Establishing a Link between Weak Social Capital, Social Exclusion and Peacebuilding in Northern Ireland
The research draws upon and develops a dataset constructed as part of a former Royal Irish Academy research project which utilises regularly collected administrative data across three domains (Social Need, Community Capacity and Community Capability) and original data under a fourth domain (Social Capital). This is augmented by the collection of qualitative data through a series of interviews, focus group discussions and seminars with members of the geographic communities and communities of interest, as well as policy makers and academics working in this field. Ms Kathleen Healey has been appointed as the fieldworker for this project and she is enrolled for a PhD at the Institute of Governance at Queen’s University Belfast. In addition to doctoral supervision at Queen’s provided by Professor John Morison and Ms Ruth Jamieson, the project is overseen by a steering group comprising a range of experts who meet regularly to provide support and guidance. This group contains: Brendan McDonnell, and Dr Mike Morrissey from Community Evaluation Northern Ireland QUB, Monina Oprey and Avila Kilmurray from Community for Foundation Northern Ireland, Professor Monica McWilliams, Professor of Women’s Studies, University of Ulster, and Dr Jeremy Harbison, Northern Ireland Social Care Council. To date an extensive literature review has been undertaken, a research methodology has been developed and a series of pilot interviews have been undertaken. Full interviews are now being conducted. The following provides an overview of the project, its context, methodology and some of the initial hypotheses that are being tested currently in the fieldwork by Kat Healey.
Introduction
Currently, many communities and community organisations are contributing positively to peacebuilding, community development, building social capital and improving community relations. However, many communities (geographic communities and “communities of interest”) do not have active groups, feel excluded or alienated by those in power or feel marginalised by the
1
actual processes of peacebuilding. Much attention has been turned towards this situation through flashpoint and interface violence; sectarian attacks, fear and intimidation; as well as being highlighted as a critical support issue by organisations working in the community and voluntary sector in Northern Ireland, and by Government Departments. However, it is now recognised that “throwing money” at these problems (in the form of a variety of funding programmes) cannot answer the deeper questions surrounding issues of exclusion and, in some cases, may even serve to reinforce the problems. Conducting research into establishing links and deeper learning around the issues of weak community infrastructure, low levels of social capital, peacebuilding and social exclusion may provide better options for promoting inclusion and development, contributing to a lasting benefit.
Context for the Research Project
In Northern Ireland, a substantial amount of research, funding, community development work and Government interest is currently being targeted at the concepts of social capital, community infrastructure, deprivation and social inclusion. The Peace II Programme included specific measures dedicated to improving community infrastructure and social inclusion. NI Government Programmes, especially those managed by the Department for Social Development (DSD), also have specific reference to all of these themes: namely, Urban Regeneration, the Neighbourhood Renewal Programme, the Local Community Fund and the Outreach Programme, as well as the Minister’s recent announcement of £33 million for Protestant areas through the Renewing Communities initiative. In 2004, the DSD commissioned Deloitte MCS to carry out research to develop a (quantitative) methodology for identifying areas of weak community infrastructure and, although this controversial research report was shelved, it was initially intended to form the primary means for allocating substantial community development funding through the Department. The Northern Ireland Programme for Government focuses on the issues of deprivation and social inclusion through its Targeting Social Need and Promoting Social Inclusion policies. Many policy consultations in Northern Ireland are also currently being carried out (with some being recently completed) that should have a real impact on these issues: A Shared Future, New Targeting Social Need/NI Anti-Poverty Strategy/Lifetime Opportunities Strategy, Pathways for Change, Investing Together, Positive Steps, the Review of Public Administration, the Future of Structural Funds, etc. However, there has to date been little evidence of research taking an integrated, strategic approach to these issues, in particular of research attempting to examine the interrelationship between the concepts of social capital, community infrastructure, deprivation, social inclusion and peacebuilding and the holistic (or fragmented, as the case may be) impact of programmes targeting these issues. Furthermore, specific evidence exists to show that particular communities of interest seem to be becoming further socially excluded in Northern Ireland.
2
Relevant examples include the recent increase in the number of racist attacks in Northern Ireland,1 an increasing number of attacks and robberies on elderly residents and the fact that suicide rates for young men aged eighteen to twentyfour in Northern Ireland are nearly double the per capita rate of the UK as a whole.2 By their very nature, areas of weak community infrastructure and low social capital are unable to articulate and advocate their own needs in the way that more organised and developed communities can. Vulnerable and socially excluded groups within these areas are often even further marginalised due to the lack of proper support networks/groups/organisations. This research will seek to further examine whether these groups and individuals are further socially and/or politically excluded from peacebuilding processes. It will also assess the effectiveness of particular local community-led inclusion interventions. It will attempt to explain how some communities have built confidence and social capital and have become engaged with peacebuilding processes, with institutions and with other communities, while other communities have been unable or unwilling to do so. In the context of the work of the Community Foundation for Northern Ireland’s Communities in Transition Programme and collaborative research into social capital with Community Evaluation Northern Ireland (see below) this research is in a unique position to examine the interrelation between all of these elements. Exploring the different experiences of several communities of interest across a number of different types of geographic communities, determining what barriers to social inclusion and peacebuilding they experience and assessing the impact of levels of social capital and degrees of deprivation on these barriers will certainly prove to be groundbreaking research. In the present context, this research should have a profound effect on those tasked with designing programmes and making policy regarding social capital, community infrastructure, deprivation and social inclusion. Above all, it should highlight the needs of some of Northern Ireland’s most marginalised communities and make recommendations towards addressing those needs. The main objectives of this research are to: • Examine the relationship between weak community infrastructure, low social capital, social inclusion/exclusion and their impact on peacebuilding
Over the 2003-2004 period the NI Policing Board reported 453 racist incidents in Northern Ireland, more than double that of the previous year. (Review and Report of the Work of the Northern Ireland Policing Board, 2003-2004). The 2004-2005 figures show that the number of racist incidents nearly doubled again, standing at 813, including 322 incidents of criminal damage and 187 assaults over the 2004-2005 period. (Review and Report of the Work of the Northern Ireland Policing Board, 2004-2005). 2 BBC News World Edition (news.bbc.co.uk), 8th August 2004.
1
3
•
• • • • • •
in a wide range of geographic communities and communities of interest across Northern Ireland; Identify the particular needs and experiences of members of these geographic communities and communities of interest and assess the effectiveness of particular community-led peacebuilding initiatives at overcoming local barriers; Determine if any of the communities of interest is more socially excluded than the others in areas of low social capital; Assess whether excluded groups are even more excluded if they live in areas of low social capital and/or areas experiencing high levels of deprivation; Assess whether the politically divided nature of a community plays any part in determining the levels of social exclusion experienced by these communities of interest; Compare the needs and experiences of members of each of the communities, in order to highlight the common issues and comparatively analyse the impact of peacebuilding initiatives; Embed the process and findings of the research in the relevant policy context and keep policy makers with an interest in these issues informed of the progress and outcomes of the study; Locate the unique findings in the Northern Ireland context within national and international research and theory, as well as carrying out comparative research.
The research project will adopt two distinct, but complementary, approaches: 1. The further development and analysis of a dataset constructed as part of the Royal Irish Academy research project (see below) and utilising regularly collected administrative data across three domains (Social Need, Community Capacity and Community Capability) and original data under a fourth domain (Social Capital). 2. The collection of qualitative data through a series of interviews, focus group discussions and seminars with members of the geographic communities and communities of interest, as well as policy makers and academics working in this field.
Background to the Research
The author is currently employed as Policy, Research and Evaluation Officer with the Community Foundation for Northern Ireland. This position has 4
facilitated the development of contacts, experience and learning necessary to carry out this research project. Indeed, it was through this work that the need for this research project was established. In particular, involvement in the Communities in Transition Programme over the past four years has involved carrying out action research with members of each of the communities of interest that will be the focus of this present research project, across a wide range of geographic communities experiencing high levels of community tension, deprivation, weak community infrastructure and low levels of social capital. The author also became involved in the social capital research project, funded by the Royal Irish Academy’s Third Sector Research Programme (see below), through her employment with CFNI. These two projects – Communities in Transition (CIT) and the Royal Irish Academy (RIA) project – formed the basis for the two approaches adopted in this research: the quantitative aspect, based directly on the dataset and methodology used in the RIA project; and the qualitative aspect, derived from the Longitudinal Studies and the Capacity Building Initiative strands of the Communities in Transition Project. The model developed for this research is also derived from a combination of the CIT and RIA models: CIT focuses predominantly on weak community infrastructure and peacebuilding and RIA focuses predominantly on social capital and social need, and both focus on community capacity and community capability. This research has attempted to combine these six elements into a multidimensional matrix in order to analyse and describe their interrelation. This section will provide a background summary of the CIT and RIA projects, in order to place the methodology for this research in an appropriate context. Since these two projects have such an explicit link to the methodology involved in this research, it is worth examining both in detail. The Communities in Transition Programme3 The Community Foundation for Northern Ireland initiated the Communities in Transition Programme in 2001 based on the findings of their previous Weak Community Infrastructure Demonstration Programme over the 1996 -2000 period. Through this Demonstration Programme, the issue of weak community infrastructure was researched through an action learning and support programme. The findings of the Demonstration Programme led to a lobby for the support of work targeted at areas of weak community infrastructure and resulted in the development of Peace II Measure 2.7 and other relevant initiatives, including the Rural Community Estates Programme, the Rural Community Network’s Low Infrastructure Programme and an element of the Creating Common Ground Consortium’s work.
3
This section is taken in large part from a previous publication by the present author, K. Healy, Challenges to Community Development in Areas Affected by Community Tension: The Communities in Transition Model (ed. M. O’Prey), Community Foundation for Northern Ireland (2006).
