Community Capacity Building and Social Policy - What Can Be Achieved?

Sally Casswell
Alcohol & Public Health Research Unit
Rnanga, Wnanga, Hauora me te Paekaka
University of Auckland

Abstract

Considerable resource is currently being directed towards community initiatives in New Zealand in order to build capacity to contribute to improved social and health goals. These community initiatives differ in the extent to which they focus on processes of community development, such as alliance building and developing new organisations and leadership, and the extent to which such community development processes are part of a community action approach which is directed towards achievement of specific social goals.

These different initiatives share the need for additional resource aimed at building community capacity and the collaboration of a number of different sectors from within the community. The evaluation literature suggests both approaches have positive impacts in enhancing community processes and that, in the case of community action, where there are specific objectives measured, there is also some evidence of a positive impact.

Introduction

There is considerable enthusiasm for the funding of community initiatives intended to meet social policy outcomes, both internationally and in New Zealand. The Capacity Building for Maori programme, the Healthy Cities projects, Safer Community Councils and Stronger Communities Action Fund projects are all current examples in New Zealand. There have also been community initiatives funded on a number of particular topics (e.g. alcohol, injury prevention, nutrition and housing programmes).

This current emphasis on funding community capacity building can be explained by a number of interwoven strands in current thought. The postmodern scepticism about the role of experts in central planning and a focus on process rather than goals encourages funding of community development projects rather than a reliance on centralised social planning (Rosenau 1994).

Interest in the role of social capital and its relationship with social policy (Robinson 1997) encourages funding and evaluation of community initiatives seen as likely to enhance social capital (Robinson 1999). Another influence has been the neo-liberal legacy of suspicion over the role of central government in people's lives, opening up a gap into which communitarianism (Etzioni 1996) and related community initiatives have moved.

The emphasis on community as a site of action has not been without its critics, however. Community initiatives have been seen as a convenient panacea with a reputation for exercising a stabilising effect in society, concentrating attention on local-level planning at the expense of a recognition of broader social issues, in particular, power and control (Petersen 1994, Robertson and Minkler 1994).

At its worst, the notion of community participation and empowerment can be used to argue for greater reliance on voluntary organisations in order to allow a withdrawal of needed health and social services (Binney and Estes 1988).

Despite these cautions there is widespread acknowledgement that, unless there is capacity to identify and address social and health issues at the community level, central government's social policy initiatives will be ineffective. Many examples could be given. For instance, in the United States the response to evidence that educational achievement prevented long-term welfare dependency in unmarried teenage mothers was the 1996 legislation requiring unmarried teenage mothers to attend school and live with an adult in order to receive federal assistance.

It was accompanied by additional funding for childcare facilities. However, this partial response ignored the community-level changes that would make educational gains more likely - changes in school management styles and school policies, provision of alternative educational approaches, changed social norms and improved transport, for example (Poole 1997).

Another example closer to home was the change in the New Zealand Sale of Liquor Act in 1999. This reduced the minimum purchase age for alcohol and had the aim of better preventing access to those below the legal minimum purchase age by simplifying the legislation and specifying proof-of-age documents.

However, there was not adequate resourcing for community-level activity of enhanced enforcement or local organising to change social norms which encourage supply of alcohol to those under the minimum purchase age. Given this, the simplification of the law, availability of age-identification documents and the accompanying publicity have probably had only a partial and unsustained impact on supply to those under 18 years (Humphrey 2001, MacMaster 2001).

As these examples indicate, the community is the site of the mediating structures that intervene between the domain of everyday life of individuals and the larger social, political and economic context (McKnight 1987) and it is possible at the community level both to respond to opportunities created by changes in legislation and to facilitate the aims of social legislation.

The community initiatives discussed in this paper can be contrasted with a more centralised social planning approach to the implementation of social policy which offers less opportunity for community responsiveness and participation.

This paper draws on experiences of community action research over two decades. First a contrast is made between two different approaches to community capacity-building initiatives: community action and community development. It is argued that, although in practice there are overlaps in the processes involved, it is useful to draw conceptual distinctions between them. Second, the research literature on the impact of different community initiatives is reviewed. Third, some implications are drawn for New Zealand practice in funding and evaluating community initiatives.

