years ­ 3 years of KDP1 and 1 year of KDP2. 30 (iii) Research Writing Workshops 3 and 5: End Phase 2A and 2B We held the primary writing workshops at the end of each of the two phases of village-level research, each of which lasted for approximately two weeks. As with the other writing workshops, they were primarily used to review findings from the field, organize data, and conduct preliminary analysis. However, the most important outputs from these workshops included written case studies detailing the conflicts researched and preparing village profiles capturing general data from the village not specific to any single case. (iv) Research Writing Workshop 6: End Phase 3B A final writing workshop was held for one week at the end of Phase 3B. This gave an opportunity--with supervisors close at hand--to finalize outputs from this phase (key informant surveys, in-depth interviews, diaries, case studies, analysis pieces, etc.), before the researchers returned to their home locations. 7. Quantitative Research As noted above, this study uses both qualitative and quantitative research tools. The qualitative work was helpful for examining questions of process: why conflicts escalated or dampened; the role of particular actors (including KDP) in determining the pathways that individual conflicts take; and examining in detail such normative and cognitive constructions as `identity' and `exclusion', which cannot easily be reduced to numbers. Key informant and household surveys will complement the data derived from the qualitative fieldwork. Random sampling, and a much larger sample size, will allow us to generalize findings in a way not possible from the qualitative work. Given the dominant development discourse, numerical data from a quantitative survey will allow the research results to have a greater impact than a stand-alone qualitative study. Three basic survey instruments were, or will be, used in the study:21 7.1 PODES Survey The Government of Indonesia's Central Bureau of Statistics' (BPS) Village Potential series (PODES) is a long-standing tradition of collecting data at the lowest administrative tier of local government. It collects detailed information on a range of characteristics ­from infrastructure to village finance ­ for Indonesia's current sixty-nine thousand villages and neighborhoods. The latest PODES was fielded at the end of 2002 as part of the 2003 Agricultural Census. For the first time, the 2003 PODES included a section on politics, conflict and crime. If conflict was reported, the survey asked the type of conflict, whether it was longstanding, and the impacts of the conflict (fatalities, injuries, and material damage). The survey also asked if the conflict was resolved, and if so, by whom (citizens, village apparatus, or security apparatus). The PODES data helped to map out the incidence of conflict and violence across Indonesia. It also allowed researchers to test a range of basic hypotheses about factors correlated with higher levels of conflict (Barron, Kaiser and Pradhan, 2004). 7.2 Governance and Decentralization Survey (GDS) The Governance and Decentralization Survey (GDS) is part of the Indonesian Decentralization Empirical Analysis (IDEA) project, which is being conducted by the World Bank together with CPPS-GMU (Centre of Public Policy Studies-Gajah Mada University). The survey aims to ascertain the effects and impacts of the rapid decentralization that took place in Indonesia in 2001 on a number of factors ranging from the 21In addition to the three surveys outlined below, it is expected that questions on conflict will also be included in the national quantitative KDP evaluation, due to be administered in late 2004, early 2005. This may thus form a fourth quantitative data source for the study. 31 performance of local governments, to service delivery, to the functioning of the justice system. The initial (baseline) enumeration took place in 150 (almost half) of Indonesia's districts in 2001. As part of the follow-up survey, to be implemented in late 2004, a module on conflict and problem solving will be inserted into the survey, and the sample will be extended. The GDS will thus provide a rich data-source on levels of conflict (as reported at the household level) and data on a range of other `institutional' and `governance' factors that can be regressed against conflict/violence levels. The formulation of this module was based on the emerging hypotheses from the qualitative fieldwork, newspaper analysis and the results of the PODES survey. Methodological insights from the latter also affected the survey design. 7.3 Key Informant Survey In Phase 3 of the qualitative research, the research team spent around fifty per cent of their time administering surveys to a range of key informants at the sub-district and village levels. The questionnaires (each of which took around 1.5 hours to complete) focused largely on the role of KDP in the locality, the extent to which it was used to solve (KDP and non-KDP) problems, and the extent to which `spillover' effects could be determined. In each of the four research districts, the surveys were administered in three sub-districts: two `new' kecamatan, each of which had had KDP for at least three years; and the previous KDP `treatment' site. In each of the kecamatan, the survey was conducted in two villages, thus giving a sample of twelve kecamatan and thirty-six villages. The surveys were administered to three informants at the sub-district level, and eight at the village level ­ thus giving a sample of approximately 324 respondents.22 7.4 Aims of Quantitative Research There are four primary aims of the quantitative research: First, the PODES has gathered, and GDS surveys will gather, primary numerical data on levels and types of conflict in the research areas. Doing this is challenging, since few instruments exist that measure most of these factors, and conflict questions are conspicuously absent from most quantitative studies. Indeed, survey instruments, of either household or key informant type, are probably not the best method to collect data that can be aggregated on levels of violent conflict.23 However, the surveys can generate data on `problems' that exist in villages and the ways in which people resolve them. Developing these questions has been a long process, requiring substantial time to adapt questions to take account of local realities and sensitivities.24 In addition, the survey will be used to collect data on levels of crime, perhaps the next best proxy for levels of violent conflict (Fajnzylber, Lederman and Loazya, 2000). This, when combined with secondary data sources such as police and health provider data, as well as newspaper analyses, will produce quantitative data on levels and types of conflict in the research areas. Second, this basic mapping will allow us to test the generality of hypotheses emerging from the qualitative work about the causes and triggers of conflict. We have a number of hypotheses about the factors that determine levels of violent conflict and the ability of communities to manage them (see Section B for hypotheses). To some extent, the qualitative work alone will be able to test these ­ the `general' data collected (see next section) can be compared across villages and `regressed' against what we know about levels and types of conflict in the research areas, both from the primary fieldwork and secondary data collection. However, by increasing our sample size, and by ensuring random selection of households in the case of GDS (and thus greater representativeness), the quantitative work will provide a stronger empirical base against which to test the validity and generality of our hypotheses. In particular, we want to measure three types of variables that may have an impact on levels of conflict and communities' ability to manage them:25 22The key informant survey may be expanded to more locations in the future. 23See footnote 8. 24At least three sessions were spent developing the survey instruments with the qualitative researchers who were able to provide guidance on what questions would `work' and what would not. 25This typology of `conflict-causing factors' was developed by Barron and Madden (2003b). 32 (1) Economic/Structural Factors · economic factors, particularly economic disparities between groups (and trends) and social mobility; · religious homogeneity; · ethnic homogeneity; · demographic factors, including percentage of youth, gender ratio, etc. (2) Psychological/Cultural factors · levels of trust (between different `groups' in the village, and between villages); · levels of civic interaction and mixing within and between villages (especially inter-group mixing) in both formal and informal arenas. (3) Social and Political Institutional Factors · inclusiveness of different institutions (formal and informal); · ease of access to and participation in social, economic, and political institutions; · the problem-solving mechanisms people turn to when there are disputes, and their success/failure; · satisfaction with decision-making processes and outcomes; · information on involvement in KDP (in the KDP sub-districts). Third, using recall strategies, the survey will also collect retroactive data on the communities being studied, in partial compensation for the fact that we do not have baseline (pre-KDP intervention) data. In the absence of baseline data, this will give us a platform against which to measure post-program changes. (The qualitative work--when done well--will also do this, but the approaches can complement each other). Fourth, the key informant survey will collect data on: (a) how well KDP is working, and (b) its relation to conflict and conflict management. We can assume that a basic precondition for any of the various hypothesized ways in which KDP might improve conflict management capacity is that KDP is functioning as intended.26 We will thus measure aspects such as citizen participation in, and knowledge of, KDP processes.27 8. Types of Data The qualitative research gathers two different types of data: case data and general data; the quantitative instruments will gather the latter. This section will outline each in turn. 8.1 Case Data As discussed above, the qualitative research primarily uses a case-based approach to understanding conflict and conflict management. By looking in-depth at specific cases of conflict, we can better understand the factors that led to the conflict and the particular path it took. To this end, the research generated over sixty `conflict pathways', studies of how particular social tensions and incidents of conflict played out in their local context. Conflict pathways examine how different actors ­ villagers, facilitators, local leaders, etc.--together negotiate (or fail to negotiate) different types of conflict in different settings. Using a version of case-based process tracing (Bennet and George, 1997; Varshney, 2002), the conflict pathway case studies developed will help us to understand the discrete stages in the 26To recap, the five major hypothesized ways in which KDP might result in increased conflict management capacity are: (1) cognitive change--improvement in civic skills and trust between groups; (2) institutional change I--creation of new forums to solve problems; (3) institutional change II--increased participation of marginalized groups; (4) cultural change--value change creating a `culture of peace'; and (5) outside actor--presence of impartial non-political mediator in the form of KDP facilitators. (See Section B3). 27Further details regarding the quantitative elements of the research are forthcoming. 33 evolution of conflict. These include: conflict triggers; factors or mechanisms that sustain conflict; factors or mechanisms that allow conflict to escalate (or that control it); factors or mechanisms that resolve conflict. By following through cases of conflict, we will be able to identify the factors that transform underlying tensions into different outcomes--violence or peace. In essence, we worked backwards from the outcome (peace or violence) by asking what led to what. Overall, the case studies sought to: · Analyze processes of particular conflicts and cases of conflict management · Identify why social tensions erupt into violence (or why they do not) · Identify examples of mechanisms that work (or do not) in preventing conflict from turning violent and escalating · Disaggregate which factors (social, economic, institutional) helped cause and sustain the conflict through looking at the process and development of incidents of conflict In research Phase 2B and Phase 3 (in the KDP sub-districts) case studies also examined the way KDP interacts with conflict.28 Case studies in KDP locations sought to: · Examine the extent to which KDP acted as a conflict management forum · Identify the links between KDP and the social tension itself (e.g. did KDP actors, rules or processes help fuel tensions, or did it help to calm them?) Thus a series of discrete conflict pathway case studies was built, showing how particular conflicts developed, and the factors that led to escalation, dampening, management, or resolution of those conflicts. As explained below, primary cases were selected based on information gathered in the first phase of qualitative research. The basic `mapping' that took place in Phase 1 allowed us to pre-select a number of cases to follow (see Sections D2.7 and D2.8 for details of the pre-selected primary cases; see Sections D2.1 ­ D2.6 for the criteria for selection). In addition, secondary cases were also followed--interesting cases that were discovered during the village level research but not beforehand.29 Whenever researchers heard about a specific conflict case in an interview or focus group discussion, they filled out a case study format (see Annex C and Annex D for relevant formats). The different formats were then brought together to form multi-perspective conflict pathway case studies (see Section J on writing the conflict pathway case studies). As noted earlier, the research has resulted in over sixty of these case studies. The writing process behind these case studies, in itself one of the most useful steps in the process of unraveling and understanding conflict evolution and resolution, is detailed in Section C6.3 above. 8.2 General Data In addition to following up specific cases, the researchers also collected `general' qualitative and quantitative data on conflict, and related information on the village and the sub-district. We did this to help identify a broader picture of conflict management capacity throughout the village and sub-district. The `general' data helped to illuminate the cases we followed, gave us a broader picture of the `conflict map' of the research sites, and helped us to test the generality of the hypotheses that emerged from the conflict pathway case studies. This data provided us with non-case-specific information on the following points: 28As we were also looking in depth at the relationship between KDP, local actors, and local institutions, we allowed an extra week for the qualitative research in Phase 2B. Phase 2B therefore lasted for seven weeks rather than six (as in Phase 2A). In addition, as noted above, it was later decided that the sample of KDP locations should be expanded; as such, an additional two months (over two phases) of fieldwork was completed in additional KDP locations. 29A list of final cases completed in Phase 2 of the research is given in Figure D1. The case list from Phase 3 is still being finalized. 34 · The role of different institutions--both formal and informal--in decision-making and problem solving · The economic and social structure of the communities in the research sites · The construction and maintenance of the boundaries of the main identity groups · The forms of social cohesion and civic interaction in the research sites · The nature and extent of conflict · The main conflict resolution mechanisms, and their success or failure · The factors that may affect levels of conflict · The relationship between KDP and social tensions · The relationship between KDP mechanisms and conflict resolution (i) Collecting General Data Using Qualitative Methods As noted above, our unit of analysis is not just the conflict per se but also the sub-district. By this we mean that we are treating the sub-district as case studies in their own right. We thus want to get a broader picture of conflict management capacity throughout the sub-district. One of the best ways of doing this is to start with actual cases of conflict and its management (see above). However, other data can also be collected qualitatively that will give us a broad picture of social tensions and conflict management capacity. In the first phase of research, researchers were asked to collect `general' district- and sub-district-wide data on a range of factors including: types and levels of conflict; causes of conflict; locations of conflict; parties involved and their characteristics; and underlying tensions. This data was non-case specific as it generally referred to district- or sub-district-wide trends and phenomena. Aggregated impacts of conflict data were also collected from health care officials (see Annex B for relevant formats). In the second phase, researchers were asked to collect `general' data on the topics listed below: 1. Socio-Economic Factors a. Inequality b. Unemployment c. Socio-Economic Mobility and Illegal Activities d. Occupations e. Education 2. Ethnic Composition a. Ethnic Composition b. Main Group Identities c. Construction of Group Identities 3. Formal Governance a. Trust in, and Problem Solving by, Village Government b. Trust in, and Problem Solving by, Kecamatan Government c. Participation in Village Government Decision-Making d. Participation in Kecamatan Government Decision-Making e. Knowledge of Government Decisions and Processes, and Access to Information f. Predictability of Decision-Making g. Accountability and Representation h. New Institutions 4. Adat a. Role of Adat b. Trust in, and Problem Solving by, Adat/Traditional Leaders c. Accountability and Representativeness of Adat d. Participation in Adat Decision-Making e. Adat Practices 35 5. Religious Organizations a. Role of Religious Organizations b. Trust in, and Problem Solving by, Religious Leaders c. Inter-group Participation d. Tensions 6. Civic Interaction a. Associational Interactions b. Social Interactions 7. Law Enforcement/Justice System a. Police b. Positive Legal System c. Tensions between Positive/Adat/Religious Law 8. Military 9. Crime/Vigilantism 10. Population Flows a. Displacement/Migration b. Female-Headed Households 11. Other Conflict Resolution Mechanisms30 a. Functional Groups b. Village Security Groups c. Illegal Groups d. Other Groups 12. Types of Conflict a. Characteristics b. Causes c. Consequences d. Actors and Location 13. Socio-Cultural Factors 14. Development Projects (non-KDP) a. Presence and Knowledge of Development Projects b. Participation in Development Projects c. Tensions with Development Projects d. Outcomes and Impacts of Problems with Development Projects 15. KDP Process and Mechanisms a. Participation in KDP b. Competitive Bidding Process in KDP c. Proposal Writing Process in KDP d. Facilitation of KDP e. Tensions in KDP Process f. Complaints Mechanisms 16. KDP and Conflict/Conflict Resolution a. Where KDP Facilitators Addressed Non-KDP Related Conflict Issues b. Where KDP Forums Were Used to Address Non-KDP Related Conflict Issues c. Where KDP Has Had a Positive Effect on Relations between Groups d. Where KDP Has Led to Non-KDP Related Conflicts Data relating to the last three of these topics (development projects and KDP) was only collected in part two of the second phase of research--i.e., in the KDP sub-district. Researchers collected, coded and 30By `other', we mean mechanisms not included in the other sections--that is, not formal government (section 3), adat (section 4), religious organizations (section 5), police (7a), courts (7b), and the military (8). 