5
The findings of the Demonstration Programme, however, identified that generic practice models developed for the support of areas of weak community infrastructure were not the solution in areas where community tensions are a significant added dimension. The Community Foundation for Northern Ireland successfully sought funding from the International Fund for Ireland and a private donor, to research this issue through a further action learning and support programme – the Communities in Transition Programme. This Programme has adopted a risk-taking model for developing and supporting local work in areas of weak community infrastructure and community tension. One of the main findings of the first Demonstration Programme was that many community groups find the complexity of grant aid application forms and the level of monitoring required (for programmes such as Peace II) to be very offputting. This is especially true in areas where there is no history of managing grants. These groups and areas require particular support to enable them to “catch up” with other areas that have benefited significantly from previous funding programmes. For this reason, the Communities in Transition Programme was not designed to be a grant making programme; rather, the areas supported by the Programme were selected through a process of in-depth research and area profiling. Many of the areas selected had little history of previous community development and few, if any, active community groups. Those areas that did have community groups were either at a very early stage of development or very fragmented. None of the areas had any history of substantial previous investment in terms of finance and resources. In addition, all of the areas had experienced community tension and division (in various forms and for a wide range of reasons), marginalisation (from resources and services, in particular) and social exclusion. Perhaps most importantly in terms of the model, the Communities in Transition Programme has accepted from the outset that the engagement and ongoing participation of those most excluded is crucial to the success of development work in areas of weak community infrastructure and community tensions. Many agencies have difficulty engaging with or recognising community groups where the involvement of particular local community activists, political or paramilitary figures is evident. However, in order to effect long-term change, their inclusion is often essential. Confidence and capacity building is required for all stakeholders. This entails trying out new approaches and processes, taking risks and developing opportunities that may not have been available to date. The Communities in Transition Programme Objectives are: • To promote the establishment and development of community infrastructure in ten disadvantaged areas and to nurture and sustain the participation of those excluded or at risk of exclusion;
6
• •
To develop the capacity of the selected communities and participants to engage in inclusive community organising; To identify and address the issues of cultural, economic and political alienation of community groups within areas of weak community infrastructure that are experiencing community tension; To influence policy-making at local and regional levels in relation to areas of weak community infrastructure that are also experiencing community tension; To impact on the current perceptions of a range of political parties and their representatives with regard to the potential of community action and development.
• •
The Programme has also enabled the development of a peacebuilding initiative – the Supporting Transition Initiative – working with twelve new joint projects involving marginalised groups across several “communities of interest” from across Northern Ireland (e.g. ex-prisoners, victims of violence, women’s groups, young people, ethnic minorities and the gay and lesbian community). Although only at the mid-term stage, the progress made across all elements and against all core objectives of the Communities in Transition Programme has been significant. Local community groups have been developed and supported in ten different local areas and the capacity these groups have built – from a nil baseline in almost all cases – has been striking. To date, these projects have been able to address a range of local issues, including environmental and housing improvements, community safety, tackling antisocial behaviour, the delivery of initiatives targeting specific sections of the community (such as young people, the elderly, women and foreign nationals), training and education, health projects and community relations. Seven residential conferences have been held for members of all of the different local projects and these have been very well attended. The conferences have offered the groups the opportunity to come together and share their experiences and develop joint strategies for overcoming local barriers. They have also provided an outlet for groups and individuals, who would previously have never had an opportunity or incentive to meet with each other, to come together in a safe environment. Finally, the personal empowerment of individuals involved in the Communities in Transition Programme has been a key progress point. For this reason, the present author is currently carrying out fifty-two longitudinal studies over the life of the Programme in order to detail the stories of individuals involved in local work over the five years. The author’s ongoing involvement in the CIT action research and carrying out these longitudinal studies has allowed for the development of contacts with the individuals and groups who will be involved in the interviews and focus groups that will comprise the qualitative side of this present research project.
7
The Communities in Transition Programme uses the following definition of ‘weak community infrastructure’: ‘Weak community infrastructure exists in communities where social need and disadvantage sit alongside the absence of locally organised, locally managed, accountable and participative community development activity. It is evidenced by the lack of self-help approaches to tackling local social, educational, health, cultural, environmental and economic issues.’ 4 Fourteen characteristics contributing to weak community infrastructure have also been identified: • • • • • • • • • • • • • • An absence of previous development; Physical isolation; An absence of community venue or resources; Fear of failure; Funding-led barriers to development; Cultural barriers; Religious or ideological barriers; Political or paramilitary activity; Intra- or inter-community tensions; Development shadow; Support failure; Difficulties with neighbourhood profiles; Alienation from the sector; Rejection of community development processes.5
Three of these characteristics – intra- or inter- community tensions; political or paramilitary activity inhibiting participation; and neighbourhood profiles (in this instance, this refers to areas with high numbers of politically motivated exprisoners, victims of violence, community polarisation etc) – are specifically related to the conflict and/or community divisions in Northern Ireland. It is the experience of the CIT Programme that, in areas where these factors figure significantly, progress is much more difficult to make and new approaches are required. The Communities in Transition Programme adopted a detailed model for the identification, selection and support of areas of weak community infrastructure and community tension. CFNI carried out a mapping exercise followed by in-depth area profiling to identify geographical areas of weak community infrastructure where community tension is an added dimension,
4 5
M. O’Prey and J. Magowan, Weak Community Infrastructure: A Support Priority, NIVT (2001). Ibid.
8
inhibiting development and participation.6 This research used extensive interviews carried out in each District Council area to supplement available data on weak community infrastructure and community tensions. Each area was scored on the level of community infrastructure and community tensions, all known active community groups were identified/listed and notes were drawn up based on the qualitative interview material to complement the scores. Finally, the areas’ scores in the Noble Measures of Multiple Deprivation 2001 (see below) were taken into account. This Mapping Exercise was then presented to a panel of practice experts (the Communities in Transition Programme Advisory Group) who made recommendations to the funding body to support community development work in ten key selected neighbourhoods over a five- to six-year period. These recommendations followed on from further in-depth area profiling of the areas shortlisted based on the Mapping Exercise. A wide geographical mix of areas was selected for comparative research purposes. On the whole it was found that there were more areas in need of support in County Antrim than in any other county. This process of pro-active selection of areas was seen as the best approach as none of the areas had the capacity to make applications for funding. In most areas, no community groups existed and, where groups did exist, they were at a very early stage of development and would not have been able to make applications to Peace II or elsewhere. The ten areas selected for inclusion in the Programme (eight in Northern Ireland and two in the Border Counties) are: • • • • • • • • • • Clady and Innisrush (Magherafelt District Council) and Portglenone (Ballymena Borough Council); Crumlin, Co. Antrim; Harpur’s Hill estate in Coleraine; Harryville estate in Ballymena; The Melmount area of Strabane, including the Ballycolman, Bridge Street, Calverts, Carlton Drive and Lisnafin estates; Mosside village in the Moyle Council area; Northlands and Sunnylands estates in Carrickfergus; Seacourt estate in Larne; Newtowncunningham village in east Donegal; Six small villages in rural north Cavan – Redhills, Bunnoe, Drung, Laragh, Stonepark and Stradone.
The design of the Communities in Transition Programme was taken from the learning of the previous CFNI Demonstration Programme, in terms of the selection of areas, key elements of the practice model, nature of funding and other essential elements. However, it was recognised from the outset that more
6
M. O’Prey and J. Magowan, Mapping Areas of Weak Community Infrastructure, NIVT (2001).
9
attention to inclusion and local relationship building and negotiation would be required. The Programme was designed to provide the resources and support necessary to initiate community-led development and peacebuilding programmes in areas experiencing weak community infrastructure, community tensions, alienation, social and (in most cases) economic disadvantage. The Communities in Transition Programme seeks to build relationships within and between people and communities, to promote local participation in community development and peacebuilding and to create community cohesion by offering opportunities and support that had not been previously available to the selected areas. It recognises the particular difficulties facing communities experiencing weak community infrastructure and sees the process adopted by the Programme as an important step in redressing the imbalance created by the previous lack of investment in social capital, especially in areas that continue to directly experience either the legacy of the conflict, alienation and/or marginalisation. Community tension in Northern Ireland can manifest itself in a wide range of ways and the Communities in Transition Programme has developed a model for carrying out community development and peacebuilding work in areas affected by tensions, division and alienation. Through the experience of action research and ongoing evaluation of the Programme, the author has also developed a typology of areas characterised by community tension and this will be further developed for use in this research project. While an area may exhibit only one of these characteristics, it is likely that most areas experiencing high levels of community tension will exhibit more than one (and it is possible that one single area could experience all of the characteristics identified). This typology may prove to be a useful descriptive tool throughout the interview and focus group processes of this current research project. Intercommunity Tension – Traditional. These areas are those most frequently associated with community tension in Northern Ireland, where the two main communities have poor relations with each other, often being manifested in the form of violent conflict, sectarian displays or contested space. Interfaces are an example of traditional intercommunity tension, although there does not have to be direct geographic/physical contact between the two communities for this type of tension to exist – i.e. There does not have to be a physical interface, there may simply be rife sectarianism throughout a mixed community or, for example, rural, dispersed areas and villages. While not isolated in incidence to certain times of the year, the occurrence of traditional intercommunity tension does tend to increase over the marching season, whereas the other types of community tension do not appear to be more apparent at any particular time of the year. However, it could be said that many of them seem to be exacerbated during periods of wider political negotiations or in the run-up to elections.