Different Approaches in Community Initiatives

In this paper we draw a distinction between community action and community development. Community action attempts change in social structures and systems as well as social norms. In a community action project, community action is closely linked with the local-level implementation of specific public policies. For instance, around the issue of the use of alcohol and drugs, the Ministry of Education and ALAC funded a Community Action on Youth and Drugs (CAYAD) project, which was implemented in five localities.

Workers were employed who extensively networked with their local communities to develop specific strategies to reduce school suspension for cannabis use. The locally appropriate strategies took into account the illegal status of cannabis but also the relatively high levels of community use. The workers networked with each other to share strategies and resources as part of a formative evaluation process which involved the Alcohol & Public Health Research Unit (APHRU) and their Maori research partner, Whriki.

In contrast, a community development approach is characterised by the wider general goal of community empowerment rather than addressing more specific policy goals.

An example of a community development project is the recently established project in Ranui, part of the Stronger Community Action fund initiative, for which APHRU/Whriki is also providing evaluation input. In this case funding has been provided to the community to undertake whatever type of development work it believes it needs in order to build the community capacity of the locality.

Community action needs to include aspects of a community development approach to achieve its more directly policy-oriented goals, but is usefully conceptualised as a different enterprise. Distinguishing between these two types of activity helps to clarify understanding of what are appropriate and feasible outcomes to be expected from community initiatives of different types.

The distinction between community action and community development informs the second part of this paper, which considers the evidence for change consequential upon such resourced community initiatives

Who and What Qualifies as Community?

The word community conveys a sense of connectedness between people and their organisations. In relation to community initiatives funded from outside, this usually relates to geographical locality and, often in New Zealand, also recognises ethnic communities within a geographical area. The community is a social space, a sector made up of informal and relatively unmanaged associations.

Community can also be defined by shared interests regardless of the geographical and cultural characteristics of its members and arises as a result of political, work, recreational, sporting, or other commonalities amongst their members which keep them in contact for a meeting of their common needs. However most of the resourced and evaluated community initiatives in New Zealand have worked with geographical communities and these provide the bulk of the examples used in this paper.

In any geographical locality there are communities within communities. A specific New Zealand response to our diversity has been the development of "by Maori, for Maori" projects, either developing in response to mainstream initiatives (Stewart and Conway 2000) or funded from separate funding streams.

Community initiatives differ in the degree to which they emphasise outreach to the under-represented and in the diversity in the coalitions that are built. Outreach to the under-represented grassroots sector is likely to be an important aim for community development, more so than for community action projects in general.

A number of community action projects we have evaluated have demonstrated the problems inherent in an expectation by funders that community activities can rely on sustained input from grassroots (see, for example, Stewart and Conway 2000). Community action projects can usually rely on a reservoir of community concern about the issues being addressed which provide moral but not active support in the long term (Duignan et al. 1993).

Explanations for the lack of involvement of grassroots volunteers in such projects have focused on competing claims of the time demands of paid work, mobility of population and demoralisation related to poverty and isolation (Bjaras 1991, Maskill and Hodges 2001). Lack of involvement may also reflect the nature of the project and the extent to which it originated as a top-down initiative or a more bottom-up approach.

However, even a strong community development approach - emphasising, as it does, a bottom-up approach and the need for outreach - is still constrained in terms of volunteer involvement by the conditions of the time. A postmodern analysis of current social movements finds people reject "unreasonable" demands on their time, have no expectation of participating in long-term, hierarchical organisations and would rather engage in "networks of waxing and waning groups and strategies" (Vester cited in Rosenau 1994). A decline in civic participation over time has also been linked to the increasing demands of television watching (Brehm and Rahn 1997).

Successful community action projects have instead developed and strengthened relationships within and between the sectors most able to contribute to strategies to meet the objectives of the project, often including institutions such as the police, local government and health agencies. In rural areas, for example, farming communities have formed partnerships with regional councils and agricultural and conservation agencies (Witten et al. 2000). This cross-sectoral activity is integral to community capacity building (Bush 1997) enabling the more effective use of existing resources.