36 recorded this data on specific formats (see Annex D). In Phase 3 of research, the number of data topics was limited to ensure a focus on KDP's role and other forms of problem solving. As such, the topics list for Phase 3 of the research was as follows: 6. Civic Interaction a. Associational Interactions b. Social Interactions 14. Development Projects (Non-KDP) a. Presence and Knowledge of Development Projects b. Participation in Development Projects c. Tensions with Development Projects d. Outcomes and Impacts of Problems with Development Projects 15. KDP Process and Mechanisms a. Participation in KDP b. Competitive Bidding Process in KDP c. Proposal Writing Process in KDP d. Facilitation of KDP e. Tensions in KDP Process f. Complaints Mechanisms 16. KDP and Conflict/Conflict Resolution a. Where KDP Facilitators Addressed Non-KDP Related Conflict Issues b. Where KDP Forums Were Used to Address Non-KDP Related Conflict Issues c. Where KDP Has Had a Positive Effect on Social Relations Between Groups d. Where KDP Has Led to Non-KDP Related Conflicts Other Forms of Problem Solving 3. a. Problem Solving by Village Government 3. b. Problem Solving by Kecamatan Government 4. b. Problem Solving by Adat/Traditional Leaders 5. b. Problem Solving by Religious Leaders 7. a. Law Enforcement: Police 7. b. Positive Legal System 7. c. Tensions between Positive/Adat/Religious Law 11. a. Other Conflict Resolution Mechanisms: Functional Groups 11. b. Village Security Groups 11. c. Illegal Groups (ii) Collecting General Data using Quantitative Methods Surveys, by definition, collect general non-case-specific data. By putting together a large-N dataset consisting of indicators and variables that get at the wide range of factors we think might matter in determining levels of violent conflict, we can test the validity and generality of a range of hypotheses. Many of the areas of information on which we wanted to collect specific survey data mirrored the general data categories the qualitative researchers used. This was deliberate; looking at a problem from more than one methodological angle gives us increased confidence that what we are finding is true, and that the inevitable, although variable, bias built into any research method is not unduly distorting findings. In short, if we get the same answers to the same questions twice, but using different methods and taking a different approach, we can be reasonably confident that the answers are as true and objective as possible. It should be noted that there will be two different types of questions in the surveys, getting at different types of general data. The first aims to get at a picture of the objective reality in the village. Thus we want to know what economic inequality is actually like, how far different sub-villages actually are from the village government office, how many people were actually injured from conflict in the past year, and 37 so on. However, we are also interested in more normative data--on people's perceptions of conflict, security, social relations, and conditions in the village. These perceptions may not mirror objectively determined reality. As we argued earlier, elements of cognition are probably as important as the structural reality in determining levels and types of conflict. Even if there is massive economic inequality, if individuals do not perceive things to be unfair, it is unlikely to be a factor around which people mobilize and raise to a level of explicit claims and hence a contributing factor to open conflict. Many--but not all--of the `objective' questions will be in the key informant survey, where the informant will, in essence, provide an objective `mapping' of conditions in the village. The importance, however, of the second `subjective' questions necessitates a household survey to try to gauge the perceptions of a wider cross-section of people. (iii) Secondary General Data Secondary sources will also generate general data on the research areas. Secondary sources of data include: · Statistics collected at the national, provincial, district, sub-district, and, in some cases, village levels, from a range of sources including BPS,31 the police and army, and health care officials; · Academic sources; · Newspaper sources.32 As noted earlier, the qualitative researchers spent much of the pre-research period gathering secondary data. They also continued to collect statistics, official documentation and newspapers when they were in the field. D. Sampling of Local Level Research Sites The qualitative research in Phases 1, 2A and 2B ran in eight sub-district sites within the four districts in two provinces outlined above (see Section C3 for details of the selection process for provinces and districts). Phase 3 extended the research into another two sub-districts within each district. This section details the process of choosing the local research sites, by which we mean the kecamatan (sub-district) and below.33 1. Kecamatan (Sub-District) Selection 1.1 Kecamatan Selection: Phases 1, 2A and 2B For Phases 1 and Phases 2A and 2B of the qualitative fieldwork, we chose two sub-districts for each of the research districts--a KDP site and a non-KDP site. The former had already had KDP1 for three years (our `treatment' sites); the latter had not yet had KDP1 (our `control' sites). We decided that it would be unfair to look for an impact if the program had been in place for less than three years. In all the non-KDP sites but one (Pamekesan) KDP2 started in 2003; thus the study will provide a range of baseline data against which the performance of KDP2 can be measured in any follow-up study.34 31BPS is the statistics agency of the Government of Indonesia. 32A systematic tracking of newspaper reports on conflict in the regions we are studying is being undertaken at present. 33The sampling strategy for the quantitative key informant survey matches that used for Phase 3 of the qualitative research (details below). PODES and GDS employed their own sampling strategies; details for the former can be found in Barron, Kaiser and Pradhan (2004); information on the latter can be found on-line at www.worldbank.or.id 34KDP2, the second installment of KDP, began in 2003. While it embodies the same principles as KDP1, there are some structural changes. In particular, it places more emphasis on strengthening the internal capacity of villages by sponsoring training programs in bookkeeping and procurement, infrastructure assessments and development planning. See the Project Appraisal Document on the web: www.worldbank.or.id. 38 Within each district, the sub-districts (KDP and non-KDP) were selected to be as similar as possible. This was to control as much as possible for non-program effects that may stem from socio-economic, institutional, or other differences. Matching and selection of sub-districts was done using mixed methods. First, the propensity score35 for each sub-district in Indonesia was estimated using the PODES and SUSENAS datasets.36 KDP and non-KDP sites were then matched--a `match' occurring where the difference in the absolute deviation of the two was less than or equal to 0.05. This allowed us to `match' sub-districts by their characteristics before KDP1 was implemented. This left us with a number of `pair options' for each district. Qualitative methods were then employed: interviewing at the district level with government officials and other experts helped us check that the quantitative selections made sense, and helped us incorporate other sources of difference/similarity. The final selections are shown below in Figure D1: Figure D1: District and Sub-district Qualitative Research Locations (Phase 1, 2A and 2B) East Java NTT High Capacity Low Capacity High Capacity Low Capacity Ponorogo Pamekasan Sikka Manggarai KDP Non- KDP Non-KDP KDP Non- KDP Non- Badegan KDP Proppo Palengaan Nita KDP Lamba KDP Sampung Maumere Leda Kota Kumba 1.2 Kecamatan Selection: Phases 3A and 3B As noted above, we extended the qualitative research into an additional two kecamatan in each of our research districts for the third phase of research. We selected these kecamatan using the following criteria: (1) The kecamatan must have had KDP for at least 3 years. As before, we need a period at least this long if we expect to find any possible "spillover effects" from the program. The kecamatan may have had KDP1 for the full three years, or may have had the last two years of KDP1 and the first year of KDP2. In some cases, the kecamatan have had KDP for four years in total. These are the "reward" kecamatan that were given KDP2 for the first year, even though they had already had three years of KDP1, because of good program performance. Priority was given to KDP2 kecamatan for practical and logistical reasons: since the program is currently running in these locations, recall issues are less of a problem and KDP staff are also on-site, and thus easier to speak with. (2) The kecamatan must pass a minimum threshold of KDP performance acceptability. If we expect to find any "spillovers," a minimum requirement is that the program is working 35The propensity score is a statistical measure designed to calculate the probability of a given household or village being selected for inclusion in a program (on this, see Rosenbaum and Rubin,1985, or for an introduction Baker, 2000). 36PODES stands for Pontesi Desa (Village Potential), a key informant survey (interviews with the village head), administered every two years by BPS that contains information on each of the 60,000 plus villages in Indonesia. SUSENAS is the National Expenditure Survey of Households, administered by the Government of Indonesia's Bureau of Statistics. 39 basically as it is meant to. We can hardly expect to find any KDP effect where the FK has run off with all the money. Thus, we defined basic performance requirements relating to transparency and accountability.37 (3) In districts where our previous KDP kecamatan (Phase 2B) was not functioning properly (Manggarai and Pamekasan), one of the new kecamatan we picked for this phase `matches' our control site from Phase 2A. This site acts as a new treatment site. We selected the other kecamatan in Manggarai and Pamekasan for variation ­ in terms of the cultural, geographical and demographic situation ­ with the other KDP kecamatan we are studying. In Ponorogo and Sikka, we picked the two new kecamatan to be as different to our previous KDP kecamatan as possible. The extra kecamatan studied in Phase 3 of the research were as follows: East Java Ponorogo Pamekasan Slahung Pasaen Jenangan Pademawu NTT Sikka Manggarai Talibura Cibal Paga Ruteng38 2. Village and Primary Case Selection (Phases 2A and 2B) The village level research took place in Phases 2A, 2B, 3A and 3B of the qualitative research. This section gives details of the selection strategy for villages studied in phase 2. Section D4 provides details on the selection strategy for villages in Phase 3. 2.1 Selecting Villages For the two periods of Phase 2 research, locations were determined by conflict case, rather than by village. Concretely, this meant that we looked for interesting cases (see selection criteria below) and then worked out which villages we should go to, based on the location of these primary cases.39 The locations of major social tensions were mapped in the first phase; preliminary data--primarily spatial--on social tensions and conflict cases helped us to pick cases and thus villages. Importantly, not all villages in a selected `KDP sub-district' may actually have participated in KDP, so field staff had to consult with project staff to clarify which villages did in fact participate while they were in the research location. We used a number of criteria to select the primary cases we would follow: First, we selected cases where the same social tension was identified in both sub-districts, but where in one case it became violent and in the other it did not. By examining similar cases (within both kecamatan) we can get a better idea of what is present, or missing, that results in a different outcome (violence or peace). A similar strategy was followed for pairs of cases within each sub-district. 37See footnote 19. 38Originally, we chose Bola (Sikka) and Reok (Manggarai) for the new NTT research sites. However, the eruption of a volcano in Sikka and flash floods in Manggarai required us to replace our original choices with Paga (Sikka) and Ruteng (Manggarai). 39By primary cases we mean cases pre-selected using the criteria below. Non-primary or secondary cases are cases selected and followed during the second Phase of research, but not pre-selected. 40 Second, we used a qualitative version of a "difference in difference" strategy, in which we identified instances of peace in otherwise relatively high-conflict villages, and instances of high-conflict in otherwise relatively peaceful villages.40 This yielded insights on local-level mechanisms for peace and conflict that were as independent as possible of the broader institutional environment (and thereby replicated at the local level the broader selection strategy adopted at the district level). As such, we compared similar cases in KDP and non-KDP locations to evaluate any possible program impacts. In order to control for non-program effects, the cases chosen for comparison were in sites that were as similar as possible prior to the KDP project's arrival. The sub-district sampling strategy outlined above helped to ensure this. Moreover, villages selected also had to be as similar as possible, as had the actual cases themselves in terms of the dynamics of conflict. In a number of instances, one case of conflict met more than one of the conflict case selection criteria. Further, in one village more than one case of conflict could be identified to match our criteria. This meant, first, that it was possible to select three or four villages in each sub-district to meet all our conflict case selection criteria. It also meant that in all sub-districts we did not need to select five different and independent cases of conflict to match all the criteria. By double-counting some cases of conflict according to the criteria list, we were also able to follow certain conflict cases with more time, thus interviewing more informants and in greater depth. This was seen as preferable to covering a larger number of cases more superficially, for the qualitative research component. Thus in most cases, less than five primary cases were selected, and less than five villages were selected as research sites.41 The criteria used for the selection of the five cases are outlined below. 2.2 Selection of Cases One and Two: Similar cases, different outcomes (within same kecamatan) Two similar cases of conflict within the kecamatan but with different outcomes--one violent, one peaceful--were selected. The criteria for similarities across cases were as follows: · Same type of conflict (e.g., conflict over inherited land between clan groups) · Same underlying tensions (e.g., similar natural resource access problems) · Same scale of conflict (e.g., same numbers of actors involved, numbers of people injured) · Same types of actors (e.g., conflict between church and school institutions and the community) In cases where all of the criteria could not be met, cases were chosen that fulfilled as many as possible. In situations where all cases of a certain interesting type were resolved violently, we chose a case that differed in terms of intensity or levels of violence. This allowed us to compare cases to discover factors that determined the quality of conflict management. 2.3 Selection of Case Three: Similar cases in two kecamatan This case was chosen to match a case in the non-KDP kecamatan with a similar case in the KDP kecamatan. Similar cases were selected using the same criteria listed above, but comparing cases in the two different sub-districts (KDP and non-KDP) within the same district. Not only were the conflicts similar in type and scale, but the general characteristics of the villages were similar as well. 40An important caveat should be noted here. In the districts selected as having a higher capacity for conflict management, the distinction between peaceful and violent conflict areas and cases was much narrower than in the case of the lower capacity districts. `Violent' villages within a sub-district, so selected, may in fact only have witnessed one or two violent cases of conflict in recent years. `Violent' villages in the lower capacity district had, relatively speaking, many more cases of violent conflict to select from. 41Even for situations where a conflict case revolved around one village, it was sometimes necessary for the researchers to visit neighboring villages in order to increase the range of informants or perspectives or to cross check information, even if those informants were only a secondary party. In practice, each team visited three villages per sub-district (Pamekasan team visited four in Phase 2B). 41 2.4 Selection of Case Four: Peaceful resolution in a violent area Within a generally violent area, or an area where a higher number of violent conflicts were identified, a case was selected that had been resolved peacefully. If no conflict had been resolved peacefully, a case was selected that had been resolved with the least violence, comparatively. 2.5 Selection of Case Five: Violent case in a peaceful area In an area of peace, or an area where there were generally a lower number of violent conflicts, or where most conflicts were resolved peacefully, a case was selected which had not been resolved peacefully or which had led to violence at some stage in the case. 2.6 Selection of Case Six: KDP-related case In Phase 2B of the research, at least one KDP-related conflict case was chosen. This case may have been an example where KDP had ­ directly or indirectly ­ caused conflict, or where we had preliminary evidence that a KDP forum, actor, or other mechanism had been used to help resolve conflict. 42 2.7 Primary Case Studies: Phase 2A Team Village Phase 2A Cases: Non-KDP sites Backup Backup Case Village Criteria 1: Same case different outcome Gelangkulon Village head election (violent) Sampung Dispute between martial arts groups (violent) Pagerukir Village head election (nonviolent) Sampung Dispute between martial arts groups (nonviolent) Criteria 2: Case for comparison between KDP and non-KDP kecamatan. Non-KDP kecamatan case Sampung Dispute over impact of mining activities on 1. Pagerukir 1. Village head election community 2. Sampung (nonviolent) 2. Dispute between martial arts Ponorogo groups (either violent or not) Criteria 3: Case with a nonviolent outcome in a violent region Sampung Dispute between martial arts groups (nonviolent) Criteria 4: Case with a violent outcome in a nonviolent region Pagerukir Dispute between martial arts groups (violent) Criteria 1: Same case different outcome Palengaan Vigilante death of a person accused of Palengaan Dispute over land ownership Daya practicing black magic (violent) Daya (more violent) Palengaan Escape and protection of a person accused Palengaan Dispute over land ownership Laok of practicing black magic (nonviolent) Daya (less violent) Criteria 2: Case for comparison between KDP and non-KDP kecamatan. Non-KDP kecamatan Case Banyupelle 1. Dispute over development of abattoir Palengaan 3. Sickle duel over suspicion of 2. Mass sickle fight at the market (violent) Laok an affair (nonviolent) Pamekasan Criteria 3: Case with a nonviolent outcome in a violent region PalengaanDaya Dispute over land ownership Criteria 4: Case with a violent outcome in a nonviolent region PalengaanLaok Mob killing of a thief Criteria 1: Same case different outcome Watu Gong Nonviolent case over development assistance Nele Wutung Violent case over development assistance Criteria 2: Case for comparison between KDP and non-KDP kecamatan. Non-KDP kecamatan Case Nele Wutung Domestic violence case Magepanda Sikka Criteria 3: Case with a nonviolent outcome in a violent region Koting A Relatively nonviolent land boundary case Criteria 4: Case with a violent outcome in a nonviolent region Watu Gong Relatively violent case over development assistance Criteria 1: Same case different outcome Golo Meni Violent land conflict case Tanah Rata Nonviolent land conflict case Criteria 2: Case for comparison between KDP and non-KDP kecamatan. Non-KDP kecamatan case Golo Meni Land case involving clan groups to compare to Golo Mangu clan land conflict case (KDP) Criteria 3: Case with a nonviolent outcome in a violent region Manggarai Tanah Rata Nonviolent land case in a violent area Criteria 4: Case with a violent outcome in a nonviolent region Watu Nggene Violent case against Jamal (widows) in nonviolent area 43 2.8 Primary Case Studies: Phase 2B Phase 2B Cases: KDP sites Backup Backup Case Team Village Village Criteria 1: Same case different outcome Dayakan Village head election (nonviolent) Biting Building dam dispute Biting Village head elections (high social tensions Biting Installation of electricity dispute and demonstrations) Criteria 2: Case for comparison between KDP and non-KDP kecamatan. KDP kecamatan case 1. Dayakan Village head elections in either of these Biting 1. Dispute over building dam to 2. Biting villages to compare with village head elections compare with mining dispute in Pagerukir/Gelangkulon 2. Dispute over electricity installation (compare Sampung) Criteria 3: Case with a nonviolent outcome in a violent region Ponorogo Biting Dispute between martial arts groups (nonviolent) Criteria 4: Case with a violent outcome in a nonviolent region Dayakan Burning down the forest Criteria 5: Case of KDP which triggers conflict Biting Building the market in KDP without clear land status Criteria 1: Same case different outcome Proppo Revenge in the rice fields (very violent) Proppo Rice fields revenge (very violent) Proppo Revenge in front of the health center (less Proppo Revenge in front of the health violent) center (less violent) Criteria 2: Case for comparison between KDP and non-KDP kecamatan. KDP kecamatan case 1. Proppo 1. Revenge Seeking (one of above) to 1. Tattangoh/ 2. Development disputes 2. Panaguan compare with revenge seeking in Banyupele Panaguan 3. Offended pride over women 2. Land dispute to compare with Pal. Daya 2. Proppo Criteria 3: Case with a nonviolent outcome in a violent region Pamekasan Panaguan Dispute over land ownership Panaguan Dispute over land ownership Criteria 4: Case with a violent outcome in a nonviolent region Proppo Revenge seeking in front of the health center Criteria 5: Case of KDP which triggers conflict Tattangoh Protest over KDP implementation (blocking Panaguan Village head election and access to site) capturing KDP for water pipes Criteria 1: Same case different outcome Bloro Successfully resolved domestic violence Magepanda Unsuccessfully resolved domestic violence Criteria 2: Case for comparison between KDP and non-KDP kecamatan. KDP kecamatan case Magepanda Domestic violence case to compare to Magepanda