10
Apart from actual violent incidents, other issues tied into this type of tension include disputes over flags, emblems and other symbols, parades, murals and various forms of sectarianism. Symptoms of traditional intercommunity tension include segregated housing, education, community organising and social activities and sometimes it requires the physical erection of “peacelines”, although this is only possible where a direct interface exists. It is more difficult to deal with when it exists within a mixed area or area where one community is predominant. Racism and racist attacks, as well as homophobia and homophobic attacks, are becoming increasingly prevalent throughout Northern Ireland. While not being the result of conflict between the two “main” communities, these occurrences could also be deemed traditional intercommunity tension in that they are often violent in nature and occur between members of two “different” communities based solely on membership of that community. On the other hand, in order to deal with it more adequately, racism and/or homophobia could make up a separate category (or two separate categories) in the typology. Intercommunity Tension – Demographic Change. Many areas in Northern Ireland have experienced radical demographic change over the past years or decades. In some areas this has taken the shape of either rapid population growth or decline. Sometimes it has caused a shift in the age profile of an area. In other cases, it has resulted in the religious/political background of an area being reversed (i.e. what were largely Protestant communities have become majority Catholic communities with the influx of new residents or vice versa). Any of these demographic changes can result in community tensions. In the case of population growth, the physical and social infrastructure is often not equipped to deal with the additional number of people. Shifts in age profile can also mean a lack of available infrastructure and resources (for young people, the elderly, etc). When the demography changes in terms of community background, it can be the case that the “old” and “new” communities fail to form relationships – they do not develop “bridging” social capital. This can occur even when the population shift does not affect the religious or political make-up of the community. Sometimes the new and established populations are from different socio-economic backgrounds. Sometimes the new community seems to fail to attach itself to the new area – they may continue to work, socialise and send their children to school in the area they moved out of, making their new community essentially a “dormitory town” for them. Intercommunity Tension – Latent. This type of tension is neither violent nor overtly sectarian. However, it has the potential to turn into violent/sectarian tension, or it may be the result of “progress” from a situation that had previously been violent and/or overtly sectarian. Latent intercommunity tension exists when the two communities live in a semblance of “benign apartheid”. While these
11
communities are not characterised by overt sectarianism, the two communities fail to mix in terms of education, recreation, shopping etc. or to socialise together. Each community tends to develop and to socialise within their own comfort zones – churches, Loyal Orders, GAA, sporting clubs and societies etc. Through lack of contact and/or direct experience of each other, they often hold negative stereotypes of each other and this can sometimes develop into fear of the other community. Whereas most areas experiencing intercommunity tension in Northern Ireland tend to have some examples of the traditional type discussed above which can manifest itself violently, many border areas in the Republic (as well as some middle class, urban and rural areas in Northern Ireland) have never experienced direct violent conflict. Their legacy of the conflict is in the form of latent tension, which can be more deeply rooted and more difficult to overcome. Intracommunity Tension – Paramilitary. Loyalist and Republican feuds and conflicts within and between the various branches of paramilitarism are examples of intracommunity paramilitary tension. Over the course of the conflict in Northern Ireland, this type of intracommunity tension has had a particularly high casualty rate. Paramilitary feuds and conflicts often occur within communities where most people do not give support to either paramilitary grouping, yet the whole community experiences the negative impact of the conflict. Although it may be purely coincidental, it is perhaps worth noting that high rates of young male suicide occur in areas where control is contested between two or more paramilitary groups from the same “side”. This is also the type of intracommunity tension that is the most feared, as it is often viewed as fighting for “turf” or territory and the outcome has implications for everyone living in the area. Apart from this, innocent victims have also borne the brunt of these attacks, including families, children, cases of mistaken identity, etc.
Intracommunity Tension – Political. This occurs when two or more political groups are seeking support from an area and “use” local people and local community groups to gain it. It can also occur when local politicians, Council workers, umbrella organisations or others block initiatives or block resources from reaching an area because of the area’s perceived political affiliation (see Gatekeeping below). Conversely, local politicians and others can also champion initiatives and projects that are not community-led and are often against the wishes and articulated needs of the local community. In addition, it can occur when local politicians or others take credit for work achieved by local community groups and individuals. This type of tension is most prevalent coming up to election time and can be very divisive, particularly for nascent community groups and projects.
12
Intracommunity Tension – Socio-Economic. Intracommunity tension can also occur between people and groups from different socio-economic backgrounds. This is sometimes in the form of overt prejudice or class issues (e.g. between groups from different educational backgrounds or groups made up of Housing Executive tenants vs. private homeowners or those who are benefit dependent and those who are not etc). In other cases it can simply result in parallel development within an area: two different groups with the same objectives develop, one group from a particular socio-economic background and the other group from a different background vying for the same resources/funding but unwilling to share or work together. Community Tension – Gatekeepers. Gatekeeping occurs when an individual or group prohibits information, resources or other necessary forms of support from reaching a local area. As mentioned above, this can take the form of an umbrella organisation or other body taking control of resources (sometimes with the best of intentions) and preventing it from reaching the local area/group for which it was intended. It could involve a key local individual, seen by agencies as a “safe pair of hands”, keeping control of all the information coming into (or going out of) an area and deciding him/herself exactly who should receive this information. It can also include politicians or paramilitary representatives who dictate who participates or what happens in an area and exclude anything (or anyone) that might negatively impact upon their power/client base. Community Tension – Antisocial Behaviour. A variety of forms of antisocial behaviour can lead to community tensions. This includes underage drinking, drug abuse, vandalism, graffiti, intimidation and physical and/or verbal abuse, death drivers and punishment attacks. While this type of behaviour does cause tension within a community, it often also tends to serve as a means of uniting the community in opposition to those committing the antisocial behaviour. However, finding solutions is problematic. Throughout the conflict, many communities became used to paramilitary groups dealing with these issues. This has changed significantly and many communities feel frustrated, abandoned and angry. In terms of trying to find local solutions, some community groups try to involve those responsible for antisocial behaviour in positive community development and peacebuilding activities; but this is frequently cited as the most difficult (and often unachievable) task for local community groups. Community Tensions – External Interference. This can occur in a variety of forms. As mentioned above, external gatekeepers can impede progress at the local area; politicians can prohibit development; external paramilitary groups can have a negative impact on development. Additionally, the failure of past externally controlled attempts at initiating community development and peacebuilding work can impinge upon future progress. Government initiatives or initiatives driven purely by funding criteria can fail to meet the researched and
13
real needs of an area and this can result in frustration and increased tension on the ground. It is possible to imagine (even to pinpoint) a community experiencing all of these types of community tension: where traditional intercommunity violence has occurred; demographics have radically changed over time; unspoken stereotypes are held by members of each community about the other community; paramilitary and political groups are vying for control of the area; prejudices exist between members of different socio-economic backgrounds; gatekeepers prevent local people from gaining access to information and resources; antisocial behaviour is rife and external interference has further deepened the community divide. Few, if any, communities in Northern Ireland are completely free from some degree of one or more of these characteristics. This is what makes community development in Northern Ireland different from community development in England or the United States and what makes peacebuilding and reconciliation work a necessary part of our community development process. Similarly, using social capital as an explanatory tool in Northern Ireland requires reference to the Northern Ireland-specific brand of community tensions that affect levels of social capital (see Sections 2 and 3 below). In Bowling Alone, Robert Putnam, the leading exponent of social capital (though not the originator of the concept), defines ‘social capital’ as “the collective value of all ‘social networks’ and the inclinations that arise from these networks to do things for each other”.7 Putnam was writing about social capital in America and the way in which social capital could influence levels of democracy in society. In a Northern Ireland context, a social capital framework enables explanation and evaluation of intracommunity relations, intercommunity relations and the balance of power between communities and decision makers. This is generally achieved through employing a model that describes levels of bonding, bridging and linking social capital, although the need for a fourth type of social capital will be argued below (see Section 2). The Communities in Transition Programme uses the following definitions of these three strands of social capital: •Bonding Social Capital – How we trust and relate to others like ourselves (Intracommunity relations). •Bridging Social Capital – Trust and relationships between individuals and groups who are unlike each other (Cross-community relations). •Linking Social Capital – The quality of relations between communities and decision makers (Relationships between communities and funders, Government, police, Councils and other statutory agencies etc). While this model offers some value in terms of explanation, it is not a prescriptive process. In other words, for the Communities in Transition
7
R. Putnam, Bowling Alone: America’s Declining Social Capital (1995).
14
Programme, there is no expectation for groups and individuals to follow along a specific continuum from bonding to bridging to linking. In fact, most groups seem to have developed bonding social capital through the initial establishment of a community group within their area and then moved on to developing linking social capital by cultivating relationships with statutory agencies, funders and other service providers and building the capacity to access resources. Bridging social capital has been cultivated within and between the groups, for the most part, through attendance at joint conferences and events and groups are beginning to develop trust and relationships with each other. Evidence of this process can be further seen through the fact that several of the groups are now working on joint projects – on a cross-community and sometimes cross-border basis. The fact that these three types of social capital do not necessarily operate on a specific continuum will certainly have an impact on the analysis and description of levels of social capital within geographic communities and communities of interest for this research. While social capital has merit in detailing the relationships between individuals and groups, it is not a robust framework for explaining all the relevant conditions and characteristics of a particular community. However, when placed within a broader construct, the concept of social capital can add real detail to a picture that might previously only have taken into account economic, environmental and other purely statistical factors. Focused on weak community infrastructure and community tensions, the Communities in Transition Programme has broken these down into four constituent parts in order to look at the overall “health and welfare” of a community: Social Need, Social Inclusion, Social Cohesion and Community Capacity. In this context, ‘social need’ refers to the quality of life in an area, including factors such as levels of deprivation, environmental and physical issues and health. ‘Social inclusion’ (or exclusion) refers to whether all individuals within a community have access to the same opportunities and resources. ‘Social cohesion’ involves the building of shared values within a community and it includes the three components of social capital; it also refers to the way in which people have a shared sense of belonging within their community. ‘Community capacity’ is a measure of the structure and organisation within a community, as well as the community’s ability to attract resources, organise needs-based projects, work with other groups/communities and generally be representative of the area as a whole. The Royal Irish Academy project (see below) has adapted this model and the present research project will continue to adapt the two models to incorporate elements which deal specifically with peacebuilding and the Northern Irish peace process. For the Communities in Transition Programme, the point of developing the community tension typology and the four-pronged framework above is to be able to identify the causes, characteristics and impacts of these various forms of community tension and community infrastructure in order to tailor an appropriate approach to community development and peacebuilding for a particular local community. For this research project, the point of developing the typology and
15
the adapted multidimensional model will be to examine the relationship between weak community infrastructure, low social capital, social inclusion/exclusion and their impact on peacebuilding in a wide range of geographic communities and communities of interest across Northern Ireland. The typology and the model will be a reference point for the interviews and focus groups and will serve as a filter through which to develop additional domains and conduct further interrogation of the dataset constructed as part of the RIA research project.