Within a community action project, employees of local services and institutions are viewed as community members with key roles to play. Coalitions are built to implement specific strategies and these may bring together different political interests in a temporary alliance. Attempts to prescribe that community programmes work from a base which must include the full range of different interests have been critiqued in the evaluation literature as impeding effective change (Holder and Reynolds 1998).

Who Defines the Goals for Community Initiatives?

Community initiatives also differ in who defines the issues they deal with. A characteristic of community development is that local actions are centred on community-defined issues. Reflecting this, processes such as participation in needs assessments and community visioning (Ames 1997, National Civic League 1996) are becoming important parts of community development.

Experts provide their services and conceptual frameworks but in a way which is, at least superficially, in the service of the community members. Leadership development and organisation building are integral and primary goals of the community development approach, as an identified group has to emerge to address issues that may not have been explicitly addressed within the community before.

Community action projects, with their more specific, policy-related objectives, start with accepting government priorities and funds for community initiatives. Given the stability in priorities over time it is likely that organisations already exist within the locality with some responsibility for the issue and the focus of community action therefore tends to be on building horizontal and vertical linkages in order to collaborate on strategic objectives rather than developing local leadership or building new organisations.

A community action approach can also be used in areas with less well-developed community-level resources. The CAYAD community action mentioned above, addressing issues of cannabis use among youth in areas with high rates of school suspensions, was an example of a project that responded to a government tender and had specific goals.

However, the way in which the communities responded brought in new players and new strategies. In the predominantly Maori rural communities involved in the project, iwi-based organisations played a key role. In several communities new opportunities were created for the youth voice to be heard in the community (Conway et al. 2000).

How Are Strategies Chosen?

In community action, broad objectives in line with policy goals are agreed between funders and the initial entry point into the community (often a non-governmental organisation). The strategies employed, however, are under local community control and are responsive to the local conditions. Community action strategies therefore challenge existing organisational structures in a way that a social planning approach will not.

The community action model as developed by APHRU/Whriki within New Zealand is one in which strategies are informed by both community knowledge and research-based knowledge. Research-based knowledge of what has been shown to be effective elsewhere is contributed to the project by formative evaluators who also play the role of critical friend in reflecting on project directions.

The formative evaluation also facilitates ongoing data collection and utilisation (e.g. Stewart et al. 1997, Stewart and Conway 2000). However, data collection without a focus on community organisation is not enough to effect change (Kennedy 2001). Formative evaluators from their position of slightly outside the service provider and organisational mix have also played the role of independent broker in the vertical relationships between the community players and the funding agency.

They have also played this role in the pursuit of horizontal linkages between existing organisations and providers who may have to set aside different perspectives and interests in pursuit of a shared objective. The ideal relationship between the professionals and the community sector in this context has been described as one of a respectful partnership (Casswell 2000, Holder and Reynolds 1998).

In describing community initiatives in this paper so far, a sharp distinction has been drawn between community development and community action, but in practice there are many overlapping elements which may coexist in the one programme. Both are resourced activities in which similar community processes of networking, alliance building and skills enhancement take place.

The major difference is in the extent to which the project's objectives are defined by a funding agency (more so in community action) and the extent to which the community development processes are valued in their own right or are subsidiary to achieving these goals.

Does It Work?

Community-level initiatives are not easily amenable to evaluation using an experimental design (Casswell 1999). Random allocation of communities to "treatments" or control, ensuring no cross-transfer of knowledge and skills for an appropriate period of time, is not readily applied to communities - although in some studies quasi-experimental designs have been used (e.g. Casswell and Gilmore 1989, Holder et al. 1997, Wagenaar et al. 2000).

In many contexts, however, alternative methodologies are better science. Many evaluations of community-level activity have therefore utilised a case study methodology. Ideally they collect both quantitative and qualitative data, gain data from multiple sources to provide a multifaceted picture of what has gone on and analyse data within a theoretical framework which allows inference of causal effects (Yin 1992).