Sikka Criteria 3: Case with a nonviolent outcome in a violent region Magepanda KDP case with peaceful outcome Criteria 4: Case with a violent outcome in a nonviolent region Nita Drunken man conflict case, violent outcome Criteria 5: Case of KDP which triggers conflict Magepanda KDP water supply conflict case Criteria 1: Same case different outcome Golo Mangu Violent land case Satar Punda Nonviolent land case between sub-village Criteria 2: Case for comparison between KDP and non-KDP kecamatan. KDP kecamatan case Golo Mangu Land case involving two clans to compare with Golo Meni case Criteria 3: Case with a nonviolent outcome in a violent region Manggarai Satar Punda Nonviolent land case in violent area Criteria 4: Case with a violent outcome in a nonviolent region Golo Mangu Violent land case in nonviolent area Criteria 5: Case of KDP which triggers conflict Satar Punda FK stole 6 million Rp. from KDP 44 3. `Non-Primary' Cases (Qualitative Work) Thus, around five primary cases were followed through in the first sub-district in each district, with the exact number of primary cases varying by district, depending on the extent to which a particular case met more than one of the criteria listed above. As noted above, in some cases the pre-selected primary cases failed to meet the criteria by which they were selected. Thus some cases were followed that did not meet the selection criteria. In addition, researchers were also free to pursue cases beyond those pre-selected as they uncovered them in the village sites. Sometimes, these cases were in fact the more interesting ones as they had not been revealed earlier, but required the trust of villagers towards the researchers before they could be uncovered. We termed these cases `non-primary', or secondary, cases. These cases were followed up in detail if they were in the same village location as the primary case being studied, and if time allowed. In cases where these secondary cases were in a different village location, consultation with the provincial field supervisor would take place to establish whether it was worth changing locations. Overall, the four district teams produced fifty-six detailed conflict case studies. Each conflict case study contains rich ethnographic data on the context of the village, the actors, victims and leaders involved, the history of the conflict in the area, and other deep-level institutional, political and social background material, in order to help understand the evolution of conflict in each village site. 4. Village Selection (Phases 3A and 3B) Within each of the new kecamatan, we selected two villages using the following criteria: (1) Villages must have received KDP funding for at least two years. We used this criteria not because we believe that KDP effects will only be apparent if a village receives money ­ in theory, it is participation rather than economic gain that should matter. However, villages have at least three months more of KDP `process' if their proposals get funded. This is because only villages which get funding take part in the UDKP3 meetings at the kecamatan level (where final allocations are decided) and in the implementation stage. Both of these are key stages in the KDP cycle. (2) One village had to be `rural' (that is, far from the urban center of the kecamatan), and one should be `urban' (near to the center). This allowed us to test for the FK effect ­ almost inevitably, villages far from urban centers have less visits from FK. (3) Each village should have experienced some conflict. This is not a primary criteria because all our experience so far has shown that there will be interesting conflicts wherever we go. Nevertheless, if we hear of any particularly interesting conflict cases, we may select the villages there. Unlike selecting kecamatan, we were unable to select villages before the research began. The primary reason for this was that it is difficult to get accurate information on disbursements to villages, outside the kecamatan. As such, when the researchers spent the initial period in the kecamatan capital they also collected information on where KDP projects had been funded. They used this information, plus knowledge of the geography of the kecamatan (which villages are urban, which rural) and conflicts within the kecamatan, to make village selections. 45 E. Sampling of Respondents 1. Qualitative Research (Phase 1) The primary aim of the first phase of research, which took place over four weeks at the district and sub- district levels, was to map out the various types of conflict that exist in our research sites and their spatial distribution. Respondents were picked for this purpose. The `starred' informants are those that the researchers were required to speak to; `non-starred' informants were suggestions for others who may have been useful.42 Figure E1: Respondent List (Qualitative Research Phase 1) Government Security/Justice Non-government KDP Kabupaten (district) Bupati * Kapolres * Religious organizations KMKab * Kepala Dinas Social * Kodim * Academics KonsultunManagmen Kepala Dinas Kesehatan* Judges (Hakim) NGOs (LSMs) PMD * BPS * Advocates(Jaksir) Confederations Bappeda * BPN * Youth organizations Tim Koordinasi * PKK Other ethnic or cultural District Hospital * organizations Kecamatan (sub-district) Camat * Kapolsek * Religious organizations FK * BPS (If there is one) Koramil * NGO branches UPK * Land office Adat networks PjOK * Health office Youth organizations PKK Other ethnic or cultural organizations 2. Qualitative Research (Phases 2A and 2B) Whereas there was a bias in the first phase of the qualitative research towards speaking with authority figures and `experts', we consciously corrected for this in the second phase. Since most research on conflict tends to privilege elite views, we took steps to minimize this bias. Even if authorities have the best overview of the situation, they are often less likely to give information on concrete conflict incidents as they play out at the local level. Talking with ordinary villagers ­ whether they are directly involved in the conflict or not ­ helps to build specific cases and also gives important information on how events affect ordinary people. In order to fully understand a situation of conflict it is necessary to gather a full range of perspectives, including the perspectives of those who may not be `experts' or have a broad overview. There were two main categories of respondent selected for Phases 2A and 2B: · A cross-section of the village population (authority and non-authority figures) · `Experts' on a particular conflict case--those involved and observers In addition, researchers conducted a number of focus group discussions with particular population groups. For Phase 2B of the qualitative research, a third category of respondents was added: actors in KDP1. 2.1 Cross-Section of Population The table below outlines the main range of informants the researchers aimed to interview. Again, an asterisk (*) indicates required interviews. Researchers were to conduct interviews with a cross-section of 42See the Glossary for explanations of the different actors and institutions listed. 46 the population, balancing the number of authority and non-authority figures, and ensuring that all the major groups were covered. Specifically, the number of interviews conducted with respondents from the second two groups (ordinary villagers and village groups) had to be at least equal to the number of interviews conducted with respondents from the first two (authority figures and other groups). Figure E2: Respondent List (Qualitative Research Phases 2A and 2B) 1. Authority figures 2. Other figures 3. Ordinary villagers 4. Village groups Village government Healthcare Women Economic groups Kepala Desa Village midwife Elite/educated women* e.g. savings and loans Kepala BPD Traditional healer (e.g. wife of village head, groups Members of BPD owners of kiosk) LKMD/LMD Poor women* Kepala RT/RW Female headed-households Religious/adat leaders Education Men Production groups Kyai Teacher(s) Land owners e.g. weaving collectives, Priests Headmaster Non-land owners farmers' groups for Adat leaders Parent-teacher association Unemployed men collective use of machinery, groups who work on same Village Elders members Elders plantation Security Recreational groups, e.g. Members of village security related to education, teams culture, sports Religious groups e.g. prayer groups Political groups, e.g. mass organizations, political parties Figure E3: Additional KDP Respondent List (Qualitative Research Phases 2B) 1. Authority figures 2. Other figures 3. Ordinary villagers 4. Village groups KDP (Kec. Level) KDP (desa level) PjOK FD (desa facilitator) UPK TPK (activities management team) Tim Pelaksanah Kegiatan TTD (technical assistance team) Tenaga Teknis Desa Tim 2,3,5,6... (monitoring team) 2.2 Locating Additional Respondents Involved in a Conflict Case In addition to ensuring that the major population groups were covered (in a balanced way), the researchers used a snowball sampling approach to find interesting respondents. Particular people may have been involved in a conflict--as actors, victims, conflict managers, or observers--and some people will know more about a particular case of conflict than others. With snowball sampling, the researchers 47 asked respondents to suggest other involved parties to interview, thereby gradually expanding the group of informants. In addition, the researchers tried to build up a picture of the individuals and institutions to which villagers refer cases of conflict, and sought out members of these institutions for further interviews. Snowball sampling also allowed the researchers to track those who took part in any efforts at conflict management or resolution (whether the management was successful or not). In some cases, this involved more interviewing at the sub-district level. In cases where conflicts extended beyond the boundaries of the village, researchers spoke to actors or victims from those locations as well. 2.3 Focus Group Discussions (FGDs) Researchers also conducted FGDs with particular population groups: poor women; educated/elite women; young men; and poor men (see Section G for more details). 3. Key Informant Survey The key informant survey was administered to three persons at the kecamatan level (Figure E4) and eight at the village level in each of the kecamatan and villages, new and old (Figure E5). The informants are given in Figure E4. Figure E4: Kecamatan-level key informant survey respondents KDP Non-KDP FK Sub-district Head Head of UPK Figure E5: Village-level key informant survey respondents KDP Non-KDP FD (male) Village Head FD (female) BPD Head Member of borrowers Toko Masyarakat group (female) (community elder) Member of borrowers Midwife group (male) F. Qualitative Research Instrument #1: In-depth Interviews The qualitative research uses three major research tools: in-depth interviews, focus group discussions (FGDs), and informal interviews/participant observation. The former was used in Phase 1, Phase 2 and Phase 3 of the research. The latter two were used only in Phases 2A and 2B. This section gives an introduction to the first--in-depth interviews. Sections G and H give information on focus group discussions and informal interviews/participant observation, respectively. 1. Overview of In-Depth Interviews Our primary research tool for the qualitative research was the in-depth interview. The majority of researcher time was spent conducting (and writing up) in-depth interviews; the information collected from in-depth interviews made up the greater part of the data we will analyze. In all, we collected almost 800 in-depth interviews. 48 Interviews can be structured, semi-structured, or unstructured. In this research project we used an interview method that lay somewhere between the latter two: researchers were given a range of topics to cover (See Annex A1), but were free to choose which to address in a particular interview; and researchers were given a list of guiding questions (see Annex A2), but were free to adapt them in order to get at the information needed. The advantage of this `open' approach is that it did not put a constraining structure on what the researchers could talk about, and how they should do it. This allowed the researchers to conduct their interviews in the ways they thought were most suitable to the local conditions. The potential downside of this approach is that it is easy for the interview to become unfocused or pushed off track, and it can be hard to standardize results. The guidelines below, and the standardized recording formats, aimed to address these potential problems. 2. Objectives and Target Groups In-depth interviews were used in this study for three main purposes: first, they were used to gather background information on the factors (structural/economic, social/cultural, and institutional) that might affect levels of conflict and conflict management capacity; second, interviews were used to follow through specific cases of conflict and conflict management by looking at outcomes, causes and the processes that link the two; and, third, in-depth interviews were used to map the range and types of conflict that exist in the research areas, and the impacts of these conflicts. Interviews were conducted with a cross-section of the population (see respondent lists in Section E). In addition, snowball sampling was used to conduct in-depth interviews with others who had knowledge about specific incidents of conflict and/or conflict management. Researchers were required to speak to as many `ordinary villagers' as authority figures. 3. Methods and Procedures 3.1 Qualitative Research Phase 1 For the first phase of qualitative research, researchers were given a range of formats for data recording: a standard interview format, with six sub-headers under which to record relevant information on conflict in the districts and sub-district; a case study format to record data, under another six sub-headers, on particular conflict cases; a standard data format to record basic demographic and other background data about the interviewee; and a health officials data sheet, to record material relating to levels and types of injuries from conflicts. (These data formats can be found in Annex C). The different data formats helped to guide the researchers during interviews. While researchers did not need to ask the specific questions on the data sheets, they needed to ask questions that would generate information that would answer the questions on the sheet. Time was spent at the first two week training workshop on interview techniques. Researchers were given guidelines on how to approach different population groups (see Annex B), and workshop sessions were spent discussing interview methods. Researchers were encouraged not to talk about conflict directly with respondents, but rather to approach the subject indirectly and only after having built up trust with the interviewee. 3.2 Qualitative Research Phases 2 and 3 Interviews for the second and third phases of research differed in a number of ways. First, the range of respondents was much wider. Whereas most interviews in the first phase of research were conducted with authority figures, the researchers were required to interview a cross-section of the population for the second research phase. This, in some cases, required a different interviewing approach, with building trust taking on even greater importance. The second training workshop included discussion of interviewing approaches for the second phase. 49 Second, the range of topics we wanted to collect data on was far wider for the second phase of research than for the first. Whereas the main aim of research Phase 1 was to construct basic conflict maps, showing conflict types and their spatial distribution, the second phase of qualitative research aimed to understand the different factors that affect levels of conflict and local community's capacity to manage it. This required collecting data on a wide range of topics (see Annex A1 for a list of topic areas for the second phase of research). The guide given to the field researchers before Phase 2A of the research contains sample questions which the researchers could use to capture information on the different data topics (see Annex A2). While these were not intended to be used as a structured questionnaire, they did provide an outline for a guided conversation. The list of data topics is very long. Researchers were not required to cover every topic in every interview. However, over the course of all the research, the researchers were required to collect data on all topics for each village. In addition to the `general' data collected, researchers were also required to find out about specific cases of conflict. They recorded the information on specific conflict cases they got from in-depth interviews on the case study format. 4. Recording and Write-up There is no right or wrong way of taking notes. Some people prefer to take extensive notes during the interview. The advantage of this is that nothing is forgotten and that the respondent's actual words can most easily be recorded, but the disadvantage is that it adds an element of artificiality. Rather than giving the impression of the interview being a conversation, it reminds the respondent that what they say is being recorded. This may increase nervousness, or make the informant less likely to talk about sensitive issues. Others--including most anthropologists--often take no notes at all during the interview itself, leaving the note taking and write-up until later. The advantage of this approach is that the informant is most likely to be at ease; the disadvantage is that it is easy to forget key elements of the interview. The qualitative research in this study used an approach somewhere in between--some notes were to be taken during the interview, but with the understanding that they should be written up fully after the interview. Another strategy used was for one of the researchers to be the designated note taker, another the interviewer. This worked with varying success. Many social researchers tape record interviews. We did not use this technique in this study. Given the sensitive nature of conflict, we felt that people would be less likely to be open and honest if they knew not only their words but their voice were being recorded on tape. A more practical barrier to recording interviews was the fact that it takes immense amounts of time to transcribe even a short recorded interview. Thus, the researchers produced edited transcripts--almost 800 in all--rather than tape- recording their interviews. G. Qualitative Research Instrument # 2: Focus Group Discussions (FGDs) 1. Overview of FGDs A secondary research tool used was the Focus Group Discussion (FGD). Focus group discussions are often used as the qualitative or `social' component to otherwise econometric studies and evaluations of development projects. Often they are tacked-on to larger surveys, in an attempt to get at information that cannot easily be captured quantitatively. This study seeks, however, to use FGDs in a more serious fashion ­ as a key component of a broader mixed methods evaluation strategy. 50 The weaknesses of FGDs are well documented in the literature. By talking to people in group situations, those with minority opinions can be marginalized and may find it hard to dissent from the dominant view, or the view held by those with power. However, FGDs can--if the make-up of the group is well designed and if the FGD is well run--be a useful method for accessing groups that may be hard to reach using standard in-depth interviews. In total, fifty-five FGDs were conducted across the research sites, all during the second phase of the research. 2. Objectives and Target Groups The study used FGDs for three main purposes. First, they were used to gather information on specific conflict cases that the researchers were tracking from groups which might be hard to access in other settings. These groups included the `marginalized' (e.g., poor women), as well as those that one might have difficulty locating or accessing individually (e.g., victims of domestic abuse). Figure G1, below, shows the groups with which the researchers were required to conduct FGDs. Second, the FGDs were used to give more general background information on the villages being studied. FGDs often produced rich data on community life, on groups and ties that existed in the village, on economic conditions, and so on. Third, the FGDs were used to gather perspectives on conflict and, more broadly, security. Women, in particular, often have a different perspective on what the major problems and issues are in a village than do authority figures. In many cases focus group discussion helped to build trust between the researchers and segments of the community. Researchers often followed up with longer in-depth interviews with individuals who took part in a focus group discussion. Figure G1: Groups for FGDs Target Group Main Topics 1. Poor women's group 1. Women's problems 2. Problem Solving 2. Educated/elite 1. Women's problems women's group 2. Problem Solving 3. Young men's group 1. Crime and vigilantism 2. Police and security 4. Poor men's group 1. Crime and vigilantism 2. Police and security 3. Method and Procedure As noted above, the study used focus group discussions to access groups which can be harder to reach and who might not feel comfortable in an in-depth interview. With the poorest most marginalized groups, it was usually more effective to talk with them in a group of their peers. The second training workshop included two and a half days on the theory and practice of conducting FGDs. After a basic introduction to the method, researchers practiced conducting FGDs on communities in Bali. They were then required to write-up and submit their FGD notes for comment. 