Mapping Social Capital: A Model for Investment8
With funding from the Department for Social Development, the Policy Reference Group on Weak Community Infrastructure9 commissioned Mike Morrissey and Pat McGinn to carry out a pilot study to develop a methodology for the measurement of social capital in 2004. The idea behind this research was to arrange for groups of people (with extensive community development background and experience, as well as a detailed knowledge of specific local areas) to ‘score’ the level of social capital within wards across Northern Ireland. These would be regarded as proxies for the actual level of social capital. The scores were: • • • 1 – low social capital, weak community infrastructure; 2 – the existence of bonding social capital; 3 – the existence of bonding, bridging and linking social capital.
The purpose of the project was to search for readily available variables (from the Census, NINIS, Noble, etc) that might predict the ward level social capital scores. Regression analysis was used over a number of datasets and four regression models. In short, none of the models accounted for a sufficient proportion of the variance to suggest that this variable set could account for the social capital scores. However, this result could suggest that social capital cannot be captured by this kind of data or that the way in which scores were assembled was a flawed exercise. High levels of deprivation and high levels of residential
Much of this section is derived from Mapping Social Capital: A Model for Investment, Paper presented at a Royal Irish Academy research conference, November 2005. 9 The Communities in Transition Programme and Rural Community Network’s Low Infrastructure Programme are jointly involved in the Policy Reference Group on Weak Community Infrastructure. The PRG is chaired by Dr Jeremy Harbison and involves key stakeholders, policy makers and academics who can play an integral part in ensuring that there is an effective link between these two programmes, policy making processes and the potential for wider impact. To date the PRG has been responsible for overseeing four key research projects.
8
16
segregation were found to be the best predictors of a high bonding social capital score. Bridging social capital was found to be negatively correlated to bonding social capital and linking social capital did not seem to be related to either of the other two components.10 From this analysis, even in the most favourable cases (Belfast and Derry, which are certainly not representative of the whole region), it did not prove possible to use these variables to predict the overall level of social capital. More than fifty per cent would remain unexplained. Nevertheless, it should be said that in real world situations, being able to use regularly collected statistics to predict as much as half of the variance might be a useful tool.11 This research was exploratory with many limitations; however, the researchers felt that, with more time and resources, a more sophisticated exercise could be attempted and this led to the Communities in Transition Programme and Community Evaluation Northern Ireland jointly being awarded a grant from the Royal Irish Academy to further develop the research. A team of researchers who had been involved in the original project was put together and the group successfully applied for a grant from the RIA TSRP funding programme in 2004. The team comprised Dr Jeremy Harbison, chair of the Policy Reference Group on Weak Community Infrastructure (PRG) which had commissioned the original research; Mike Morrissey, a consultant with CENI who had carried out the original research; Brendan McDonnell, Director of Community Evaluation Northern Ireland (CENI) who had conducted a Scoping Study on Weak Community Infrastructure for the PRG; Joanne Kelly, Researcher with CENI; and the present author. The research team agreed that there was a definite need to expand on the original research in two main ways: 1) To expand the scoring system that was used in the first project so that the three types of social capital could be more clearly defined in the scores and to expand the numbers of people who carried out the scoring for each area; and 2) To expand upon the dataset that was constructed from regularly collected administrative and other data so that it more accurately reflected two of the domains, namely Community Capability and Community Capacity. The team also felt that, while it was interesting and useful to attempt to devise a predictive statistical model for the social capital scores, it was also very useful to construct this dataset in its own right. In other words, whereas the original research was very experimental in nature and used only a
10
Morrissey and McGinn, “Research to Identify Key Indicators to Measure Social Capital in Neighbourhoods”, DSD (2004). 11 Ibid.
17
limited amount of data (primarily from the Census), the new research project collected data from a number of wider sources and assembled the data into a dataset based on the four domains (Social Need, Social Capital, Community Capacity and Community Capability). At the time the research was being conducted no such collated database existed in Northern Ireland and the database itself was of value, regardless of whether the experimental part of the research (the predictive model) was found to be effective. The research initially sought to answer two fundamental questions: Do communities with weak community infrastructure also have low levels of social capital and, in turn, does this translate as forms of deprivation, anti-social behaviour, etc.? If so, can such relationships be measured or are both ideas conceptually flawed and of little value in policy terms?12 Since data was available, albeit in varying amounts and levels of quality/reliability, for three of the domains (Social Need, Community Capability and Community Capacity), although no real spatial measure of social capital was available (exceptions being the data from the NI Life and Times Survey and the Continuous Household Survey, see below), the research team decided to build on the learning from the previous project and again collect original scores on social capital across Northern Ireland. The first stage of the research was to refine the scoring system used for measuring social capital. It was agreed that, rather than using a single component score for each area as in the last research project, this time scorers would assign individual scores for each of the three types of social capital (bonding, bridging and linking). In addition, it was felt that most people would place different values on these individual scores, e.g. someone might feel that bridging social capital (in the form of cross community relations) was of more importance in a local context than bonding social capital (in the form of intracommunity relations) or vice versa. For this reason, the researchers decided to ask scorers to rate the overall level of social capital in an area based on their own individual opinion of which of the three types was the most important, rather than relying on a cumulative overall social capital score. This was done by asking scorers the question, “Would you want to live here?” in relation to the area’s levels of overall social capital. Furthermore, the researchers decided that, while they had the opportunity of having scorers assembled, they should also ask them to score Community Capability and Community Capacity. This was done in order to verify and expand upon the data collected for these two domains, for which less data was available than for Social Need.
12
Mapping Social Capital: A Model for Investment, Paper presented by the author at a Royal Irish Academy research conference, November 2005.
18
In the previous social capital project, scores had been assigned to each area by individuals. In addition to allocating a numeric score to each ward, these individuals also submitted notes on why they had scored the areas accordingly; thus allowing for the collection of a limited amount of qualitative data. For the RIA research project, the team decided to use panels of scorers over a series of sessions, in order to allow for discussion and agreement on the numeric scores assigned to geographical areas, as well as to minute these discussions so that more in-depth qualitative data would be collected to explain these scores. Panels of scorers were made up of CFNI and CENI staff members with particularly detailed knowledge of specific areas, local community development representatives and representatives of statutory agencies and funding bodies (e.g. Councils, Local Strategy Partnerships, Housing Executive, network organisations and, in one instance, the PSNI). Panels were assembled in the District Council areas in which CFNI and CENI had the most contacts and this led to the formation of scoring panels for twelve District Council areas: Antrim, Armagh, Ballymena, Belfast, Carrickfergus, Coleraine, Derry, Larne, Magherafelt, Moyle, Newry and Mourne and Strabane. Because the areas covered by Belfast and Derry were felt to be too large, these areas were broken up into smaller areas covered by separate panels: one panel each for North, South, East and West Belfast and one each for the Waterside and Cityside areas of Derry. Each panel consisted of at least five scorers (at least three of whom were independent from CFNI and CENI) and at most twenty-one scorers. Nominal Group Technique (NGT) was the process used for scoring each geographical area. This technique, from the Treasury’s Magenta Book, seeks to tap into local stakeholders’ knowledge about their communities based on their extensive experience. All the scoring panel members were briefed on the background and purpose of the research and they were given definitions of the three types of social capital. They were informed of the scaling to be used and asked to look at their area and assign a score each from one (very low) to five (very high) for bonding, bridging and linking social capital, as well as the overall level of social capital in the area based on the question, “Would you want to live here?” Whereas the previous research had generated social capital scores at ward level, this project worked at the level of ‘Super Output Area’, designed for the new Northern Ireland Multiple Deprivation Measure (see below). This had the advantage of being smaller than ward level, and thus to generate scores and qualitative data for more specific geographic communities. Although as administrative entities Super Output Areas are still not exact descriptions of actual communities, they do more closely approximate “real” neighbourhoods than wards do. The primary advantage of using Super Output Areas is that deprivation scores have now been presented at this level in the new Noble index. Moreover, the Northern Ireland Statistics and Research Agency (NISRA) agreed 19
to transform the spatial co-ordinates of other datasets to Super Output Area level in order to produce a uniform spatial format. Maps of the Super Output Areas for each of the District Council areas involved in the exercise were produced and made available for the Nominal Group sessions. Panel members were first asked to score all the Super Output Areas individually and then to discuss the scores as a group to see if a consensus could emerge. Finally, stakeholders were also asked to assign scores (again between one and five) to Community Capacity (defined as the number and density of community organisations in an area) and to their overall level of Capability (“How able are they at drawing in resources, carrying out programmes, being representative and cooperating?”). Scorers were advised that these two variables were not necessarily related. In other words, a community might have a large number of community and voluntary organisations (i.e. score a 5 for Capacity), but these groups might not be at all capable of managing resources, bringing in funding, working together or representing the community as a whole (i.e. they might score a 1 or 2 for Capability). Conversely, an area might have very few groups (scoring a 2 for Capacity), but the groups that the area had might be very capable at all of the above functions (thus scoring a 4 or 5 for Capability). It was realised that these scores would be weak, but their purpose was to compare these with other kinds of data. Qualitative data relating to Social Capital, Community Capacity and Community Capability was also collected via discussions with those stakeholders involved in the scoring process throughout the research process and this has been used to add further explanatory value and detail to the findings.13 Scoring using the Nominal Group Technique involved three stages: First, panel members were asked to assign their own individual scores to each Super Output Area without discussing these scores or the reasons behind them with any other member of the panel. After each person had finished assigning scores to all of the SOAs, participants were invited to share their scores for each area and discuss why they had assigned these scores. After the panel had finished discussing these individual scores, the facilitator collected each score sheet. The panel was then asked to score the SOAs again, making any changes they felt were necessary based on the previous discussion. This second set of scores was then discussed, with a focus on changes that anyone had made. Individual score sheets were collected again. Finally, the group was asked to reach a consensus on one score for each type of social capital for each SOA. While the discussion that led to this single set of scores was often lengthy and sometimes rather heated, there were only a handful of (fewer than five) cases for all of the 447 SOAs scored in which a member of one of the panels refused to agree with the group’s overall scores. In addition to the comments on individual score sheets, all stages of the group discussions were minuted and these comments and minutes provided valuable qualitative data for each area.