Unexpected Outcomes

Anticipating likely change as a consequence of a community initiative is only ever partial at the start of the initiative, because of the flexibility of goals in the case of community development, and of strategies in the case of community action.

Evaluation plans therefore need to allow for the assessment of unexpected outcomes as well as progress measured against agreed objectives. A community action project, which was a collaboration with two Maori Trusts and responded to government goals of reducing the extent of "drink driving" among Maori, was assessed against the agreed objectives. These objectives were:

• developing and implementing a marae-focused programme aimed at increasing support among Maori for culturally appropriate strategies to prevent alcohol-related traffic crashes;
• developing and implementing a coordinated media strategy, including media advocacy and paid Maori mass media, to increase support for culturally appropriate strategies;
• developing and implementing strategies to reduce drunkenness in drinking environments in which Maori drink; and
• developing strategies aimed at increasing the mutual supportiveness of compulsory breath testing and the programme components.

However, the evaluation also found evidence of enhanced social cohesion in the local communities, particularly in terms of the perception by Maori communities of the police (who had become active partners in the projects), and of Maori by the police (Moewaka Barnes 2000). This same project also demonstrated its impact on the community's perception of the participating organisations, the Trusts. Thus the location of the project in Maori Trusts and consequential ownership by Maori not only enhanced the impact of the project (as expected), but the project also had a positive effect on the standing and visibility of the Trusts in the broader community (ibid.).

Measurement Issues

From the published evaluations of community development projects there is considerable evidence that resourced community development achieves its objectives when these are couched as improvements in community processes. Throughout the literature there is a strong assumption that this enhancement of process is in itself valuable. Reviews of community development projects in New Zealand and overseas (Ehrhardt 2000, Gillies 1998, Maskill and Hodges 2001, Smith and Herbert 1997) have shown increased awareness of and interest in the issue, identifying the following as processes which have been consistently achieved in community initiatives:

• improved linkages between organisations resulting in coordinated activities and shared strategic direction;
• involvement of new actors and new solutions;
• bringing in new resources and pooling existing resource (information, skills and money); and
• modification of services and/or institutional change to meet local needs.

If community initiatives are viewed as part of social policy development, as often happens in community action, other changes become relevant measures of effectiveness. The extent to which the community voices its own understanding of issues and possible solutions, collaborates with media to advocate for community positions, and manages conflict and controversy when they arise, are all measures of community processes that are relevant to assessing the impact of community action as a contributor to policy development.

Sustainability can also be considered as an evaluation measure. Both securing ongoing funding and institutional change may be seen as measures of effectiveness where ongoing structures are needed in a community to address an issue. In a Scandinavian community action project aimed at the goal of injury prevention, for example, the project was viewed as successful when, eight years after its inception, the municipality established a Safety Board (Lindqvist et al. 1996).

In New Zealand an evaluated community action project on alcohol carried out in the 1980s resulted in local funding continuing the work of the local health promotion workers and an extension of the concept to about 25 communities around New Zealand (Stewart et al. 1993). The development of community alliances and data-gathering tools has also shown considerable sustainability. For example, the liquor-licensing liaison groups and the collection of data by police to identify poorly managed licensed premises for use by this alliance are ongoing strategies throughout New Zealand that were developed in the context of a community action project (Stewart et al. 1997).

Models of Impact

The evaluation results showing enhancement of community processes do not give direct evidence of an impact on health and well-being at the individual level. There are, however, possible models for the way in which enhanced community processes will enhance health and well-being. In one model, enhancement of health and well-being is due to an increased level of interpersonal trust and social connectedness within the community.

There is an established link between levels of social connectedness in a community and the health and well-being of its members (Berkman 1995) and, also, between civic participation, interpersonal trust and trust in public institutions (Brehm and Rahn 1997). While some level of relationship has been demonstrated in a number of contexts there is as yet apparently no direct evidence of an increase in social connections, established in community development projects, resulting in an improvement in health and well-being.

In a second model, the development or implementation of specific interventions directly change behaviour and risk exposure. Community initiatives showing improvements in health and well-being in this way have been described in this paper as community action.