3.1 Focus Groups with Women As noted above, FGDs were conducted with two groups of women, covering women's problems and problem solving. Researchers used these topics as the basis of discussion, starting the FGD by talking about women's problems in general. Examples of some of the specific themes researchers used to start 51 interviews were: access to water, children's health, access to the market, access to health services and schools in the village, and marital problems. As the researchers discussed these topics, they probed further to find out where women go for help in solving these problems. This helped researchers understand women's perspectives on village government, the adat system and religious organizations, and their role in problem solving--a key aspect of conflict resolution--without asking them directly about conflict. When women were asked directly about formal problem-solving processes, they normally said that they know nothing about village government. However, they often had extensive knowledge of informal problem-solving, particularly institutions and individuals who help solve the types of problems that affect women specifically (like child health issues), which we needed to record. Researchers recorded the information they got from the FGD on the appropriate formats, using the data topics and sub-topics to code the data (see Annex A1 for data topics, and Annex D for the FGD formats). The two all-male research teams hired a local woman to help them facilitate most of the FGDs with women. 3.2 Focus Group Discussions with Men The focus group discussions with men also addressed problems in the village, but, in some cases, covered other issues such as crime and vigilantism, and the role of the police and security apparatus. In a number of cases, researchers held FGDs related to a specific conflict with groups of men. For example, in cases were two villages were disputing village and land boundaries, separate FGDs were held with groups of men from each side. 4. Recording and Write-up As with in-depth interviewing, the researchers were given detailed guidelines on recording and writing-up the data from FGDs. In many ways the recording techniques were very similar, with the main difference being that the researchers had to record who said what. Researchers were also encouraged to record the atmosphere of the FGD and the context in which it took place: who was present, what the setting was like, were people shy or nervous, or relaxed? When looking at the completed FGD transcripts, the recording of these signals often helped to explain the pattern of the FGD itself. As with in-depth interviews, any information from the FGD that related to a specific case was recorded on a separate case study format. H. Qualitative Research Instrument # 3: Informal Interviews and Participant Observation 1. Overview of Informal Interviews and Participant Observation The third research tool used in the qualitative research really consists of two different methods: the informal interview and participant observation.43 Where the other research tools collect information in the semi-formalized environment of the interview or the focus group discussion, less formal methods are also needed in order to understand conflict, and the environments that sustain or limit it. People tend to chat more freely when they do not feel like they are being interviewed, and the dynamics and relationships observed within a community often give a vital piece of the puzzle when we are trying to understand a social system. This is particularly true when we are seeking to understand something as sensitive as conflict. In addition to the formal interviews and FGDs, the researchers spent a lot of time talking informally with people in the villages where they were staying (e.g., in the evenings chatting with the host in the house 43We group informal interviews and participant observation together in this paper because they both involved immersion in the local community. The more the researchers immersed themselves in community life, the more they learned. 52 where the researchers were staying, or in conversation with the Ibu--mother--when helping her prepare the dinner). The researcher's drivers were also able to give information about the local situation. Maintaining a regular presence in the village afforded many informal occasions for the researchers to discuss life with villagers. All of these situations were opportunities to get key data. Observing was often as important as interviewing. Whereas the other research tools relied on spoken answers as a source of data, participant observation derives insight from observed actions. Here, key sources (and types) of data include the different informal relationships and structures of the research communities, and people's values, attitudes and ways of acting. Categories of `observed data' include: the physical setting and environment of the villages; activities and rituals that take place; norms and types of behaviour observed; informal interactions between villagers; and forms of non-verbal communication. 2. Objectives and Target Groups The study uses these `informal' methods for two main purposes: first, to gather key `unspoken' information from the way people act, their relationships and so on; and, second, to gather spoken information people give in informal environments. These techniques can be used at all times and with all people. Unlike the other tools, there was no formal sampling of respondents. There were no `quotas' of people the researchers needed to interview or observe. Rather anyone the researchers met, and everyone they saw, was a potential source of information. However, these techniques were especially useful at getting information from `marginalized' and `silenced' groups. In some cases, women were not confident being formally interviewed, but were happy to talk when one of the researchers was helping her in the house. Time was spent at the second training workshop discussing informal interviews and participant observation. 3. Recording and Write-up Unlike for in-depth interviews and focus group discussions, we did not use a specific format for recording the data gathered using informal interviews and participant observation. Instead a number of formats were used: 3.1 Informal Interviews For informal interviews and discussions, the format to be used depended on the amount and quality of the information gathered. If there was a lot of information the standard formats were used (data sheet, interview transcript, and, in some cases, the case data format). However, more often than not, the researchers recorded the information gained from informal interviews in their diaries. 3.2 Participant Observation Observations were also recorded in the researchers' diaries. Here, evidence and analysis could both be recorded, although the importance of making it clear which is which, and sourcing all evidence, was reiterated in the training. When the researchers were conducting in-depth interviews and FGDs, they also wrote their observations on `context'--­the atmosphere, dynamics, and non-verbal signals observed--on the appropriate cover sheet (see Annex D for formats). 53 I. Qualitative Research: Recording the Data The qualitative research involved eight to fifteen researchers spending seven months in field sites.44 The size and scope of the study presented a number of data collection challenges. First, there were four different research teams (one for each district studied), each made up of individuals with varying experience, expertise and interest. Ensuring that there was consistency of data recording was extremely important for undertaking serious cross-district comparisons. As a result, while the development of the methodology benefited from the diverse inputs of the different research teams, these inputs resulted in standardized formats that would be used in all four districts. Training was also conducted with the team as a whole to help ensure common understanding across district teams. Second, the data being collected will be, in part, analyzed by people who are not part of the field research teams.45 This leads to a number of data recording requirements which may not have been as salient if there was not such a clear role and task divide between data collection and analysis. The need for clear writing was an obvious concern, as was the necessity of keeping raw data and analysis apart (and clearly labeled). In response to the first challenge, researchers typed their formats in the time spent at the district and sub-district level (time, electricity, and hardware allowing), and (in some cases) hired local data imputers. In response to the second, the formats used in both phases of the qualitative research kept raw and analyzed data apart (see below). Third, it was important that all primary material be sourced clearly and systematically, so that those conducting the analyses could return, if necessary, to the primary source material. The guidelines on writing case studies (see Section J) set out how this should be done in the context of writing multiple perspective cases of conflict and conflict management. The formats themselves (and particularly those for the second phase of research) also indicate where and how sourcing should be done. Fourth, the research operated in a multiple language setting. Given this, there was, and is, an opportunity for lost or distortion of meaning in the data collection and analysis processes. To guard against this, data was recorded as much as possible in Indonesian, but with particularly meaningful local words recorded directly (with explanations in Bahasa Indonesia) given beside. Researchers produced glossaries of important local words. Although most of the researchers spoke good English, they were not asked to record the data in English. Rather, experienced translators were hired for this (see next Section L on managing the data). The different data recording formats are given in the Annexes. The rest of this section explains what formats were used, when, and how, for the two phases of qualitative research. The formats are divided into `raw data formats' (those used to capture primary data), and `analysis formats' (see Sections I1 and I2, respectively). The formats were developed and refined iteratively during the training workshop periods. Research Phase 1 allowed some of the basic instruments to be piloted. As a result, the instruments used in the second phase of research differ from those used in the first. 1. Raw Data Formats 44As noted earlier, fifteen researchers spent approximately five months in the field (Phases 1, 2A and 2B). For Phase 3, eight researchers spent two months conducting fieldwork. 45Although, in reality, the divide is not absolute. Two of the team who will be writing the final report (Claire Smith and Patrick Barron) spent most of the research period in the field, supervising the different research teams. During these `supervision' trips, they also conducted their own research. The other (Michael Woolcock), although based in D.C., had a number of field visits over the course of the project. 54 1.1 Phase 1 Four data recording formats were used in the first phase of the research. (The formats are reproduced in Annex C). (i) Interview Data Sheet For every interview conducted, researchers filled out a standardized `background' data sheet. This `background' sheet recorded basic information: the sex, age, position, occupation, basic wealth categorization, ethnicity, and religion of the interviewee; the date, time, location and length of the interview, as well as who else was present. The `background' sheets were coded by interview to match the standard interview format (in case they became separated). Researchers were encouraged to incorporate these questions into an opening discussion rather than asking them outright. (ii) Standard Interview Format The standard interview format was completed for every interview. The form also acted as a basic guideline for the questions to be asked during the interview, but was not to be filled in during the interview. Information on this format was recorded in the informant's words as much as possible. The forms were filled in at the end of every interview by the interviewer(s). Researchers were free to ask the questions in any order and to use their own choice of wording. (iii) Case Study Format The case study format was used to record information on specific cases of conflict and/or conflict management that came up during interviews. In some interviews, more than one case study format was used when more than one conflict case came up. Information on this format was, again, recorded in the informant's own words as much as was possible. Researchers were, also again, free to ask the questions in any order and to use their own choice of wording. As for the standard interview format, case study formats were filled out after the interview. (iv) Health Officials Data Format This was an experimental format aimed at collecting data from health care providers, at the district and sub-district levels. These officials often have the best sense of levels and types of violent conflict. The format was a complicated attempt to allow for the collection of standardized data across the research areas. The format collected `informal' statistics on levels of injuries/deaths from violent conflict. Researchers also filled out the semi-standardized questionnaire form and, in some cases, the case study form(s) after interviewing health officials. Researchers filled out the formats after the interview but used them during the interviews to guide their questions. 1.2 Phase 2 Three data formats were used to record raw data in Phase 2 (see Annex D for the formats used). These formats were used in both Phase 2A and 2B. However, the versions used in Phase 2B were slightly modified to take into account the lessons learned from the first phase of village-level research. (i) Interview Data Sheet As above. No modifications of the format were deemed necessary after Phase 1. (ii) Case Study Format The case study format was modified slightly from that used in Phase 1 of the qualitative research. Researchers had new categories to find information on (compare the formats in Annex C and Annex D). (iii) Standard Interview Format (Coded Transcripts) The first phase of research constructed `conflict maps' of the research areas. The second phase (the village-level research) aimed to investigate in far more detail processes of conflict and conflict management. As a result, the number of research topics we investigated in the second phase of research 55 was much larger and, as such, we changed the standard interview format for Phase 2. Researchers recorded general data gathered during in-depth interviews and, in some cases, informal interviews, on this format. When researchers finished their interview they wrote up their notes in their notebooks first; then they rearranged the transcript of the interview onto this format. Researchers reordered their transcripts and coded the information based on the general data topics (see Annex A). This involved coding the data--working out which of the data categories the information collected related to. The format has a left margin where the `data code' was written; the actual data went next to the appropriate code. Researchers were given substantial training in coding material at the second training workshop. (iv) Standard Focus Group Discussion Transcript Recording Format This format (new to the second phase of the qualitative research) was used to record general data gathered in focus group discussions. Case data was not recorded here. The format is very similar to the one used for interviews, with the exception that it has an extra column to write the respondent's name next to the information recorded. The same principles of coding as for the Standard Interview Transcript Recording Format were used for this format. (v) Standard FGD Data Format This format is the same as the interview data sheet, with the exception that it has room to record the details of more respondents. It was used to collect background information on respondents who took part in focus group discussions. Like the Interview Data Format, there was also room on the form to record contextual data on atmosphere, problems in the interview, dynamics, and non-verbal signals from respondents. 1.3 Phase 3 The raw data formats for the qualitative component of Phase 3 of the research were the same as those used in Phase 2 with the exception of the FGD formats that were not used in this phase. We also added the key informant questionnaire (see Annex X). 2.Analysis Formats and Pieces 2.1 Phase 1 In addition to the raw data formats, researchers recorded their analyses in two formats. First, researchers were required to keep a diary each day. The diaries contained researchers' analyses of the data they were collecting, as well as any methodological challenges they were encountering. Second, each team was also required to submit analysis pieces at the end of Phase 1 which, together, provided a basic conflict mapping for each district. 2.2 Phase 2 In the second phase of research, the researchers continued to keep diaries. These proved useful, both for the researchers and for those of us doing the analysis. In particular, many key insights into the cultures of the research settings and the institutions and social relations within them were recorded in the diaries. The main analytical pieces the researchers produced in Phase 2 of the research were conflict pathway case studies. These present the pathways of fifty-six individual cases of conflict, and involved bringing together information from different respondents as well as the researchers' analysis. Details are given below (Section J). In addition, basic village profiles were written. Researchers were also asked to write other analysis pieces including comparisons of conflicts and research sites, and of the roles different actors and institutions play, and summaries of KDP performance and its interaction with conflict. 56 2.3 Phase 3 Researchers continued to keep diaries and to write case studies (although these were a lower priority than for other phases of the research). In comparison with previous phases, the analytical pieces had greater importance in Phase 3. Each district team produced three analysis pieces after each phase (3A and 3B): · Analytical Piece #1: Problem-Solving, Dispute Resolution and Meditation This document outlined how different communities solve their problems and manage disputes. It first looked at problem solving and dispute resolution individually in each of the two villages, and then compared the two villages ­ looking for similarities and differences. In particular, it focused on why mediators were successful or not. When attempts at mediation failed, was it because of a lack of legitimacy on the part of the mediator, because of a lack of willingness of the mediator to act, or because of a lack of capacity? What were the conditions that needed to be met for a mediation to be successful? · Analytical Piece #2: Civic Interaction and Identity Groups This document explored in more detail the different identity groups that exist in the kecamatan and villages studied. It summarized the main identities that exist in the research locations, and then examined the extent to which bridging associations or interactions exist between the groups. · Analytical Piece #3: KDP and its spillover effects The final (and most important) analytical piece examined the performance of KDP in the new kecamatan, considering whether KDP helps communities manage conflict or not, and, if so, the cases in which it does and the mechanisms by which it does so. In addition, each research team completed four different types of in-field summary forms: · 3 KDP forms on the role of KDP (at kecamatan level) · 3 demographics forms (at kecamatan level) · 7 KDP forms on the role of KDP (at village level) · 7 demographics forms (at village level) J. Writing Case Studies 1. Overview of Case Study Writing At the end of Phase 2A and Phase 2B, we held two-week writing workshops for each of the provincial research teams. (Researchers wrote case studies in their home locations after Phase 3A and during a final workshop after Phase 3B.) During the in-field writing periods, the researchers brought together the information from the individual interviews, focus group discussions and participant observation they had conducted to write fully-formed conflict pathway case studies (see Section C6.3 on the writing workshops). This task involved bringing together the information from the individual case study formats that related to a specific case. However, it also involved analysis: trying to work out what was important in the case, what the dominant narrative was, and so on. At the beginning of the first writing workshop, researchers received a day of training on how to write multi-perspective case studies and were given samples of good case studies. All in all they collectively produced fifty-six case studies ­ averaging 10-20 pages each ­ over the course of the two writing workshops in Phase 2; another ten were prepared after Phase 3. The researchers also produced short village profiles in note form for the twenty-five villages studied in Phase 2. In Phases 3A and 3B, we introduced new village summary formats which the researchers used to both produce profiles of the new research areas and reformat the summaries from Phase 2. 