13
Mapping Social Capital: A Model for Investment, Paper presented at a Royal Irish Academy research conference, November 2005.
20
Using this process Social Capital scores were collected for just over half (447 out of 890) of the Super Output Areas in Northern Ireland. This included complete sets of scores (i.e. scores for each of the four Social Capital variables for every Super Output Area within a District Council) for the twelve District Council areas listed above. Incomplete sets of scores were also generated for a few areas (namely, Castlereagh, Omagh, Fermanagh and Craigavon), but these scores were not included in the database since they either did not cover the full area, did not have at least five scorers for the area or did not include scores for each of the Social Capital variables. In terms of constructing the database, Social Need indicators were derived from the 2001 Census data at Super Output Area level and from the 2005 Measure of Multiple Deprivation. Data on the number of community-based organisations and their funding incomes were taken from the NI Council for Voluntary Action State of the Sector survey and the Voluntary and Community Sector database at the Department of Social Development. A final data set was constructed that consisted of: • Fifty-four Census variables (based on the percentage of the relevant population of the Super Output Area) that related to demographic and socio-economic characteristics (including ‘community background’ attributed religion); Six variables from the Northern Ireland Measure of Multiple Deprivation; Four social capital variables (bonding, bridging, linking and overall social capital) (generated by Nominal Group Technique); Two variables (also generated by Nominal Group Technique) on community capacity and community capability; A variable on the number of community-based organisations per 1,000 population, constructed from the Department of Social Development database.14
• • • •
Utilising these data, statistical techniques were used to identify those variables that best predicted the level of social capital and weak community infrastructure (as indicated by a low density of community-based organisations). Analysis of the dataset suggested that: • The associations amongst the four social capital variables did not indicate an internally coherent concept. For example, there was an inverse correlation between bonding and bridging social capital. Bridging and linking social capital correlated more closely with the overall social capital
14
The database contained postcodes for just over 5,000 organisations. These were translated by NISRA into Super Output Area locations. 4884 postcodes were successfully translated.
21
score than bonding did and this may indicate that people tended to place more value on these two types of social capital than on bonding • Variables relating to the highest level of qualification in the Super Output Area significantly correlated with both the bonding and bridging variables, but in different ways. Areas with high percentages of residents with no qualifications correlated with bonding, whereas areas with high percentages of residents with level four and level five qualifications correlated significantly with bridging; The bonding variable correlated positively with high unemployment rates, the bridging variable with low unemployment rates; Similarly, the bonding variable correlated positively with high Multiple Deprivation scores and the bridging with low Multiple Deprivation scores.15
• •
Such findings point to the possibility of a social class gradient in the Social Capital domains, in particular between bonding and bridging. There was also an indication that levels of bonding social capital are not necessarily associated with bridging. Curiously, the linking domain was more ‘neutral’ to social class differences. Of some interest in Northern Ireland is the relationship between levels of social capital and community background. Analysis of this data set pointed to small, but significant differences in the relationships between Protestant and Catholic community backgrounds respectively and levels of social capital – indeed, there was an inverse relationship between Protestant community background and all four social capital scores. Since this finding tended to contradict previous research (see, for example, Cairns et al 2003), the relationship between the community background and social capital variables was explored further – i.e. were the differences observed on community background a reflection of other variables? To test this, differences by community background were explored for occupation, unemployment and highest level of qualification – significant differences were found between areas characterised as high Protestant or Catholic community background and unemployment, though not occupation nor highest level of qualification. A new variable (residential segregation) was constructed with five values: • • • • •
15
40 – 60 per cent of the population of either Protestant or Catholic community background; 75- 90 per cent Catholic community background; Greater than 90 per cent Catholic community background; 75-90 per cent Protestant community background; Greater than 90 per cent Protestant community background.
Ibid.
22
Higher scores for the bonding social capital variable were associated with higher levels of residential segregation. The converse was true for bridging social capital. It is thus possible that the degree of concentration is as important as the actual community background of the area.16 One of the central concerns of the team was that using Nominal Group Technique would generate a set of scores that was more reflective of the mindset of the stakeholder participants than of the communities they were attempting to score. To explore this, some checks were undertaken using part of the Continuous Household Survey 2003/04 dataset that included a number of questions related to social capital. From this set of questions it was difficult to identify one that was fully representative of bridging social capital. However, two questions were chosen as reflecting bonding (“How much would you agree that is area is a close, tight-knit community?”) and linking (“Do you agree that by working together, people can influence decisions affecting the area?”). In the CHS the “close, tight-knit community” variable showed significant differences by: highest educational qualification; occupation; car ownership; and religion. While the “influence” variable showed significant differences by: highest educational qualification; occupation; and car ownership. Significant differences on both variables appeared by religion.17 This project set out to explore a model of community infrastructure using a four variable model, mindful that the term community infrastructure and the concept of social capital are both contested. The central challenge was to see if the model could be populated from existing datasets and whether an indicator of social capital could be derived without the expense of multiple local surveys. While the effort to find appropriate secondary data was less than completely successful, substantial relevant amounts of data were obtained. Also, the use of the Nominal Group Technique did offer a relatively low cost method of developing social capital indicators (as well as community capacity and community capability indicators) for small spatial areas and checks using CHS data suggested that the results were not completely arbitrary. The relationships between the variables did raise some questions about the coherence of the community infrastructure construct. In particular, the bonding and bridging variables correlated inversely with each other and differently with important social variables, raising questions about whether the Social Capital domains do, collectively, form a single concept. However, as mentioned above, the Communities in Transition Programme has attempted to explain this phenomenon through the recognition that the three components of social capital do not operate on a continuous spectrum within Northern Ireland.
16 17
Ibid. Ibid.
23
In other words, the particular brands of community tension in operation within local communities here quite often do mean that an area where bonding social capital is particularly high will be less likely to build strong bridging social capital. This poses more questions than just challenges to the theoretical construct of social capital and its applicability to Northern Ireland. For example, it could call into question the value of specific types of single identity work in Northern Ireland and whether the further strengthening of bonding social capital within a community might, in effect, limit the ability of this community to engage in cross community (bridging social capital) development. The interrelation between the various types of social capital and other social variables may also lead to questions about whether funding and development support might be better targeted at the causes and effects of these social variables (e.g. education) rather than specifically focused on community development and peacebuilding work. This current research project will attempt to further examine the relationships between these variables (both quantitatively through further interrogation of the dataset and qualitatively through the interviews and focus groups) in order to shed more light on some of these questions as well as the value of the concept of social capital in a Northern Ireland context.