57 2. Content of Cases The case studies were meant to reflect all the elements of a conflict pathway. This meant that each stage of a conflict, and attempts at its resolution, needed to be noted, as well as the events that linked different stages. Each case study included: summary of the case; the `pre-history' of the case; the history of the case; attempts at resolution; impacts of the case; and the aftermath of the case. The researchers were given a detailed list of elements that needed to be included in their case studies: Summary of case: small box (paragraph) at the beginning · Five sentences describing the case · Type · Scale · Outcome Pre-history of case (context) · Brief context and background of the village/cultural/economic/social issues relevant to the case History of the case · What happened and how? (the chain of events) · Conflict Management/resolution o When did it happen? o Where did it happen? o Why did it happen? · Underlying causes (wood of the fire) · Trigger event (matches that lit the fire) · Who was involved in the conflict case (perpetrators, victims)? o Who was not involved? o Who was affected (victim and wider impacts)? · Who was involved in conflict management/resolution? o Who was involved? o Who was not involved? Impacts · # Deaths · # Injuries · Amount of destruction (property/other) · Social and psychological effects (on the individual) · Changes in group relationships · Economic impacts Aftermath · Summarize the current state of conflict · Is the case finished--what does this mean? · Is the case resolved? · Will the case likely lead to further conflicts/violence? · Are people satisfied with the outcome? · If so, who? · If not, who is not, and why? · Number of similar cases (impressions, quantified if possible, but not necessary) 58 3. Sourcing and Glossaries As noted above, the importance of consistent and clear sourcing cannot be understated. While the case studies will be used when the final analysis work is being done, this may involve going back to the primary data. Researchers were given firm guidelines on sourcing. In addition, researchers wrote glossaries of local terms. These helped the translators when the cases were being translated. They will later be combined into an overall glossary. K. Data Entry and Analysis 1. Data Entry and Translation Procedures 1.1 Qualitative Research Phase 1 Researchers collected data from the first round of qualitative research, which is primarily categorical and spatial, on a range of formats (see last section and Annex C). In addition, the provincial field supervisors collated the data into master spreadsheets (one for each district). These spreadsheets are in English. The exact conflict categories were partially determined a priori, but were then refined as the early rounds of data began to come in, allowing us to establish a basic typology characterizing the types, triggers, intensity, frequency, duration, and impact of conflict in our eight selected sub-districts. We will use this typology as a basis and structure for the final analysis work, and to help determine categories for inclusion in the quantitative surveys. The field supervisors used the master spreadsheets to pinpoint interesting conflict cases to follow in the second phase of the research. This preliminary conflict case selection provided a basis for discussion at the second training workshop, where final decisions on which primary cases to follow were made. Researchers typed both their original formats and their analytical pieces in the break between the two phases of research. Phase 1 material was translated by the field supervisors and some of the researchers. 1.2 Qualitative Fieldwork Phases 2 and 3: Primary Data Processing and managing the data flow from the second round of the qualitative field research was far more complicated, both in terms of the amount of material collected, and the range of possible responses to be incorporated. Around eight hundred interview transcripts were produced in these phases, with many accompanied by case study formats--almost 10,000 pages of primary data. Matters were further complicated by the tension between wanting to keep the written narratives in Bahasa as the aggregation process proceeds (to retain as long as possible the original sense of what was being conveyed in the interviews and observations), accompanied by the need to translate this material into English at some appropriate point for the purposes of analysis, write-up, and (external) dissemination. After consultation with a handful of people at the Bank with experience managing such a large, sensitive, and diverse research project (in both DC and Jakarta), and a number of academics--anthropologists, sociologists, and political scientists at Harvard, LSE, Tufts and the University of London--we developed the following procedure. There is no escaping the large time commitment that the three people doing the final analysis will have to put in to fully familiarize themselves with the raw material--final reports cannot really be written by committee. Yet given that we have interview and focus group discussion data from upwards of a thousand people, we needed to establish a more manageable interim level at which this familiarization could happen. 59 As a result we made a decision to filter the number of transcripts that were translated and analyzed to a more manageable level: 200. Overall, there are over 1,000 pages of conflict pathway case studies. These were translated as well as a selection of transcripts. There was a need to balance carefully concerns of representativeness of data translated--to control for selection bias--with a need to ensure that the most illuminating interviews are also translated. To this end, our strategy was to (a) translate every sixth transcript (regardless of perceived `quality'); and (b) select the most interesting transcripts.46 Three full time native English speakers were hired to translate the case studies and formats. The provincial supervisors also translated some material. Selection of transcripts for translation was a task of the four provincial supervisors, with support and guidance from the Jakarta management team. In some cases, parts of rather than the whole transcript were translated. For each `information piece' the standard data sheet was attached. This ensures that the original sources can be readily traced and checked should the need emerge during the aggregation process. The two hundred information pieces, the case studies, the additional translated analytical pieces, and some extra translations by the researchers constitute the primary source material used to prepare the final report(s). 1.3 Qualitative Fieldwork Phases 2 and 3: Analyzed Data In addition, the research teams themselves--in consultation with the provincial coordinators--were given responsibility for making early forms of aggregation in their summary field reports. Researchers wrote a range of summary analytical pieces (see Section J above). These reports, though written on the basis of interviews/observations with several people (and thus representing an initial "aggregation" of views), are self-contained and coherent documents. 2. Aggregation and Analysis of Data The procedure outlined above has the broad goal of yielding about 200 information pieces that represent a manageable interim step between eight hundred individual interviews and a general summary statement of our findings. When these narratives are appropriately mapped onto our site selection design, they should yield as good a story as we can likely get with qualitative data. Preliminary coding has already been done by the researchers during the data collection period. Once in English, qualitative software analysis tools will be used to more formally manage and codify the material, though this will not be a substitute for doing the hard work of reading and familiarizing oneself with each transcript. 3. Review Plans Preliminary drafts of all the outputs from this study (see below) were, and will be, circulated among the research team and among the advisory group to solicit suggestions for improvement. Preliminary findings will also be circulated widely for comments. In addition, a series of `brownbag lunches' were held to outline ongoing findings and to receive input. 46We originally decided to translate every fifth transcript but later changed this to every sixth in order to increase the number of `interesting' transcripts that could be translated. 60 L. Final Outputs Outputs from the research will take four primary forms. 1. Village Feedback and Capacity Building Local field staff will have the opportunity to use the material from their case study narratives to contribute to regional newspaper articles, and (importantly) to provide feedback to the communities from whence the data was obtained. Too often data collection is a one-way process, with researchers `taking' data but rarely returning to share the results. Funds for disseminating findings back to village communities will be used from the `Empowerment' trust fund grant, which explicitly recognizes the importance of this process. We plan to host eight workshops at the sub-district level to disseminate (and discuss) results in May 2004. The researchers will be responsible for running these workshops. A one week workshop will be held in Jakarta for participating researchers to discuss how we might run, and document, these events. The material collected in the research will also be available in the public domain, once it has been appropriately edited to ensure confidentiality. Other researchers, academics, NGOs, the government, etc. will be free to use the data we have collected. In addition, it is planned that the complete set of transcripts and other material will be archived--probably at UCLA or Cornell--for the use of other researchers. 2. Operational Support Though KDP is primarily a participatory mechanism designed to more accurately and effectively provide appropriate development assistance to poor communities, Indonesia itself (as noted at the beginning of this paper) is rife with conflict. As such, participatory projects that may be able to play some role in helping stem that conflict are especially important. As the next iterations of KDP are implemented (KDP2 and KDP3), the results of this research will help to maximize their impact. Moreover, a new project explicitly designed to address conflict issues (SERP) is currently in preparation. The results of this research will increase our understanding of the opportunities and constraints that external agents face as they attempt to facilitate the improvement of local conflict mediation processes. A formal mechanism to input emerging data and findings from the research study into the operational project has been established. Ongoing brown bag discussions, circulation of analysis reports and back to the office reports, and so on will also help. 3. Policy Research Articles As far as we can tell, this is the first systematic and comprehensive effort to establish whether and how a CDD (community-driven development) project helps to improve local conflict mediation processes. Because of the topic, the integrated (qualitative and quantitative) data sources it has called upon, and the relatively long period of time that field staff have spent in villages collecting data, this study is being closely followed by others with interest in conflict reduction. The findings will also be of interest to a broader range of practitioners, scholars, and policymakers, with interest in a variety of areas including (but not limited to) state transition, human security, and development theory. Several policy research articles, and possibly a book, will be prepared to communicate our findings to them. 61 4. Methodological Tools Lastly, a complementary goal of this study is to develop and share innovative research methods and tools with other interested researchers and development practitioners. In principle, the framing of the study as an advanced version of "thinking quantitatively, acting qualitatively" (Woolcock, 2001) can easily be adapted to other settings. From this starting point, prospective research teams can reinterpret and scale the details of implementation to match available resources. Thus we believe the conceptual underpinning and basic evaluation principles can serve as a foundation for even the smallest, most resource-constrained study. In addition, the full set of training materials used to prepare the qualitative research teams (see Barron, Diprose and Smith, 2004) will soon be freely available online to those considering or already engaged in similar studies elsewhere. The set includes detailed field guides for all three phases of research, data recording formats, key informant questionnaires and practical implementation tips. However, successful replication of this study will require fundamental adaptation of the research hypotheses, questions and field tools to new environments. This endeavor has evolved over several years and benefited from a unique alignment of resources, talents, and commitments (both individual and institutional). We cannot overemphasize the importance of serious, extended consideration of the local context, and a deep commitment to training and engaging all levels of the research team. Substantial advanced fieldwork was crucial for generating specific hypotheses to test and designing research instruments that were relevant to the Indonesian context. The quality of the data collected, and hence of all subsequent analysis, depended on the comprehensive training, mentoring and buy-in of field researchers. Commitment to the study and to ensuring its quality involved socialization and group learning for all levels of the research team. For any research environment, the adaptive decision-making process--including making time and space for ideas to gestate--is a critical prerequisite to crafting a strategy that more comprehensively uncovers the `truth' regarding the key questions under investigation. 62 M. References Alkire, T., A. Bebbington, T. Esmail, E. Ostrom, M. Polski, A. Ryan, J. Van Domelen, W. Wakeman, and P. Dongier (2001). PRSP Sourcebook. Available online at: www.worldbank.org/poverty/strategies/sourcons.htm Baker, Judy (2000). Evaluating the Impact of Development Projects on Poverty. Washington D.C.: World Bank. Bates, Robert (2000). Violence and Prosperity: The Political Economy of Development. New York: Norton. Barron, Patrick, Rachel Diprose, and Claire Q. Smith (2004). `Training Manual for the KDP & Community Conflict Negotiation Study'. Mimeo. Washington, D.C.: The World Bank. (Will soon be freely available online). Barron, Patrick, and David Madden (2003a). `Violence and Conflict Resolution in Non-Conflict Regions: The Case of Lampung, Indonesia'. Washington D.C.: World Bank East Asia & Pacific Region. Barron, Patrick and David Madden (2003b). `Beyond Structural Factors: Towards a Model for Understanding Conflict and Conflict Resolution in Indonesia'. World Bank Office Jakarta Social Development Unit working paper. Mimeo. Barron, Patrick, Kai Kaiser, and Menno Pradhan (2004). `Local Conflict in Indonesia: Measuring Incidence and Identifying Patterns'. World Bank Office Jakarta. Mimeo. Bennett, Andrew and Alexander George (1997). `Process Tracing in Case Study Research'. Paper presented to MacArthur Foundation Workshop on Case Study Methods. Available at: http://web.mit.edu/17.878/www/Bennett/Process%20Tracing%20in%20CS%20Research.htm Esman, Milton and Ronald J. Herring (eds.) (2001). Carrots, Sticks, and Ethnic Conflict: Rethinking Development Assistance. Ann Arbor: The University of Michigan Press. Fajnzylber, Pablo, Daniel Lederman, and Norman Loayza (2000). `What Causes Violent Crime'. Paper presented at conference on Asset Distribution, Poverty and Economic Growth'. World Bank conference, July 14th ­ 17th, 1998. Brazil. Friend, Theodore (2003). Indonesian Destinies. Cambridge, MA: Harvard University Press. Government of Indonesia (2002). `Kecamatan Development Program: KDP in Conflict Areas'. Ministry of Home Affairs, Community Development Agency (PMD) and National Management Consultants (NMC). Grimes C., T. Therik, B.D. Grimes, and M. Jacob (1997). A Guide to the People and Languages of Nusa Tengarra. Kupang, NTT: Center for Regional Studies, UKAW. Horowitz, Donald (1985). Ethnic Groups in Conflict. University of California Press. Kanbur, Ravi (ed.) (2002). Q-Squared: Qualitative and Quantitative Methods of Poverty Appraisal. New Delhi: Permanent Black Publishers. Rao, Vijayenda and Michael Woolcock (2003). `Integrating Qualitative and Quantitative Approaches in Program Evaluation', in Francois J. Bourguignon and Luiz Pereira da Silva (eds.) Evaluating the Poverty and Distributional Impact of Economic Policies. Washington, D.C.: The World Bank. Rosenbaum, Paul R. and Donald B. Rubin (1985). `Constructing a control group using multivariate matched sampling methods that incorporate the propensity score'. The American Statistician 39: 33- 38. Smith, Claire Q. (2004). `Ethnic Conflict in Indonesia: Violence and Reconciliation in Central Kalimantan, East Java and Lampung'. Washington D.C.: World Bank East Asia & Pacific Region. United Nations (2003). United Nations Human Development Report 2003: Deepening Democracy in a Fragmented World. New York, NY: UNDP. Varshney, Ashutosh (2002). Ethnic Conflict and Civic Life: Hindus and Muslims in India. New Haven: Yale University Press. 63 Annex A: Research Topics, Data Codes and Field Guides A1: General Data Topics and Codes 1. Socio-Economic Factors a. Inequality b. Unemployment c. Socio-Economic Mobility and Illegal Activities d. Occupations e. Education 2. Ethnic Composition a. Ethnic Composition b. Main Group Identities c. Construction of Group Identities 3. Formal Governance a. Trust in, and Problem Solving by, Village Government b. Trust in, and Problem Solving by, Kecamatan Government c. Participation in Village Government Decision-Making d. Participation in Kecamatan Government Decision-Making e. Knowledge of Government Decisions and Processes and Access to Information f. Predictability of Decision-Making g. Accountability and Representation h. New Institutions 4. Adat a. Role of Adat b. Trust in, and Problem Solving by, Adat/Traditional Leaders c. Accountability and Representativeness of Adat d. Participation in Adat Decision-Making e. Adat Practices 5. Religious Organizations a. Role of Religious Organizations b. Trust in, and Problem Solving by, Religious Leaders c. Inter-group Participation d. Tensions 6. Civic interaction a. Associational Interactions b. Social Interactions 7. Law enforcement/Justice system a. Police b. Positive Legal System c. Tensions between Positive/Adat/Religious Law 8. Military 9. Crime/Vigilantism 64 10. Population flows a. Displacement/Migration b. Female-Headed Households 11. Other Conflict Resolution Mechanisms47 a. Functional Groups b. Village Security Groups c. Illegal Groups d. Other Groups 12. Types of Conflict a. Characteristics b. Causes c. Consequences d. Actors and Location 13. Socio-cultural factors 14. Development Projects (non-KDP)48 a. Presence and Knowledge of Development Projects b. Participation in Development Projects c. Tensions with Development Projects d. Outcomes and Impacts of Problems with Development Projects 15. KDP Process and Mechanisms a. Participation in KDP b. Competitive Bidding Process in KDP c. Proposal Writing Process in KDP d. Facilitation of KDP e. Tensions in KDP Process f. Complaints Mechanisms 16. KDP and Conflict/Conflict Resolution a. Where KDP Facilitators Addressed non-KDP Related Conflict Issues b. Where KDP Forums Were Used to Address Non-KDP Related Conflict Issues c. Where KDP Has Had a Positive Effect on Social Relations Between Groups d. Where KDP Has Led to Non-KDP Related Conflicts A2: General Data Field Question Guide 1. Economic Factors (hypotheses 1.1a and 1.1b) a. Inequality (1.1a) · What proportion of households are poor in this desa/dusun? · Who is rich in this desa/dusun? Who is poor? (differentiate by occupation, age, ethnicity) · Has this situation changed over the last four years? · Why are some people rich and others poor? 47By `other', we mean mechanisms that are not included in other sections--i.e., formal government (section 3), adat (section 4), religious organizations (section 5), police (7a), courts (7b), and the military (8). 48Research topics 14-16 were only investigated in Phase 2B in the KDP sub-district. 65 b. Unemployment (1.1b) · Is unemployment a problem in this desa/dusun? What proportion of people are unemployed? · Of those who are unemployed, what were their previous jobs? What gender are they? What age are they? · Has unemployment gone up or down in the last four years? 49 c. Socio-Economic Mobility and Illegal Activities (1.1b) · What are the prospects for the poorest households in this desa/dusun to improve their economic position? Has this changed over the last four years? · What are the prospects for the poorest households in this desa/dusun to access education? Has this changed over the last four years? · If the poorest households feel they are unable to improve their economic or education position, why? If the poorest households feel able to improve their economic/educational position, why? · To what extent do people in this desa/dusun gain income from illegal activities? What illegal activities? · Why do people engage in illegal activities to make money in this desa/dusun? d. Occupations · What are the main occupations in the desa / dusun? · Has this changed over the last four years? e. Education · How many children go to / finish school? · What level do education do most people get? · Has this changed? 2. Ethnic Composition (hypothesis 1.1c) a. Ethnic Composition (1.1c) · What is the ethnic make-up of this desa/dusun? · How has this changed over the past four years/in recent years?50 b. Main Group Identities (1.1c) · What groups do people define themselves as part of? · Which are the most important group identities? (e.g., Clan, ethnic group, religion, silat (martial arts group), class/employment type, origin). · To what extent do these group identities overlap? · To what extent have these group identities changed over the last four years/recent years? c. Construction of Group Identities (1.1c) · What is the size of each group? · Who can be a member of each group? · To what extent do groups face disputes with groups of similar types (e.g., martial arts groups/religious groups/clashing house-based culture identities)? · How serious are these disputes? Do these disputes ever become violent and why? 3. Formal Governance (hypotheses 1.3a, 1.3b, 1.3d, 2a, 2b and 2c) a. Trust in, and Problem Solving by, Village Government (1.3a) · Which village government institutions do people trust/accept the authority of most? · What is the role of the Kepala Desa/RT/RW/BPD/LKMD or LMD/PKK in solving problems? 49Four years is our key measurement because it is during this time that KDP has been in existent in these districts. 50When we say `in recent years' we just want some idea of longer-term trends. Interviewers should note the trend and the rough time period over which it has taken place. 66 · To which village government institutions do people turn to help them solve their problems? Which kinds of problems? · Do people feel the village government is more able/less able to solve problems than four years ago? Why? b. Trust in, and Problem Solving by, Kecamatan Government (1.3a) · Which government institutions do people trust/accept the authority of most? · What is the role of the camat/camat's office in solving problems? · To which kecamatan government institutions do people turn to help them solve their problems (if any)? Which kinds of problems? · Do people feel the kecamatan government is more able/less able to solve problems than four years ago? Why? c. Participation in Village Government Decision-Making (1.3b, 1.3d, 2a, 2b and 2c) · How does village government make decisions? Who participates? · To what extent do women participate in village government decisions? · Has the way village government institutions makes decisions changed in the last four years? · Who do decision-makers consult in making decisions? · Has this changed over the past four years? · How many women are in leadership positions in village government? d. Participation in Kecamatan Government Decision-Making (1.3b, 1.3d, 2a. 2b and 2c) · How does the kecamatan office make decisions? Who participates? · To what extent do women participate in kecamatan government decisions? · Has the way the kecamatan office makes decisions changed in the last four years? · Who do decision-makers consult in making decisions? · Has this changed over the past four years? · How many women are in leadership positions in kecamatan government? e. Knowledge of Government Decisions and Processes and Access to Information (1.3b) · How do you find out about government decisions? · Do you know when public meetings are held? Who tells you? f. Predictability of Decision-Making (1.3b) · To what extent do village government institutions compete or cooperate with each other? · If they are in competition with each other, does this have an effect on their ability to help solve problems/resolve disputes? · To what extent do kecamatan government institutions compete or cooperate with each other? · If they are in competition with each other, does this have an effect on their ability to help solve problems/resolve disputes? g. Accountability and Representation (1.3b, 2a, 2b and 2c) · Do the village government institutions represent all people or certain groups? If the latter, which groups? · Do the government institutions represent all people or certain groups? If the latter, which groups? · How do the leaders of village government get appointed/elected? Has this changed in the past four years? · How do the leaders of kecamatan government get appointed/elected? Has this changed in the past four years? h. New Institutions (1.3a and 1.3b) · Are there any new government institutions in the village? · Are there any new government institutions in the kecamatan? · Have these new institutions had a positive/negative effect? Why and in what ways? · In what ways are they different? 67 4. Adat (hypotheses 1.3a, 1.3c, 1.3d, 2d) a. Role of Adat (1.3c, 2d) · What type of adat or customary systems exist and at what levels? · To what extent are these institutions involved in legal decision-making? · To what extent are these institutions involved in government decision-making? b. Trust in, and Problem Solving by, Adat/Traditional Leaders (1.3a, 2d) · Who accepts adat authority, to what extent and why? · Who rejects adat authority, to what extent and why? · Has the composition of the group of people covered by adat changed over the past four years/in recent years? · Do those who accept adat authority trust adat law/structures/decisions? · What role do these institutions/leaders play in dispute resolution at each level? c. Accountability and Representativeness of Adat (1.3c, 2d) · Do the adat institutions involve one clan/ethnic group only? · Do they ever involve other clans/ethnic groups? How has this changed over time? · How do adat leaders get appointed/elected? Has this changed in the past four years? · Are there ever any disputes/problems over the appointments/elections? d. Participation in Adat Decision-Making (1.3c, 1.3d, 2d) · How do the adat institutions make decisions? Who participates? · Has the way the adat institutions makes decisions changed in the last four years? · Who do decision-makers consult in making decisions? · Has this changed over the past four years? · To what extent do women participate in adat decision-making? e. Adat Practices · What particular Adat practices are there? · When do these take place? · Who is involved? · What is the significance? 5. Religious Organizations (hypotheses 1.3a, 1.3c, 1.3d, 2a, 2b and 2c) a. Role of Religious Organizations (1.3c) · What type of religious organizations exist and at what levels? · In what areas of life do religious organizations play a role? At what level? In what issues do they get involved? b. Trust in, and Problem Solving by, Religious Leaders (1.3a) · What is the extent of the authority of different religious organizations (informal/formal)? · Who accepts religious authority and to what extent and why? · Who rejects religious authority and to what extent and why? · What role do religious leaders play in resolving disputes? · How successful are they? c. Inter-group Participation (1.3c, 1.3d, 2a. 2b and 2c) · Do religious organizations incorporate different groups (of other sorts ­ ethnic/clan etc)? · What role do these institutions play in dispute resolution at each level? · To what extent do women occupy leadership roles in religious organizations? d. Tensions (1.3a and 1.3c) · Are there any tensions between different religious organizations? · How pervasive are these tensions? 68 6. Civic Interaction (hypotheses 2a, 2b, 2c and 2d) a. Associational Interactions (2a, 2b, 2c and 2d) · To what extent are the memberships of the following functional groups made up of people of different identities? (repeat for each of the different identities that people identify): o Credit/Finance groups (e.g. savings and loans groups) o Production groups (related to availability of material inputs and/or creation of outputs--weaving collectives, farmers' groups for collective use of machinery, groups who work on same plantation, etc.) o Recreational groups (related to education, arts, culture, and sports) o Religious groups (e.g. prayer groups, churches) o Political groups (political parties, confederations) · For the above mentioned groups, has membership composition changed by identity groupings in the past four years/in recent times? b. Social Interactions(2a, 2b, 2c and 2d) · When you have a party/celebration do you tend to invite people from other ethnic groups (repeat for each identity factor mentioned)? If not, why not? Has this changed over the last four years/in recent times? · What parties/celebrations do you get invited to? How often do people from other ethnic groups invite you (repeat for each identity mentioned)? Has this changed over the last four years/in recent times? · Do your children play with children from other ethnic groups (repeat for each identity mentioned)? Has this changed over the last four years/in recent times? · If a neighbor had a problem would you try to help? Would it make a difference if they were from a different ethnic group (repeat for each identity)? Has this changed over the last four years/in recent times? · Do you use kiosks/market stalls run by people from other ethnic groups (repeat for each identity)? Has this changed over the last four years/in recent times? 7. Law Enforcement/Justice System (hypotheses 1.2a, 1.2b, 1.3a and 1.3b) a. Police (1.2a and 1.2b) · How many police are there in this kecamatan/kabupaten? · How often do they come to your dusun/village? · What kinds of disputes do police get involved in? · At what point in the dispute/conflict do police get involved? · Does police involvement resolve or end disputes/conflict? If so, when and why? If not, why not? · Do people trust police involvement and why? · Who tends to appeal to the police for help? · Who does not trust police involvement and why? · Do police arrest criminals? Do they release criminals according to the law? b. Positive Legal System (1.2a and 1.2b) · Which types of disputes go to the district court? · At what stage of disputes do they go to the district court? · Do people trust the jaksir/hakim in the district courts? · Do people accept the decisions made in the district courts and why? · Do cases ever get referred to higher courts and at what levels and why? · Do court decisions tend to end disputes/conflicts? If so, why? If not, why not? c. Tensions between Positive/Adat/Religious Law (1.2a, 1.2b, 1.3a and 1.3b) · Are there ever any conflicts between positive and religious/adat law? 69 · How are conflicts between positive and religious/adat law resolved? · Do people feel more satisfied with decisions (solutions to disputes) made by the positive legal system/adat system/ religious system? · Which system do people prefer and why? 8. Military (hypotheses 1.2a and 1.2b) · To what extent does the military get involved in resolving village level disputes? · Which level of the military gets involved and in which ways? · For what types of dispute? · How successful is the military at intervening in and resolving local level conflicts? · When the military gets involved, do you think their involvement is useful? · Has military involvement in dispute resolution changed over the last four years/in recent years? · Have there been any tensions between the military and police? How have these played out? 9. Crime/Vigilantism (general) · To what extent is crime a problem in this desa/dusun? · What kinds of crime are a problem? · Who tend to be the main actors, who are the main victims? · Why do people engage in crime? · Have crime levels and patterns changed in the last four years/recent years? · How much crime is reported to the police? · What responses do people have to crime? · Do people feel they can do anything about crime? · To what extent do people in this dusun/desa carry out their own forms of justice and what does this involve? (e.g. lynching suspected criminals) · If it does happen, to what extent is it socially accepted? Does anyone disagree with it and why? 10. Population Flows (general) a. Displacement/Migration · Over the past four years, how many people have moved here from outside the desa? Where were they from? · Do you know why they came here? · Are newcomers to the desa richer or poorer than you? · Over the past four years, how many people have left the desa? · Where did they go? · Do you know why they left? · Have any come back? · Have households who have had a member(s) leave grown richer or poorer? b. Female-Headed Households · How many female-headed households are there in this desa/dusun due to male household members migrating overseas for work? · How many of the male migrant workers come back? · What is the situation of the female-headed households if the male household members do not come back? 11. Other Conflict Resolution Mechanisms 70 (general) a. Functional Groups · Which functional groups do people use for conflict resolution?51 · For what types of problem/conflict do people use functional groups? · Are functional groups ever used to solve inter-ethnic problems? · How successful are these groups at solving problems? Are there any specific types of conflict they are good at resolving? What types are they weak at solving? b. Village Security Groups · Is there a village security group? · If so, how long has it been in operation? · Who takes part? · What is its role? · Do people ever turn to this group--or people within the group--to solve problems? · If so, what types of problems? · How effective is it? c. Illegal Groups · Do villagers ever turn to illegal groups to solve problems? · What kinds of illegal groups are involved in problem solving/conflict resolution? · What kinds of problems/conflicts do they address? · How effective are they? d. Other Groups · What other groups exist to solve problems? · How long have they been in existence? · What problems do they address? · How effective are they? · Who are these groups initiated by? 12. Types of Conflict (general) a. Characteristics · Which forms of conflict exist? · How often do they lead to violence? b. Causes · What tend to be the triggers of the dispute? · What tend to be the underlying tensions? c. Consequences · What are the physical consequences in terms of deaths/injuries/destruction? · What are the other impacts of the conflict? d. Actors and Location · Which actors or parties are involved in the disputes? · What is the extent of wider community involvement? · What are the locations of disputes? 13. Socio-Cultural Factors (general) 14. Development Projects (non-KDP)52 51By functional groups, we mean the categories listed in section 6a--credit/finance, production, recreational, political groups. 71 (hypotheses 3) a. Presence and Knowledge of Development Projects · What development programs have there been in the desa/dusun? · How many years have they been in the desa/dusun? · Have there been any new projects in the last five years? · Which ones did you have in this dusun? · What were they for? · Do you know how much funding the projects had? · Do you know where their source of funding was from? b. Participation in Development Projects · Which people in the dusun/desa were involved in the projects? · Were you involved? · If so, in what ways? · Did you ever go to any meetings about the projects? · Were you involved in any decision or planning meetings about the projects? · Who made the decisions about these projects? c. Tensions with Development Projects · Were there any tensions or problems caused by these development projects? · What kinds of tensions or problems did they cause? · Why did they cause those tensions or problems? · Can you remember what happened? · Were these problems resolved? · If so, how were they resolved? · If not, why were they not resolved? d. Outcomes and Impacts of Problems with Development Projects · Have any of the problems over development projects become violent? · If so, in what ways? · What other impacts did the problems over the project have in the dusun/desa? 15. KDP Process and Mechanisms (hypotheses 3) a. Participation in KDP There are four kinds of meetings to ask informants questions about: · KDP socialization meetings? o dusun, desa, kecamatan levels · KDP planning meetings? o dusun, desa, kecamatan levels · KDP decision making meetings? o dusun, desa, kecamatan levels · KDP implementation meetings? o dusun, desa, kecamatan levels For each of the meetings listed above, ask the following questions: · If you went to any KDP meetings, which ones did you go to? · If you went, who invited you, or how did you hear about it? · If you were not invited, could you still go to the meeting? · How many times in one year did they have meetings in this dusun/desa? · If they did not have meetings, do you know why not? 52Including government development projects and external donor or NGO projects. 72 · Were the meetings different in any way to other meetings for development projects? · If so, in what ways were they different? · Who were the key decision-makers in the meetings? · What was the role of the kepala desa in the meetings? · Were decisions made by consensus/voting? · If not, how were the decisions made? · What were the meetings like in the first year? · What were the meetings like in the second year? · What were the meetings like in the third year? · Was there any change in the meetings from year to year? · If you did not go to any meetings, why not? · If you did not go, did anyone in your household go to the meetings? b. Competitive Bidding Process in KDP · Was there a competitive process to select proposals in this dusun/desa? · During the bidding process, how did they prioritize projects? · Who made those decisions? · Was there any discussion about prioritizing projects? · If there was no consensus, how did they make the decision about which proposal to prioritize? · At the UDKP meetings, how did they make the decisions about prioritizing proposals? · Was there any discussion about the selection? · If there was no consensus, how did they make the decision about which proposal to prioritize? c. Proposal Writing Process in KDP · How did the proposal writing team get selected in this desa? · Did you know the results of the proposal writing? · Who informed you about the result of the proposal writing? · Was there any socialization in this desa about the results of the proposal? d. Facilitation of KDP · Was there an FD in this dusun/desa? · What was their role? · How were they selected to be the FD? · Was the FD related to any of the village leaders? · At the KDP meetings, was there anyone present from outside the dusun/desa? (i.e. FK) · What was their role? e. Tensions in KDP Process · Were there any tensions about project design/selection/funding/ implementation? · Tensions over allocation of projects to one dusun/desa and not others? · Tensions over manipulation of project funds · How serious were these tensions or problems? · Did the tensions ever lead to violence of any kind? · Who resolved the tensions or problems? · How were they resolved? · Why was that person involved? · If the problems were not resolved, why was that? f. Complaints Mechanisms · If there were any problems or complaints you had with KDP, where did you address the problem? · Who addressed your problems with KDP? · How effective was that person/institution? · If you did not have access to any complaints mechanism about KDP, why was that? · Did other people have access to complaints mechanisms? 73 · Was the problem was solved through this person/institution? · If not, why not? 16. KDP and Conflict/Conflict Resolution (hypotheses 3) a. Where KDP Facilitators Addressed Non-KDP Related Conflict Issues · Were any problems not related to KDP addressed by KDP facilitators (FD or FK)? · If so, why was that? · What kind of problems did they address? · Were they effective in solving the problems? · If so, why? · If not, why not? b. Where KDP Forums Were Used to Address Non-KDP Related Conflict Issues · What kind of forums did you have in this desa before KDP? · Did you have any inter-village forums before KDP? · Were any problems not related to KDP addressed through KDP forums? · Ask for dusun/desa/kecamatan · If so, why was that? · What kind of problems were addressed in the forums? · Were the forums an effective place to solve the problems? · If so, why? · What was it about the forums that mean the problems could be resolved? · If not, why was that? · What were the limitations of the forums? c. Where KDP Has Had a Positive Effect on Social Relations between Groups · Did members of different groups53 work together on any of the projects in this desa/dusun? · If not, why not? · If so, which groups worked together? · If so, did working together on KDP projects increase "solidarity" between groups? · If so, was that unusual or normal for those groups to work together? · If it was unusual, why did they work together on this project? · Did members of different groups meet together at the KDP forums? · Which groups came to the meetings? · Did any groups not attend the meetings? · Do you know why not? · Did any groups come to the meetings that did not normally attend meetings? · If so, do you know why they came to these meetings? d. Where KDP Has Led to Non-KDP Related Conflicts · Have disputes over projects ever spilled over and led to other disputes, outside the project? · Have these disputes ever become violent? · Who was involved in resolving these disputes? A3: Case Data Topics, Codes and Field Question Guide 1. Background to the Dispute/Conflict · When did the conflict occur (date, time)? 