Initial Hypotheses
Using the Communities in Transition Programme and the work of the RIA project as a starting point provides a distinct advantage to carrying out research to explain the interrelationship between weak community infrastructure, low social capital, community capacity, community capability, community tension, social need and peacebuilding. These two projects have used slightly different models and very different methodologies to explain similar phenomena and answer a series of related questions. This research project will build upon the two models and methodologies in order to broaden their scope of explanatory powers in a Northern Ireland context. What follows is an outline of a few initial hypotheses formulated prior to conducting the local interviews, against each of the main objectives of the research: 1. The Relationship between Weak Community Infrastructure and Low Social Capital: It seems unlikely that there will prove to be a direct correlation between weak community infrastructure and low social capital as a cumulative concept (i.e. bonding + bridging + linking). However, it seems very likely that areas of weak community infrastructure (using the CIT definition on pages 6-7 above) will be found to have, on the whole, considerably lower levels of bridging social capital and lower levels of
24
linking social capital than do areas of strong community infrastructure. It may also follow that areas of weak community infrastructure have higher levels of bonding social capital and that, in many instances, this actually precludes the development of not only bridging and linking social capital, but also of community infrastructure. However, if this is found to be the case, it may well only be true in areas of deprivation and degrees of residential segregation will likely play an additional part. 2. Social Capital, Weak Community Infrastructure and Deprivation: It does not seem likely that any significant correlation will be found between these three concepts as a whole. In other words, there are affluent geographic areas with weak community infrastructure and low social capital, just as there are deprived areas. However, the community tension typology used by the CIT Programme should go some way towards explaining the causes and effects of this. Affluent areas are more likely to have low levels of social capital when the “dormitory town” phenomenon is in place and their community tension is more apt to manifest itself in the form of “latent” tension. If this is found to be the case, it will be additional evidence in favour of combining the community tension typology with the multidimensional model (Social Capital, Social Need, Community Capacity, Community Capability and Peacebuilding), since merely examining (and ranking) each of the model’s components will not explain their causes and effects. 3. Weak Community Infrastructure, Low Social Capital, Deprivation and Peacebuilding: It does not seem probable that there will be a correlation between most of these components; however, specific communities of interest may be found to have an added political dimension. That is to say, successful community-based peacebuilding initiatives will only be found to occur in areas where community infrastructure is being developed (such as the Communities in Transition projects where no peacebuilding work was being carried out prior to the initiation of the local projects, but several successful initiatives are now being undertaken). The relationship between social capital and peacebuilding will depend on the type of social capital. It seems obvious, but is perhaps worth stating, that only single identity work would be possible in an area with strong bonding and low bridging social capital. The fact that there does not appear to be a specific continuum between the three types of social capital may mean that levels of linking social capital are not necessarily correlated to the occurrence of successful local peacebuilding initiatives. In terms of deprivation, there should not be any significant correlation, but
25
if there is, it will be more likely that peacebuilding initiatives will be found to occur more frequently (and with higher success rates) within and between working class communities than affluent ones – this is perhaps a stereotyped assumption based on an implicit supposition that there is less recognition of the need for peacebuilding work to take place within affluent and middle class communities which have been more characterised by latent tensions than the other types of tensions and have perhaps felt less of the fear and experienced less of the very real effects of the ‘Troubles’. In terms of certain communities of interest, for specific groups (namely dissident Republicans and certain brands of Loyalism) it is probable that political factors will be more responsible for the failure to engage with peacebuilding initiatives, and the peace process in general, than will issues relating to social capital, community infrastructure and/or deprivation. For other marginalised communities of interest (for example, ethnic minorities), the opportunity to engage with peacebuilding will not have arisen yet and this may or may not be related to social capital, community infrastructure and/or deprivation, although it seems more likely that communities with strong infrastructure and high levels of (at least bridging) social capital would be more likely to engage with this type of peacebuilding work. 4. Community Capacity and Community Capability: As stated above, there will most likely not be any necessary correlation between these two concepts – some areas may have a very large number of groups, but very little capability and others may have a high degree of capability, but few groups. Furthermore, these two concepts may prove more useful as a tool for measurement than for explanation. While it is necessary to measure the number of groups within a geographic community (or, for that matter, a community of interest) and how able they are at bringing in funding and resources, working in partnership, managing projects and representing their community in order to paint an accurate picture of that community, it is possible that neither of these concepts will have as great an impact on describing that community (and its engagement with peacebuilding) as will social capital and community tension, for example. Furthermore, based on the findings from the RIA project, these two concepts will be the most difficult to gather reliable data on for a number of reasons: In terms of community capacity, there is a great difficulty of “comparing like with like” when scoring an area – do small, local community development groups count the same as large network organisations? What consideration is given to sector specific organisations, such as youth groups, church groups and sporting organisations? How are we to measure the area which a particular 26
organisation services? In other words, just because an organisation’s headquarters is located within a particular Super Output Area, this does not mean that those that use or benefit from the organisation are located within that Super Output Area. In terms of community capability, this is the variable for which the least data exists and for which it is perhaps the most difficult to collect original data. Some data exists in terms of the amount of money organisations handle (see above), but most of this data is based on annual income and this may not be an accurate reflection. No database exists which details all the funding groups across Northern Ireland have drawn down, for example, never mind how much they have (unsuccessfully) applied for. There is no source to tell us, furthermore, how representative groups are of their area/community, nor how well they work in partnership with other organisations and communities and these are, perhaps, variables that should not even be quantified in the first place. Nevertheless, some account must be taken of both community capacity and community capability.
Overview of the Project
Section 2 below, summarises some of the primary and secondary literature on two of the model’s main concepts, deprivation and social capital. While much literature exists on the theoretical debate over using the concepts poverty, deprivation or social exclusion, there is almost no secondary literature on the evolution of the means for measuring deprivation in Northern Ireland. The author was able to find only one secondary source (a research paper written for the NI Assembly) and it only compares Robson with the 2001 Noble Index (omitting Townsend and being written prior to the 2005 Noble update). For this reason, a large portion of this section traces the changes made to the main recognised means for measuring deprivation in Northern Ireland from the 1980s to the present. It also provides a brief outline of how the concept of social capital has become increasingly popular and influential, not only within Northern Ireland, over the past half century. This section looks at the concept in general, whereas Section 3 looks at different approaches to studying and measuring social capital in Northern Ireland specifically. The development of the model to be used throughout this research is discussed in Section 3 and placed in the context of other relevant research (both quantitative and qualitative) that is being conducted in Northern Ireland. This section also includes a further discussion about the conceptual relationship between social exclusion and social capital in Northern Ireland and explains the 27
reasons for including community capacity and community capability in the model. It also pins down the definition of peacebuilding that will be used in this research and summarises current peacebuilding work and theories within Northern Ireland. It concludes with a brief synopsis of comparative international research. The fourth section is divided into two halves: one on policy and one on politics. The policy section contains an abstract of relevant policy initiatives and programmes that are currently operating or being developed within Northern Ireland. The politics section looks at the recent history of the four main political parties, different paramilitary organisations and specific victims’ groups and their positions in relation to peacebuilding in general and the peace process specifically. A further description of the two methodologies is given in Section 5 and Sections 6 and 7 present the findings from the two approaches. Section 6 contains the findings from further analysis of the dataset and Section 7 presents the findings from the interviews and focus groups. These are the most important sections of the thesis. Finally, Sections 8, 9 and 10 give further details about the outcomes, implications and conclusions of the research project.
Methodology
The primary purpose of this research project is to examine the relationship between weak community infrastructure, low social capital, social inclusion/exclusion and their impact on peacebuilding in a wide range of geographic communities and communities of interest across Northern Ireland. This will be achieved through the use of two distinct, but complementary, methodologies: 1. The development and analysis of a database constructed from regularly collected administrative data and other data specific to the Voluntary and Community Sectors across three Domains (Social Need, Community Capacity and Community Capability) and original data under a fourth Domain (Social Capital). 2. The collection of qualitative data through a series of interviews, focus group discussions and seminars with members of geographic communities and communities of interest, as well as policy makers and academics working in this field. Due to the availability of data, the model developed will operate on a number of levels, beginning with Northern Ireland-wide analysis and then looking
28
at a number of specific geographic communities and communities of interest for which more detailed data has been collected.
The Model, Database and Sources
The model will be constructed from four distinct Domains: 1) Social Need; 2) Community Capacity; 3) Community Capability; and 4) Social Capital. Each of these Domains is made up of a number of indicators, derived from a range of sources. Only the first three of these Domains have associated data that is regularly collected (and available) on a Northern Ireland-wide basis. The first step in developing the database will be to compile and aggregate data on these three Domains for each of the 890 Super Output Areas in Northern Ireland. Because all of this data is now available for Super Output Areas,18 SOAs will be the geographical unit for analysis, as they more closely approximate “real” neighbourhoods than do Wards. The District Council within which each SOA is located will also be included in the database, in order to enable more general analysis. Postcodes will also be included for data, where available, in case there is a future revision of the boundaries or geographical units regularly used in Northern Ireland.19 The Northern Ireland-Wide Database Social Need. Each Super Output Area in Northern Ireland, along with its District Council area, will be entered into an MS Excel file and a corresponding SPSS file. Information used for each of the individual Noble Domains will be included for each Super Output Area, along with other relevant variables from the Census. This information is already publicly available from the Northern Ireland Neighbourhood Information Service; it must just be adapted slightly to fit the purposes of this research. Rather than using separate subsets for each individual Domain of Noble and then one composite dataset for the Multiple Deprivation ratings and rankings, this database will include a collated version of all the Domains and the Multiple Deprivation Measure for each SOA. This will enable cross-tabulations and other analyses to be run between each of the Domains and other variables across the SOAs. Furthermore, the data available is provided by NINIS in MS Excel, which enables easy sorting and a number of other basic tasks; however, putting a copy of the complete database into SPSS
18
Census data and data from the Noble Measures of Multiple Deprivation are provided for each Super Output Area through the NISRA website. Data from the DSD and NICVA were provided with postcodes and then an area generator was used to place each postcode within a Super Output Area and District Council Area. This required a considerable amount of data cleaning and checks for accuracy after the areas were generated due to a fault in the coding programme (provided by NISRA). The data has now been cleaned and verified and NISRA and the DSD will be notified of the inaccuracies in their original dataset. 19 This will be particularly relevant to the development of the new Local Government Boundaries under the Review of Public Administration. It will also allow the SOAs to be translated back into Electoral Wards, if and where, appropriate.