53This could be different groups based on identity (religion, ethnicity, ancestry, political affiliation) or on income-level or other structural groups. 74 · How long has it lasted/did it last? · Where did the conflict take place? · Who was involved? · Was it an individual or group-based conflict? · How many people were involved and which groups? 2. Causes (Tensions and Triggers) · What were the underlying tensions that led to the conflict? · What were the trigger events that led to the conflict/that escalated the conflict? · Why did the conflict take place? 3. What Happened? (History) · What was the chronology of events--the sequential evolution of the conflict? · At what points did it escalate? Why? · At what points did it dissipate? Why? · What is the current status of the conflict? 4. Impact and Effect of the Dispute/Conflict · How many people were injured/died in the conflict? · Who were they? · How did they get injured/die? · Where did they get injured/die? · Was there any property destruction? · If so, who did it and what was destroyed? · If so, why did they destroy property? · If so, whose property was destroyed? And why were they targeted? · What were the indirect impacts of the conflict? · Has the conflict changed relations between individuals/groups? · If so, in what ways? · Have there been any behavioral changes since the conflict? · If so, what changes and by whom? · Have there been any economic impacts from the conflict? · If so, what are they and who is affected? 5. Mediation/Intervention · Were they any attempts at intervention? · If so, by whom? · What happened? · What was the effect: positive, negative or did it make no difference? · Why was there that effect? · Did formal state institutions/authorities get involved? · If so, which ones and in what ways? · If not, why not? · Did informal organizations/authority figures get involved? · If so, which ones and in what ways? · If not, why not? 6. Outcome of Mediation/Intervention · Was the conflict resolved? 75 · If so, what does this mean? · If not, what are the prospects for resolution? 76 Annex B: Interview Guidelines54 B1: General Interview Guidelines 1. Get concrete incidents, not theories · Always ask for a specific real incident about something that actually happened, rather than a generalization (seek observations, not conclusions) · For example: Don't ask the respondent if they think levels of crime are high. Ask if they can remember a specific incident, and get them to tell you about it · Don't let the respondent do the analysis; leave it for later 2. Keep your questions `neutral' · Don't lead the respondent to a certain answer (the one you want) · Don't put words in the respondent's mouth · The respondent should speak far more than you! · Harder than you might think! 3. Let respondents bring up what you looking for · In most cases, don't ask directly about what you want · Ask more general questions that give respondents an opportunity to talk about what you want to hear 4. Don't assume what you are hearing is always true · Verify information you hear from multiple sources (triangulation) · Get perspectives from all sides · Truth is built from many perspectives ­ and all are interesting! 5. Take your time ... · A first meeting is partly about establishing an interviewing partnership · Take time to build relationships with interviewees · Time spent talking about `unimportant' things is never wasted 6. ... But keep focused · You are not writing an ethnography about all elements of culture · Selectively record relevant information 7. Interview a cross-section of the population · Often you get the best information from the most unexpected source · Be wary of turning to authorities and `experts'--ordinary villagers will probably be more helpful 8. Record respondents' own words · They are probably more revealing (and more powerful) than your own! · Especially true when writing case studies 9. Keep the analysis separate · Don't mix analysis and evidence taking in your notes · Use certain formats for recording evidence (what they said); some for doing analysis 54This annex reprints the different guides on interviewing the researchers were given. While some may be particular to Indonesia (and the local situation), other `tips' may be useful for those conducting qualitative research on conflict elsewhere. 77 · Don't spend time developing theories and typologies ­ that is for later 10. Write-up notes as soon as possible · Either take good notes during the interview or straight after · Do not wait until the evening--you will have forgotten much · Always write up your notes on the same day · If you work in pairs, while one person asks questions, another takes notes 11. Don't mention conflict directly! · Ask about problems in the village, or difficulties they have--at most about disputes · Use questions about crime, young men, the adat system, etc., as ways into the discussion about conflict · Paint things in a positive way--don't make it seem like you are making judgments 12. Quality not quantity · Better to get a few good interviews than lots of bad ones · Take time to write up everything you have heard (that is relevant) 13. Work as a team · If you are working in pairs, decide who is leading the interview (take it in turns) · Never interrupt each other. It makes you both look bad! B2: Guidelines for Accessing and Interviewing Government Officials Accessing and interviewing government officials requires some special considerations. The researchers should adapt these as necessary depending on the local context. 1. Aim for a first interview with the informants listed, or the head of the listed institution or department. a. You will have a copy of the surat (letter) to help you access the official. b. Do this interview for general access purposes even if you think they do not have the information you are looking for. c. Spend as much time as you need with the head, don't rush any first or second appointments. d. But don't take too much of their time if they seem busy. Just introduce yourself, and ask if you may come back to them later. e. If you sense any problems, or think that you have reached a sensitive point, retrace your steps and go back to talking about "neutral" issues until they are comfortable again. f. Take it especially slowly with the military and police at the district levels. If they say they have five minutes, only stay longer if they indicate they want you to. 2. If you cannot get an appointment with the head, talk to their nearest representative. 3. Your objective is to get access to other officials within the head's department as well as to get information from the head. Make sure this is ok with the head. 4. Always interview other officials or representatives to cross- check information. a. For example, at the land department, interview the land titling officer, not just the head. b. If the head won't leave you alone for these interviews, come back another time when you can talk to other informants on their own. B3: Guidelines for Accessing and Interviewing Military and Police 78 The military and police are some of the most sensitive informants you will interview. These sources also have some of the key information we need on crime and violence. These guidelines should give you some help on how to access this information. 1. Your first interview should be for introductions only. Ask to meet the head of the district police or military, or their representative if they are not available. 2. At the district office it may be that chatting to the secretary will be a much more useful exercise than the "real" interview. But don't let them be suspicious of you--remember that you have to work in this area for five months. 3. Briefly tell the chief about the research and say that you hope they will be able to help you with information. Be vague about your research objectives until you feel they are comfortable and trust you. 4. Entry points: discuss the "local situation and problems." Ask them about logistics, how they "manage" the area, the structure of their organization, their "strategy" etc. 5. Flatter their expertise and knowledge of the local context and situation. 6. Be very nice and very soft. 7. DO NOT talk directly about conflict or violence at the first meeting. Simply tell them you are there to get information about "local issues and problems in the community." Soon enough, they will talk about crime and violence. 8. DO NOT take notes until you have established a very comfortable rapport--you may never get this at the kabupaten police office, so just stick to introductions. 9. In any interview, if they start to appear angry, be patient and slow down. 10. If you sense any problems, or think that you have reached a sensitive point, retrace your steps and go back to talking about "neutral" issues until they are comfortable again. 11. Take the discussion very slowly. If they say they have five minutes, only stay longer if they indicate they want you to. This is unusual at kecamatan, but common at kabupaten. 12. In the kecamatan, talking outside of the office, in the evening, or early in the morning, can be the best time to get them to open up. 13. The military have very detailed data at the kecamatan level on social tensions, political conflicts, etc. But it is hard to get them to reveal it. Try to get it. If they do show you their data, do not write it down in front of them. B4: Guidelines for Accessing and Interviewing Health Officials Another key source of information is health officials. They may hold data (formal or informal) on injuries and deaths from violence--a key proxy for levels of violent conflict. The following guidelines should help you obtain data from officials at the kabupaten and kecamatan levels. Kabupaten level · Introduce yourself to Kepala DINAS Health (or his representative). 79 · Say you are doing research on `disputes' and want to find out information on levels of injuries from disputes. · Ask if they have any information on injuries from violent conflict. · Even if they say no, ask if you can talk to them about their opinions. · You can generally be more direct with health officials than you can with other government officials. · Ask for permission to interview relevant staff (e.g. at district hospital). · Talk to the chief of hospital (or his representative). · The Chief of hospital may refer you to a nurse or doctor. · Interview as many as you can. The main areas to get information on are: · Levels of injuries/deaths from violent conflict · Injuries/deaths from different `types' of violent conflict · Locations where injuries/deaths took place (where referred from ­ by kecamatan, desa if they have it) · Characteristics of victims (religion, ethnicity, class, sex, age) · Types of injury (including severity) · Any changes in injury rates/types Information will mainly be `informal' data--e.g., unofficial estimates. However, you may be lucky to get some `official' statistics. Kecamatan Level · Introduce yourself to head of Puskesmas · You need to talk to: (1) doctor; (2) paramedic; (3) nurse(s); (4) midwife · Interview as many of the above as you can · Say you are doing research on `disputes' and want to find out information on levels of injuries from disputes. · Ask if they have any information on injuries from violent conflict. · Even if they say no, ask if you can talk to them about their opinions. The main areas to get information on relate to levels of injuries. You should be able to gather data on: · Levels of injuries/deaths from violent conflict · Injuries/deaths from different `types' of violent conflict · Locations where injuries/deaths took place (where referred from--by desa) · Characteristics of victims (religion, ethnicity, class, sex, age) · Types of injury (including severity) · Any changes in injury rates/types In addition, health officials may be able to give information on population flows. Again, data will mainly be of an `informal' type--unofficial estimates. You may also be able to gather stories/cases of injuries from conflict--write these on your case study format. 80 Annex C: Data Recording Formats--Qualitative Research Phase 155 C1: Standard Data Format Code No: Researchers present: Interviewer(s): Note-taker(s): Date: Time: Length (hours, minutes): Where held: Kab: Kec: Desa: Dusun: Location/Place: Other people present: Respondent: Gender: Age: < 15 41-50 15-21 51-60 22-30 > 61 31-40 Organization (if applicable): Position held: Ethnicity: Religion: Place of birth: Time lived in present location (Kabupaten): 55These are English versions of the data formats; the versions used in the field were in Bahasa Indonesia. For fully formatted and printable version of all data collection formats used throughout the research, see Barron, Diprose and Smith (2004). 81 C2: Standard Interview Recording Format Date: Code # (Interview): Kabupaten location: Interviewee: Instructions: - Summarize all interviews on these sheets after the interview. - As much as possible, use the respondents' own words. - Very important: Do not do the analysis yourself at this point. - Record exactly what the informant says here, not what you think. - Put actual case studies on the separate format sheet. - Use your diary/analysis sheets for your own analysis. - You do not need to fill the whole sheet. 1. Characteristics of dispute/conflict in kabupaten/kecamatan 2. Cause of dispute/conflict in kabupaten / kecamatan 3. Levels/seriousness of dispute/conflict 4. Locations of disputes (by kecamatan / by desa) 5. Parties involved in disputes (where they live, ethnicity, religion, age, sex) 6. Underlying tensions 82 C3: Case Study Recording Format Date: Coding #: Kabupaten location: Interviewee: Instructions: - Summarize the key aspects of any story/case an interviewee tells you on this sheet - Each sheet should only record the story of interview. You will not combine the stories from several different interviews (that is, from all respondents) to create a full case until later. - As much as possible, use the respondents' own words. - If there is more than one respondent in the interview, note which respondent provided the information - Mark where it is the respondent's own words (or a translated version) in quotation marks. - Very important: Do not do the analysis yourself at this point. - Record exactly what the informant says here, not what you think. - Put other parts of the interview not related to the story on the separate format sheet. - Use your diary/analysis sheets for your own analysis. - You do not need to fill in the whole sheet. 1. Background to the dispute/conflict 2. What happened? 3.Why it happened? 4. Impact and effect of the dispute/conflict 5. Status of the conflict 6. Attempts at conflict management/resolution 83 C4: Health Officials Recording Format Date: Coding #: Kabupaten location: Interviewee: 1. Number of `accidents' related to conflict referred Estimate / Official (a) # Per Month (2003 and 2002): (b) # Per Year: 2003 2002 2001 2000 1999 1998 1997 (c) Trend: up down around the same Timespan: Level of confidence: 2. Number of referred accidents related to conflict that end in death (at the hospital) (a) # Per Month (2003 and 2002): 84 (b) # Per Year: 2003 2002 2001 2000 1999 1998 1997 (c) Trend: up down around the same Time span: Level of confidence: 3. `Types' of accident Injuries (deaths) 2003 2002 2001 2000 1999 1998 1997 Stabbings Shootings Burnings Beatings Other categories related to conflict: Trends: Injury type Timespan up down same (please circle) 85 up down same up down same up down same If official statistics, note sources here: -------------------------------------- 4. Locations where violent conflict-related injuries/deaths took place Across kecamatan Kecamatan with highest levels in kabupaten: Desa with highest levels: Non-KDP kecamatan Level in our non-KDP kecamatan: 86 Desa within our non-KDP kecamatan: KDP kecamatan Level in our KDP kecamatan: Desa within our KDP kecamatan: 5. Characteristics of victims (a) Victims by ethnic group (%) Timespan: (b) Victims by religious group (%) Timespan: Catholic Muslim Protestant Hindu Buddhist Other (c) Victims by sex (%) Timespan: Male Female (d) Victims by age group (%) Timespan: < 15 51-60 15-21 > 61 22-30 31-40 41-50 7. Notes 8. Other Information 87 - For other information use the standardized sheets, but staple them together with this sheet. - In addition to health `stats' you can get information on: o main types/patterns of conflicts o their views on the conflict situation o who may be inflicting the injuries o the extent to which violent conflict is big problem for health providers - Try to record as much as possible in their own words - Use case study format to record actual cases--e.g., stories about people they treated 88 Annex D: Data Recording Formats--Qualitative Research Phases 2A and 2B56 D1: Standard Data Format Code No: Researchers present: Interviewer(s): Note-taker(s): Date: Time: Length of interview: Where held: Kab: Kec: Desa: Dusun: Location/Place: People present: Respondent: Gender: Age: < 15 41-50 15-21 51-60 22-30 > 61 31-40 Organization (if applicable): Place of birth: Position held: Time lived in present location (Kabupaten): Ethnicity: Context 56The data formats used in Phases 2A and 2B of the qualitative research were by and large the same. However, some minor modifications were made to the 2A formats based on the experience of the researchers in that Phase, primarily to make it clearer to track which transcript relates to which case and which informant. The formats presented here are the updated formats used in Phase 2B. The actual formats used in Phase 2A can be found in the Field Guide for that Phase (see Barron, Diprose and Smith, 2004). The versions presented here are in English. The actual versions used were is Bahasa Indonesia. See the Bahasa versions of the two field guides for the Indonesian versions. 89 D2: Standard FGD Data Format Code No: Where held: Kab: Researchers present: Kec: Interviewer(s): Desa: Note-taker(s): Dusun: Date: Location/Place: Time: Length (hours, minutes): FGD participants: No. Name Age Education Occupation Ethnicity Religion 1 2 3 4 5 6 7 8 Notes: Include here comments on (a) atmosphere, (b) any problems during discussion, (c) non-verbal signals from respondents Please carry onto separate sheet if necessary. 90 D3: Standard Interview Transcript Recording Format code here> D4: Focus Group Discussion Transcript Recording Format Participant> 91 D5: Case Study Recording Format Date: Coding #: Kabupaten location: Interviewee: Case Name: Instructions: - Summarize the key aspects of any story/case an interviewee tells you on this sheet - Each sheet should only record the story of interview. You will not combine the stories from several different interviews (that is, from all respondents) to create a full case until later. - As much as possible, use the respondents' own words. - If there is more than one respondent in the interview, note which respondent provided the information - Mark where it is the respondent's own words (or a translated version) in quotation marks. - Very important: Do not do the analysis yourself at this point. - Record exactly what the informant says here, not what you think. - Put other parts of the interview not related to the story on the separate format sheet. - Use your diary/analysis sheets for your own analysis. - You do not need to fill in the whole sheet. 1. Background to the dispute/conflict 2. Causes (tensions and triggers) 3.What happened? (history) 4. Impact and effect of the dispute/conflict 5. Mediation/Intervention 6. Outcome of mediation/intervention Annex E: Data Recording Formats--Qualitative Research Phase 357 ______________________________________________________________________________ E1: Standard Data Format See Annex D1 E2: Standard Interview Recording Format See Annex D2 E3: Case Study Recording Format See Annex D5 57The key informant surveys use in Phase 3 of the research can be found in Barron, Diprose and Smith (2004). 93 E4: KDP Data Format: Kecamatan level Kecamatan: Instructions: After each piece of data you enter in this form (including in all the boxes in the tables) write the interview code(s) in brackets ( , , ). If other interviews verify the information, write the other codes using commas as indicated. If another source gives a different answer, write both answers in the same box or section with the interview code after each answer. While you are in the kecamatan, make sure you collect any of the "Kecamatan Dalam Angka" or other statistical books/sources that the kecamatan government may hold. Please collect these books/sources for the most recent year as well as the years KDP was implemented (or as many as possible). For each of the sections/ columns referring to other information or additional comments, you can fill in either your own clarifications or quotations from the informants. If it is your own clarification, please precede the comments with your initials (eg RD:). If you directly quote the informant, use the quotation marks ( see the field guide instructions) and mark their code and position clearly (eg. Camat, 152 "The participation of women increased significantly over the three years [of KDP]") PART 1: Basic Information on Kecamatan 1. Kecamatan name: 2. Kecamatan received KDP funding for which year(s): 3. Current population of Kecamatan ­ most recent data (no. inhabitants, no. households, year, source): 4. Population of Kecamatan (no. inhabitants, no. households, year, source) (KDP Year 1): (KDP Year 2): (KDP Year 3): Extra Year: PART 2: Basic information on FK Table 1: FK details for each year of KDP1 and KDP 2 (if relevant) 1. KDP 2. FK ­ male: 3. FK ­ female: 4. Other relevant (Year) name (place living prior to name (place living prior to becoming information becoming FK) FK) KDP 1 Yr 1 ( ) KDP 1 Yr 2 ( ) KDP 1 Yr 3 ( ) KDP 2 Yr 1 94 ( ) Additional Comments for Table 1 (e.g. Describe unusual changes in FK with reasons, as well as problems they faced in selection process) _____________ PART 3: Basic information on Participation in KDP Kecamatan-Level Meetings Table 2: List of Villages in Kecamatan and Participation in KDP Kecamatan-level forums 1. Village Name 2. Participate in 3.Participate 4. 5. 6. Reason for not (List all villages in Kec-level in Kec-level Participate Participate participating in a the Kecamatan) Forums Forums in Kec-level in Kec-level particular year (State Year 1 ( ) Year 2( ) Forums forums year and reason) Year 3 ( KDP 2, Yr 1 ) 1. Additional Comments for Table 2: 5. In general, how many people attended the kecamatan-level forum meetings in each of the following years (give approximations of average number per meeting, as well as percentages for men and women): KDP 1, Year 1 (if applicable): KDP 1, Year 2: KDP 1, Year 3: KDP 2, Year 1 ( if applicable): PART 4: Funding for KDP Projects and Villages Table 3: Villages Receiving Funding from KDP (NB: If you cannot get the information on the amounts of the projects, try and fill this in from the village level interviews) 95 1. Name of 2. Year 1 3. 4. Year 2 5. 6. Year 3 7. 8. KDP 9. Village ( ) Amount ( ) Amount ( ) Amount 2, Yr 1 Amount (distance Activity Rp (if Activity Rp Activity (s) Rp Rp from kec. (s) known) (s) capital near/far) 1. Additional Comments for Table 3 _____________ PART 5: Other Development Projects in this Kecamatan 6. Over the past 5 years what other development programs/projects have been implemented in this Kecamatan? Fill in Table 4 Table 4: Other Development Programs in Kecamatan 1. Name of 2.Years 3. Purpose 4. Source 5. Grass 6. Grass 7. Villages that Program/Project Implemented / of Program of roots Roots received project Received Funding community Community (include list funding (Year participatio participation separately if ____ to _____) n in in long) planning? implementati (yes, no) on? (yes, no) 1. Additional Comments for Table 4 (please include here any specific problems with any of these programs (name program, village where problem occurred), explain the problem, what happened and how it was resolved, as well as any spillover impacts of the program on the community) 96 E5: KDP Data Format: Village level Village Instructions: After each piece of data you enter in this form (including in all the boxes in the tables) write the interview code (s) in brackets ( , , ). If other interviews verify the information, write the other codes using commas as indicated. If another source gives a different answer, write both answers in the same box or section with the interview code after each answer. While you are in the village, make sure you collect any of the statistical books/sources that the village government may hold. Please collect these books/sources for the most recent year as well as the years KDP was implemented (or as many as possible). For each of the sections/ columns referring to other information or additional comments, you can filled in either your own clarifications or quotations from the informants. If it is your own clarification, please precede the comments with your initials (eg RD:). If you directly quote the informant, use the quotation marks (see the field guide instructions) and mark their code and position clearly (eg. Camat, 152 "The participation of women increased significantly over the three years [of KDP]") PART 1: Basic Information on Village 1. Village name: 2. Village received KDP funding for which year(s) of KDP 1: PART 2: Basic information on FD Table 1: Names of FD for each year of KDP1, and KDP 2/reward year (if relevant) 1. KDP 2. FD ­ male: 3. FD ­ female: 4. Other (Year) name (place living prior to name (place living prior to becoming relevant becoming FD, place of FD, place of origin) information origin) KDP 1 Yr 1 ( ) KDP 1 Yr 2 ( ) KDP 1 Yr 3 ( ) KDP 2 Yr ( ) Additional Comments for Table 1 (eg. Describe unusual changes in FD with reasons, as well as problems they faced in selection process) 97 PART 3: Basic information on Participation of Villages in KDP Kecamatan-Level Meetings Table 2: Participation in KDP Kecamatan-level forums This will be cross referenced with the information from the Kecamatan 1. Village 2. Participate 3.Participate 4. Participate 5. 5. Reason for Name in Kecamatan- in Kecamatan in Kecamatan Participate in not level Forums - level Forums - level Kecamatan- participating in Year 1 ( ) Year 2 ( ) Forums Year level forums a particular 3 ( ) KDP 2, Yr year (State 1/Extra Year year and KDP reason) Additional Comments for Table 2: _____________ PART 4: Funding for KDP Projects at the Village level Table 2: Funding from KDP (NB: This will be cross referenced with the kecamatan level information. Only fill in the rows for when the village received funding, draw a line through the rows which are not relevant) 1. Year 2. 3. 4. 5. 6. 7. 8. 9. of KDP Activity Amount Activity Amount Activity Amount Activity Amount (Year) 1 Rp 2 Rp 3 Rp 4 Rp KDP 1, Year 1 ( ) KDP 1, Year 2 ( ) KDP 1, Year 3 ( ) KDP 2, Year 1/ reward year KDP 1 ( ) Additional Comments for Table 2 98 PART 5: Other Development Projects in this Village 7. Over the past 5 years what other development programs/projects have been implemented in this Village? Fill in Table 4 Table 3: Other Development Programs in Village __________________ 1. Name of 2. Years 3. Purpose 4. Source 5. Amount 6. Grass 7. Grass Program/Project Implemented / of Program of of Funding roots Roots Received Funding Rp. community Community funding (Year participatio participation ____ to _____) n in in planning? implementati (yes/no) on (yes/no) 1. Additional Comments for Table 3 (please include here any specific problems with any of these programs (name program, explain the problem, what happened and how it was resolved, as well as any spillover impacts of the program on the community). If the information is very detailed, fill in the interview transcripts and write here the codes of the transcripts referring to this information. _____________ 99 E6: Kecamatan Demographic Form Name of Kecamatan:_______________________________ Basic Instructions After each piece of data you enter in this form (including in all the boxes in the tables) write the interview code (s) in brackets ( , , ). If other interviews verify the information, write the other codes using commas as indicated. If another source gives a different answer, write both answers in the same box or section with the interview code after each different answer. ALL INFORMATION ON THIS FORM MUST HAVE A SOURCE. Remember that you are not recording this information in quote form to save time, but this does not mean that you do not need to record the source of the information. Be vigilant!!! Part A: Kecamatan Demographics Table 1: Ethnic and Religious Groups at the Kecamatan Level 1. Main Ethnicities or lineages 2. Main Religions (as percentage of (as percentage of total or figure total, or figure for a particular year, for a particular year, note year) note year) 1. 1. 2. 2. 3. 3. 4. 4. 5. 5. Additional Comments for Table 1 (Have there been any significant population changes? Such as large influxes or outflows of migrants. Do people practice their religion or is it a nominal identity? Etc. Note if particular villages are experiencing these changes) (1) List of Key Occupations 1. List here the main occupations in the kecamatan (include figures if available, and year) 1. 6. 2. 7. 3. 8. 100 4. 9. 5. 10. Year ________________Source:________________________________ Additional Comments (include here any occupational changes over the last few years eg. TKW, new project work, note the villages) (2) Education We want to gauge education levels in the kecamatan. 1. Compared with other kecamatan in the kabupaten is the education level: 1. High 2. Average 3. Low 2. Approximately what percentages of people in this kecamatan complete education to the following levels: 1. No education _________________(%) Source: ________________ 2. SD or equivalent ______________ (%) 3. SMTP or equivalent____________(%) 4. SMTA or equivalent____________(%) 5. Higher education_______________(%) 3. Which villages are more/less educated (indicate village, general level of education, and degrees of difference) 4. Describe any difficulties in accessing educational facilities for particular villages (3) Poverty & Unemployment 1. Is the kecamatan particularly rich or poor? 2. What are the levels of unemployment in the villages in this kecamatan, and the kecamatan in general. Are there any villages which stand out? Why? What about the kecamatan (include information on the levels of unemployment, and if there are seasonal differences) 101 (4) Health 1. What health facilities are available at the kecamatan level - List? (Puskesmas, doctors, nurses etc)? 2. Are there any particular health problems in particular villages? (5) Kecamatan Geography Table 2: Geography in the Kecamatan 1. 2. Borders with 3. Access to Capital 4. Other Geographic Information Area other kecamatan (road type/quality, (kmē) distance from kab. capital) Additional Comments for Table 2 (list here any cleavages/tensions or community polarization that occurs along geographic lines, including the stereotypes) _____________ Part B: Kecamatan-Level Interaction Aside from Government Institutions, and political parties, list the other institutions which exist at the kecamatan level, and how long they have been present where possible (eg. NU, Muhammadiyah, NGOs, Pesantren networks, adat groups, distinct bajingan groups, sports groups, silat groups, prayer groups, women's groups, savings and loans groups, credit groups, village head associations etc). Include information on the composition of the groups ­ across ethnic, religious, geographical, and economic lines. Table 3: Groups at the Kecamatan level 1. Group 2. Length 3.Other information (eg member numbers, of time in composition) Kecamatan Additional Comments for Table 3 102 Part C: Governance and Leadership (1) Kecamatan Government Institutions This section summarizes the government institutions at the kecamatan level. Name of Camat: Length of time as Camat in Kecamatan: Previous appointments: Place of Origin: Place living prior to appointment: (2) Tokoh Masyarakat List the tokoh masyarakat/key leaders in the kecamatan only if they have significant influence. Describe what type of tokoh they are and the extent of their influence (that is their affiliations, ties, nature of their leadership and who they influence). Table 4: Tokoh Masyarakat in the Kecamatan Name of Tokoh Masyarakat Type of Tokoh Nature of leadership Masyarakat Part D: Broad Types of Conflict in the Kecamatan Refer to list of types of conflict to fill in this section. This can be used as a partial basis for your decision making on which villages to visit. Table 5: Types of Conflict in the Kecamatan 1. Type of Conflict 2. Villages where prevalent Additional Comments for Table 5 (note here if there are villages which are more or less violent than others and for what types of cases) _____________ 103 E7: Village Demographic Form Name of Village: Name of Kecamatan: Basic Instructions After each piece of data you enter in this form (including in all the boxes in the tables) write the interview code (s) in brackets ( , , ). If other interviews verify the information, write the other codes using commas as indicated. If another source gives a different answer, write both answers in the same box or section with the interview code after each different answer. ALL INFORMATION ON THIS FORM MUST HAVE A SOURCE. Remember that you are not recording this information in quote form to save time, but this does not mean that you do not need to record the source of the information. Be vigilant!!! Complete this form for every village that you visit. Part A: Village Demographics (When filling in the following table write the interview codes in brackets after each piece of information. If another source is used, such as a village profile, or kecamatan statistics, use a footnote for the source. If only the latest year of information is available, write N/A in `b' and `c') Table 1. Population, Religion and Ethnicity in the Village 1. Total Population Year 2. Total Males, 3. Total 4. Main Ethnicities or 5. Main Religions (list more than one year Year Females, Year lineages (as (as percentage of if available) percentage of total or total, or figure for a figure for a particular particular year, note year, note year) year) Yr a. a. a. 1. 1. ( ) 2. 2. Yr b. b. b. 3. 3. ( ) 4. 4. Yr c. c. c. 5. 5. ( ) Additional Comments about Table 1(Have there been any significant population changes? Such as large influxes or outflows of migrants. Do people practice their religion or is it a nominal identity? Etc.) ___________________ 104 (1) List of Key Occupations List here the main occupations in the village (include figures/percentages if available, and year) 1. 6. 2. 7. 3. 8. 4. 9. 5. 10. Year _____Source: _______________________________ Additional Comments (include here any occupational changes over the last few years eg. TKW, new project work)__________________________________________________________________________ _____________________________________________________________________________________ (2) Inequality ­ General 1. To what extent is there inequality between ethnic groups in the village? 2. To what extent is there inequality between religious groups in the village? 3. To what extent is there inequality between other groups in the village (write in)? (3) Education We want to gauge education levels in the village. 1. Compared with other villages in the kecmatan is the education level: 1. High 2. Average 3. Low 2. Approximately what percentages of people in this village complete education to the following levels: 1. No education _______________(%) Source: ________________ 105 2. SD or equivalent _____________(%) 4. SMTP or equivalent___________(%) 5. SMTA or equivalent___________(%) 6. Higher education______________(%) 3. What schools/other educational facilities/institutions exist in the village? (eg. TK, SD, SMA, SMP, Madrasah, Pesantren, etc.) 4. Describe any difficulties in accessing educational facilities for the village or dusun. (4) Poverty & Unemployment 1. Is the village particularly rich or poor? 2. What are the levels of unemployment in the village? Are there any dusun which stand out? Why? What about the village in general (include information on the levels of unemployment, underemployment and if there are seasonal differences in unemployment levels) (5) Inequality ­ Economic and Power 1. In the village is the difference between rich and poor: 1. Very High 2. High 3. Average 4. Low 5. Very Low 2. Are there any marked differences in welfare or access to power between different ethnic groups? 3. Are there any marked differences in welfare or access to power between different religious groups? 4. Are there any marked differences in welfare or access to power between other different groups? (6) Health 1. What health facilities are available in the village - List? (eg. Posyandu, polyindes, Puskesmas, bidan, etc.) 106 2. Are there any particular health problems in the village? (7) Village Geography Table 2: Village Geography 1. Area 2. List of Dusun 3. Borders with 4. Access to Capital 5. Other Geographic (kmē) other villages, (road type/quality, Information kecamatan distance from kec. capital) Additional Comments about Table 2 (list here any cleavages/tensions or community polarization that occurs along geographic lines, including the stereotypes) Part B: Village Activities and Interaction (1) Village Activities and Organizations This section is organized to record the main associational and social interactions in the village List the main forms of village interactions in terms of associational (formal) and social (informal) interactions. Formal interactions may include those interactions based on belonging to a group in the village such as prayer groups, Koran recitals, sporting activities etc; Informal interactions may include those interactions which are more spontaneous or not based on belonging to a group such as market days, wedding ceremonies, gambling, etc. Include information on the composition of the groups ­ across ethnic, religious, geographical, and economic lines. Table 3: Village Interaction 1. Associational Interactions 2. Social Interactions Additional Comments for Table 3 (2) Non-government institutions Aside from Government Institutions, the BPD and political parties, list the other institutions which exist in the village, and how long they have been present where possible (eg. NU, Muhammadiyah, NGOs, 107 Pesantren networks, adat groups, distinct bajingan groups, sports groups, silat groups, prayer groups, woman's groups, savings and loans groups, credit groups, etc) Table 4: Groups in the Village 1. Group 2. Length of time in 3.Other information (eg Village member numbers and composition) Additional Comments about Table 4 Part C: Governance and Leadership (1) Village Government Institutions This section summarizes the village government institutions in the village. This includes both the Village Government and its associated institutions, as well as the BPD. Fill in the following table in regards to village government institutions and the administration. It can include everything which has government association, including PKK and similar government organizations in the village. There is a special table for BPD information Table 5 ­ Village Governance (non-BPD) 1. Name of 2. Other Positions 3. Positions 4. Number of 5. List the 6. Other Village current in the Village which are Women in positions these Government Village Government currently Village women occupy Institutions Head, which are filled unfilled in the Administration in the village (length of village Structure administration term, administration elected in what year) Additional Comments about Table 5 (3) BPD Fill in the following table for the information on the BPD Table 6: BPD in the Village 108 1. Name of BPD 2. BPD Formed 3. Number of 4. Number of 5. Political Head in (Year) Members Women Party Domination? Additional Comments about Table 6 (include here any tension between the BPD and Village Government, any problems with formation etc) (4) Political Parties in the Village 1. What is the dominant political party in the village according to the last election results (include figures or percentages if possible)? 2. List the other political parties in the village in order of dominance, according to the last election results (include figures or percentages where possible) 3. Other comments on political parties in the village (5) Tokoh Masyarakat (non-government leadership) List the tokoh masyarakat in the village. Describe what type of tokoh they are and the extent of their influence (that is their affiliations, ties, nature of their leadership and who they influence). Table 7: Tokoh Masyarakat in the Village 1. Name of Tokoh 2. Type of Tokoh 3. Nature of leadership Masyarakat Masyarakat Other comments on Table 7 109 Part D: Broad Types of Conflict in the Village This section provides an overview of the type and incidence of conflicts in the village. The information should be based on actual cases that the informants know of. This is necessary to gain an overall picture of conflict in the village without following the cases that are not related to KDP. Where the conflict occurs frequently such as crime, you may only get data on the last year. Where it occurs infrequently, there may only be data on the past 5 years. Fill in the table wherever you can, with the most relevant information to each type of conflict. Write N/A if the information is not available, and the source in brackets. For the last 3 columns, indicate who is generally involved in the conflicts, that is whom the conflicts are between (eg. Individuals, families, groups, community vs govt, govt vs govt, govt vs private sector etc) and who generally seeks to resolve the conflicts. Indicate if in the majority of cases violence erupts or not. Table 8: Types and Frequencies of Conflict in the Village 1. Type of 2. No. of 3. No. of 4. No. of 5. No. of 6. General 7. General 8. Conflict Incidenc incidences incidenc incidences Actors Intermediarie Generall es in the in the last es in the in the 5 s y violent last 3 last 5 years prior (yes/no) year/ years/num years/ to number ber which number reformasi/ which were which number were violent were which were violent violent violent Additional Comments for Table 8 (note here if there are dusun which are more or less violent than others and for what types of cases). _____________ 110 E8: Informant List Kecamatan __________________ Desa _______________________ Research Informant Name Position in Dusun Case Study? Standard Key Informant Code Community (Yes / No) Interview Survey? Format? (Yes / No) (Yes / No)