29
will also allow for more sophisticated statistical analysis. The Excel format also allows for relatively simple “cutting and pasting” in order to add data from other sources to the database; however, where it is necessary to perform tasks such as translating postcodes into SOAs or aggregating data for each SOA/District Council, an intermediary file in MS Access will be constructed in order to use/develop programmes to carry out these functions. The data will then be moved back into Excel and SPSS for analysis. Community Capacity. The Department for Social Development has collected raw data from its own and other (including Local Authority and other Departments’) programmes, as well as from the Northern Ireland Council for Voluntary Action’s most recent State of the Sector Survey and has made this data available to the research team involved in the Royal Irish Academy project. Two indicators will be constructed from this data for each SOA in Northern Ireland: 1) Community Capability and 2) Community Capacity. The DSD data includes the postcodes of all known active community groups. These postcodes will be translated into SOAs and the number of groups within each area will then be aggregated for each SOA. This will provide an indicator for the total number of known, active groups within each SOA (and District Council) – a very basic indicator of Community Capacity. (An alternative indicator for Community Capacity is presented in the next section.) At present the DSD/NICVA dataset provides the most complete set of indicators for the number of active groups within a given area. However, there is an issue of weighting which needs to be addressed: namely, within the DSD/NICVA data, organisations and groups are simply counted and no distinction is made between different sizes and types of organisations. In other words, the headquarters of a large agency such as the NI Housing Executive within an area would be counted as one organisation, equivalent in score to a small community-based residents’ group located in another area. Furthermore, no distinction is made between groups that are single interest in focus (such as youth groups, church groups, women’s groups, sporting organisations, Loyal Orders, etc) and organisations with a wider range and membership. This latter issue is not something that can be resolved within the Northern Ireland-wide database constructed for this research project; although it is an issue which has now been raised by the RIA research team with the DSD, NISRA and the NICVA team responsible for the State of the Sector Survey and will hopefully be resolved in future Surveys through the inclusion of a variable in the database which labels the type of community/voluntary group. The former issue – the problem of scale – should be at least partially resolved through the construction of an indicator for Community Capability. Community Capability. The DSD has also provided data on the annual income for each of the groups in its dataset. This data does not give actual amounts of annual income, but a ranking of 1 - 8 (8 being the largest annual income) for each group. These individual scores will be aggregated at SOA level
30
in order to illustrate the differences in total amounts of income for all groups in one area compared with another. In other words, an SOA with a score of 1 would indicate that only one group in an area had a modest annual income, whereas if an area had ten organisations, all with very large incomes, it would score 80. In this way, not only will the indicator provide a simplistic picture of Community Capability, it will also go some way towards solving the problem of scale mentioned above. For instance, in the example given above, the headquarters of the large agency might score an 8, while the small community group might score a 1. Providing an aggregate score for Community Capability, but distinct from Community Capacity, also enables the issue of fragmented community infrastructure to be taken into account: i.e., if there is a large number of groups in an area but the aggregated annual income for the area is relatively low, this will mostly likely indicate fragmented community infrastructure (and capability). Analysing the Database. Structuring the data available at a Northern Ireland-wide level in this way will allow for a number of interrogations. Crosstabulations/correlations can be carried out between all the different indicators and Domains at the geographic level of SOAs and District Councils. Regression analysis and other simple statistical tests can also be conducted to further explain these relationships. This will provide some answers to the initial question of how community infrastructure, in general, and community capacity and community capability, in particular, are influenced by social need, if at all, as well as providing a geographic “map” of these concepts across Northern Ireland. In other words, basic analysis will begin to explore the relationship between these concepts across Northern Ireland, but the database will also have the added utility of being able to rank all 890 SOAs in Northern Ireland according to each of the Domains individually or any (weighted or unweighted) combination thereof. Similar to Noble in its capacity to rank areas in terms of the individual Domains or combined Measure of Multiple Deprivation, this utility will certainly be of interest to funders, policy makers and others working in the Statutory, Voluntary and Community sectors, as well as academics. This database has the advantage of containing a complete set of indicators for each of these three Domains for all 890 SOAs in Northern Ireland. However, it has the disadvantage of not including adequate detail to fully capture all the factors influencing community dynamics within an area and the relationships between all the Domains. As more detailed data is only available for a subset of the sample, it will be necessary to construct a number of additional databases which include more variables, details and the fourth Domain (Social Capital) but which only include data for some of the SOAs in Northern Ireland. Social Capital Database
31
As mentioned above, original data has been collected through the Royal Irish Academy funded research project to enable the construction of a Social Capital Domain. This data is available for 12 complete District Councils and 447 (just over half) of the Super Output Areas. The second database will, therefore, include all of the information in the database described above for these 447 SOAs/12 District Councils, as well as four indicators of Social Capital: • • Bonding Social Capital – How do local communities trust and relate to others like themselves (intra-community relations). Bridging Social Capital – Trust and relationships between individuals and groups who are in other communities or unlike each other (crosscommunity relations). Linking Social Capital – The quality of relations between communities and decision makers, those in positions of power (Local Government, service providers, funders, PSNI etc). Overall Social Capital – Based on the question ‘Would you want to live here?’
•
•
Each of these indicators was scored from 1 (very low) to 5 (very high) through use of the Nominal Group Technique, as described above. The ‘consensus scores’ agreed for each indicator for each SOA by the NGT panels will be added to this sub-database of 447 areas. This will allow for additional analyses between all four Domains across half of Northern Ireland. The sample includes complete sets of data for each SOA in Antrim, Armagh, Ballymena, Belfast, Carrickfergus, Coleraine, Derry, Larne, Magherafelt, Moyle, Newry and Mourne and Strabane. Finally, data collected through the Continuous Household Survey relating to Social Capital will be used to validate the scores collected for each of these areas. Alternative Community Capacity and Community Capability Database In addition to collecting scores on Social Capital, the NGT panels also generated scores on Community Capacity and Community Capability. During the scoring sessions, these two concepts were defined as: • Capacity – The number and density of community organisations in an area, accounting for a variety of different types of organisations (1 = none/very few, 5 = many, across a range of different purposes/interests). Capability – How able are these organisations at drawing in resources, carrying out programmes, being representative and cooperating? (1 = 32
•
not at all able or there are no organisations, 5 = very able at each of these things).
These scores were collected across 5 District Councils (Ballymena, Belfast, Derry, Magherafelt and Strabane) for 276 SOAs. A third database will be constructed which includes all of the above information for these 276 SOAs, as well as these separate scores for Community Capacity and Community Capability. This will enable not only the further development of the model for exploring the relationship between the Domains for these 276 SOAs; but will also serve to validate the use of the ‘simpler’ measures of Community Capacity and Community Capability from the DSD’s data. In other words, if the scores collected from local experts on how many/what types of groups are in an area are directly reflected in the DSD’s information, then the Northern Ireland-wide sample with the ‘simple’ indicator will prove robust enough and it will be possible to use the larger sample for any measure of Community Capacity. The same holds true for the relationship between the NGT scores based on a broader definition of Community Capability: if these scores directly correlate to the DSD’s data on levels of annual income, then the larger dataset will be used for any measure of Community Capability. In addition to collecting numeric scores for these six indicators (Bonding, Bridging, Linking and Overall Social Capital, Community Capacity and Community Capability), qualitative information was also gathered through the NGT panel sessions. Each individual scorer provided notes on why s/he had rated each element/area as s/he did and all of the group discussions through which the panels arrived at consensus scores were minuted. The use of this qualitative data will provide further explanations for the relationships between the Domains, as well as potential causes for low (or high) scores within an area. It will also shed light on possible causes and effects of community tension within the areas and the relationship between community tension and the four Domains (see below). Finally, as above with the Northern Ireland-wide sample, these two additional databases will have the added value of being able to rank their constituent areas in terms of Social Capital, Community Capability and Community Capacity or any combination of all of these Domains. If the ‘simple’ Capacity and Capability indicators are found to be accurate, this means that a set of just over half of Northern Ireland’s 890 SOAs could be ranked according to all four Domains. Of further use to funders, policy makers etc, the qualitative information will serve to back up the rationale for these rankings. Community Tension Database and Case Studies
33
The final database will contain a much smaller spatial sample (only 25 SOAs), but in much more detail. This data comes from survey work carried out through the Communities in Transition Programme across 21 SOAs and a very similar survey conducted by the Rural Community Network’s Low Infrastructure Programme across 4 SOAs. The present author was responsible for designing and analysing the Communities in Transition surveys and worked closely with RCN in the design and analysis of their survey. RCN provided the raw data generated from their survey for inclusion in this research project. The Communities in Transition survey included 4,497 respondents from 21 SOAs and the total rate of return was 31.64%, so it was a relatively large sample. A number of indicators will be generated from this survey. These do not necessarily fit under the four Domains outlined above, although they may be found to have some correlation to the scores which individual areas were assigned within some or all of the Domains. Many of the indicators are more closely related to Community Cohesion or levels of Community Tension. For this reason, the indicators will be tested in terms of how they correlate to the four Domains (Social Capital in particular), but many of them may have further relevance to the final element of the model – Peacebuilding. These indicators include: • The length of time people had lived in their local area. This indicator clearly describes an element of Social Cohesion, but may also be found to relate to Social Capital, as well as Community Capability and Community Capacity. For example, areas with a large amount of recent newcomers will probably be found to have lower Community Capacity and Community Capability than areas with relatively settled, long established populations; however, this hypothesis will be tested for validity across all 25 SOAs. Levels of participation in community groups and activities. This also relates to Social Capital, Community Capacity and Community Capability and the relationship between this indicator and the NGT scores for each of these Domains will also be tested. Intuitively, it seems that very high levels of community participation should correlate to high scores for each of the Social Capital indicators as well as Community Capacity and Community Capability. This will be tested and the qualitative data from CIT surveys, as well as the NGT sessions, will be analysed to assess why this may or may not hold true. In addition to being asked whether they participated in community groups and activities, how regularly they participated and how many different groups/activities they were involved in,
•
34
respondents to the CIT survey were also asked to list all the groups to which they belonged. The types of groups listed will be analysed in order to ascertain the numbers and types of groups active in an area, a component of Community Capability detailed above. • Whether residents were aware of community groups and activities in their area. Respondents were asked to indicate their level of awareness of community activity and were also asked to select one of the following options: 1) I actively participate in community activities; 2) I am aware of community activities, but I choose not to participate; 3) I would participate in community activities if I were informed of them; 4) I would like to know more about community activities; 5) I do not wish to know about community activities. Levels of ‘community spirit’. Respondents were asked to rate the community spirit in their local area by selecting one of the following options: 1) We have a strong community spirit. 2) We have a good community spirit, but some people do not participate in the life of the local community. 3) We have a good community spirit, but most people do not participate in the life of the local community. 4) Only a few individuals or groups control everything in the life of the local community. 5) There is no sense of community spirit at all. Sense of belonging. Respondents were asked to select one of the following options: 1) I do feel a sense of belonging to my local community. 2) I do not feel a sense of belonging to my local community. Local influence and community tensions. Respondents were asked to consider a number of different groups, agencies and other factors and rate whether these had a positive effect on their local area; a negative effect on their local area; both positive and negative effects on their local area; or were not applicable to their local area. They were also asked to comment on each of these groups, which included the following: 1) Key local individuals; 2) Local community groups or organisations; 3) Local business; 4) Local political leaders or organisations; 5) Local church leaders or organisations; 6) Housing Executive; 7) Local Council; 8) Health services; 9) Police; 10) Schools; 11) Other agencies; 12) Paramilitaries; 13) Drugs; 14) Antisocial behaviour; 15) Graffiti; 16) Murals.
•
•
•
35
•
Primary local concerns. Respondents were asked to list their key concerns for their local area and community. This question was intentionally left open-ended, yet a surprising number of respondents listed the same concerns. These included: drugs, intimidation, antisocial behaviour, vandalism, lack of policing, paramilitaries, underage drinking and road safety.
This survey across 25 local areas contains a wealth of descriptive information, as well as a number of indicators which can be used to complement the four Domains and to describe additional factors contributing to community tension, conditions for peacebuilding and social cohesion. In addition, Social Capital, Community Capability and Community Capacity scores were collected for all 25 of these SOAs. For this reason, a database will be constructed which includes all of the data listed in the three preceding sections as well as the additional indicators drawn from the community surveys for these 25 Super Output Areas. This database will facilitate more extensive analyses of the four Domains and other factors contributing to social cohesion, community participation, community tension and peacebuilding. The relationship between each of these variables will be explored. In addition, a considerable amount of qualitative data has been collected over the past five years on each of these local communities, as they have been participants in the Communities in Transition Programme and the Low Infrastructure Programme. These areas were selected because of their low community infrastructure and (in the case of the CIT Programme areas) because of high levels of community tension. This will enable a qualitative examination of the factors necessary for the creation of community infrastructure over time and the impact that community tension has on this process. The qualitative information collected for these 25 SOAs will also serve to create a collection of Case Studies to further complement the quantitative dataset.
Database Summary In total, four separate databases will be constructed and analysed, beginning at the Northern Ireland meta-level and becoming more narrowly geographically focused while increasing in detail and types of information. The first database will include data on Social Exclusion, Community Capacity and Community Capability for all 890 Super Output Areas. The second database will include data on Social Exclusion, Community Capacity, Community Capability and Social Capital for 447 Super Output Areas. The third database will include data on these four Domains, plus an expanded set of scores on Community
36
Capacity and Community Capability for 276 SOAs. The fourth database will include all of the above information, plus a number of additional indicators of community tension and social cohesion collected from local area surveys for 25 Super Output Areas. The last three of these databases also have a large amount of corresponding qualitative information to contribute to their explanatory powers. Each database will be analysed separately in two ways: first, a few simple statistical tests will be applied to each (primarily cross-tabulations between all of the Domains/indicators and regression analysis) in order to establish any statistically significant correlations between the variables included. Second, each database will be “sorted” by the rankings of the different SOAs according to each Domain/indicator separately, according to a composite (unweighted combination) of all the Domains/indicators and a few experimental formulas and weighting systems will be applied to rank the areas. This list of rankings will be used as a tool to inform the panel discussions and interviews (see below) in which individuals and groups will be asked to comment on the model in general, each of its components and to assess the extent to which they agree with any of the ways in which the SOAs have been ranked through the use of the four databases. The diagrams below show the composition of the individual datasets and how each of the indicators will be used within the Four Domain Model.
rmation Collected through Interviews and Panel Sessions
Northern Ireland-Wide Dataset Covering: 890 Super Output Areas Includes: Noble Measure of Multiple Deprivation and Individual Noble Domains; Census Data; DSD Data on Number of Community Organisations; DSD Data on Annual Income for Organisations Continuous Household Survey Data on Social Capital
Social Capital Dataset Covering: 447 Super Output Areas Includes: Noble Measure of Multiple Deprivation and Individual Noble Domains; Census Data; DSD Data on Number of Community Organisations; DSD Data on Annual Income for Organisations; Original Scores on 4 Types of Social Capital (Bonding, Bridging, Linking and Overall Social Capital)
Community Capacity/Capability Dataset Covering: 276 Super Output Areas Includes: Noble Measure of Multiple Deprivation and Individual Noble Domains; Census Data; DSD Data on Number of Community Organisations; DSD Data on Annual Income for Organisations; Original Scores on 4 Types of Social Capital (Bonding, Bridging, Linking and Overall Social Capital);
37
Case Studies Covering: 25 Super Output Areas Includes: Noble Measure of Multiple Deprivation and Individual Noble Domains; Census Data; DSD Data on Number of Community Organisations; DSD Data on Annual Income for Organisations; Original Scores on 4 Types of Social Capital (Bonding, Bridging, Linking and Overall Social Capital); Original Scores for Community Capacity and Community Capability; Survey Data relating to Community Tension and Community Cohesion
The Four Domain Model and Indicators
Social Need Noble Measures Census Data
Community Capacity DSD Data on Number of Groups Original Scores
Peacebuilding Interviews
Community Capability DSD Data on Annual Income Original Scores
38
Social Capital Original Scores Continuous Household Survey Data
Interviews and Panel Sessions
A number of individual interviews and group seminars will be conducted in order to get feedback on the model used, test the validity of the ranking of Super Output Areas based on different combinations of data, further inform the qualitative information collected about specific areas and, most importantly, gather information and opinions on the Peacebuilding element of the model. Another main objective of the research project is to establish whether certain geographic communities or communities of interest are more likely to engage in locally based peacebuilding initiatives and in building Social Capital (in particular Bridging Social Capital), whether there are any conditions which limit the likelihood of success for these initiatives or, indeed, alienate members of the local community from participating or supporting them. The interviews will be the main method for investigating this research question. Individuals representing a wide range of political and other interests will be interviewed, including: the main political parties (DUP, UUP, Alliance Party, Women’s Coalition, SDLP and Sinn Fein); politically motivated ex-prisoners and/or individuals representing various organisations (LVF, UVF, UDA, PIRA,
39
INLA, Dissident Republicans and Unaligned Republicans) and representatives from a cross-section of victims’ groups. These interviews will have three main purposes: • To discuss the model and its four Domains, establishing whether the individual feels the Domains are capable of capturing an adequate picture of the community (geographic or ‘of interest’) s/he represents, as well as to determine if s/he feels any of the Domains should be given priority over the others. (E.g. Is Social Need more important than Social Capital in terms of the relative development of an area?) To discuss the specific impact of any of these elements on peacebuilding and attitudes to the Peace Process in general. (E.g. Are there any required local conditions in order for peacebuilding work to commence or to succeed? Does the politically divided nature of a community impact upon its levels of social exclusion and vice versa? Are any of the four Domains relevant to attitudes and perceptions towards the Peace Process generally for any particular groups? In other words, what is the relationship between community dynamics, peacebuilding and politics?) To discuss the scores given to specific geographical areas and the way these areas rank in relation to each other in order to ascertain how the individuals interviewed would have expected the areas to score/rank.
•
•
A series of pilot interviews will be conducted with four to five individuals in order to refine the model and the interview structure and content. Then a minimum of eighteen interviews will be held to cover each of the representative groups listed above. A further set of group seminars will be held with people who have a detailed knowledge about specific geographical communities. This will be carried out with groups from the Communities in Transition Programme areas for which more detailed data is available. These groups will be made up of people with a range of political, community, socio-economic and professional backgrounds (including local residents, members of community groups, special interest groups and representatives of the statutory sector), but who all share a detailed knowledge of a specific geographic area. Similar to the interviews, these group seminars will look at the model in general, as well as the four Domains and other indicators and the relationships between them. The seminars will also look at the way the geographic area concerned (in this case usually between two and four SOAs per group seminar) scored across all the Domains and indicators and how it ranks compared to other areas. In addition to validating (or disproving!) and refining the model and its components and scores, these seminars will also serve
40
to collect further qualitative data that may help to explain the causes and effects of these elements on a local community, as well as their impact on peacebuilding initiatives. Comparisons will be made between the interviews and the group seminars in order to determine whether specific communities of interest (or political groups) tend to have different views than those expressed by a wider crosssection of a local geographic community. Each of these elements will be written up and analysed and then a final refinement of the model and the databases will be carried out. Finally, at least one seminar will be held with a group of funders, policy makers and academics working in the field to present all of the findings and discuss the theoretical, policy and practice implications of the research. The conclusions drawn from this seminar will be included in the final thesis. Throughout the course of the project, the Research Steering Group20 will also continue to meet regularly to discuss the findings, implications and progress of the research and to offer theoretical and practical suggestions as the model and the methodology develop.
20
Prof. John Morison, Dr Jeremy Harbison, Dr Mike Morrissey, Brendan McDonnell, Monina O’Prey and Avila Kilmurray.
41