The objectives of these initiatives focus on environmental changes which have immediate impact on behaviours and are therefore likely to elicit measurable change in the time period covered by the evaluation (Carr 2000).

A number of such findings of effectiveness have come from the alcohol field. One example of successful community action in Surfers Paradise increased regulation of licensed premises and the implementation of policies and a code of practice within the premises.

This improved bar staff practice and reduced alcohol-related violence (Hauritz et al. 1998, Homel et al. 1997). Closer to home, a case study of a community action at Piha, which resulted in bans on public drinking, found fewer incidents of injury and crime, and an improved sense of well-being (Conway 1998).

A controlled evaluation of a United States community action project found a reduction in drink driving in those aged 18-19 and in disorderly conduct among those aged 15-17. The community action project had achieved these results by reducing underage access to alcohol via a number of strategies, including pseudo-patron operations with alcohol outlets, keg registration and shortening hours of sale of alcohol (Wagenaar et al. 2000).

Another controlled evaluation in the United States found that bar staff training, implementation of policies in licensed premises, increased threat of enforcement, increased age verification checks, and highly visible "drink driving" enforcement, which were all results of a community action project, achieved a reduction in injury (Holder et al. 1997).

Community organisation in rural communities in the United States resulted, after four years, in a positive effect on social norms about women smoking and also on quit rates (Secker-Walker et al. 2000). In the injury-prevention field, an evaluation of the WHO Safe Community model in Sweden found a reduction in utilisation of health care due to injury as a result of the work of local action groups (Timpka et al. 1999).

The Waitakere Community Injury Prevention Project has also demonstrated a reduction in child injury hospitalisation (Coggan et al. 2000).

These are examples of projects with objectives around which there was a considerable consensus within the community (albeit some commercial imperatives operating in an unhealthy direction).

They also took place in contexts that were relatively sympathetic to the project. Existing agencies were mobilised to play a more active role following the input of a community worker with organising skills. Also, each of the projects was evaluated over a time period long enough to allow for such changes to take place, something that is not always the case (Ehrhardt 2000).

The Way Forward

Community initiatives depend upon a supportive policy environment to make a difference in people's lives. Local-level action in isolation is unlikely to ameliorate the effects of a policy environment hostile to its goals. Community initiatives inform the need for central policy change if lines of communication between the community voice and policy advisers are open.

Central government has a key role to play as a funder of community initiatives and can also encourage support by relevant agencies at the community level. It also needs to set realistic goals and time frames for funding (Carr 2000). Time periods of 10-15 years have been suggested in the literature.

In addition to measures directly related to health and well-being, resourced community initiatives, to the extent that they increase social connectedness, enhance trust and strengthen community organisations, will also contribute to government outcomes in relation to governance. To the extent that these projects facilitate capacity building in Maori communities, they will also contribute to meeting Treaty obligations.

These community initiatives have high face validity and an emerging evidence base to suggest their effectiveness. As such they are likely to continue to receive funding from a range of policy ministries. However, to ensure that this resource is well spent there needs to be investment of effort in strategic evaluation to increase the knowledge base and enable government funding streams to best support innovative and effective community capacity-building initiatives.

There is a need for knowledge specific to New Zealand as the different nature of our civil society at both central and local government levels, our ethnic mix and, in particular, government treaty obligations will affect the nature and outcomes of community initiatives. In order to achieve the accumulation of this knowledge base, capacity at the tertiary educational level needs to be developed so as to allow graduates to play an appropriate role in evaluation and training (Poole 1997).

In particular, evaluation training must be based on an understanding of the appropriate relationship between evaluator and community, where the evaluator adopts the role of learner and collaborator rather than that of expert (Perkins and Douglas 1995). Such collaboration with the kinds of community initiatives outlined in this paper will help fulfil the potential of tertiary institutions to contribute to social and economic gains (Tertiary Education Advisory Commission 2001).

References

Ames, S. (1997) "Community visioning: planning the future in Oregon's local communities" in B. Pable, B. McClendon and R. Quay (eds.) Proceedings for 1997 National Planning Conference Contrast and Transitions, APA, San Diego, California.

Berkman, L.F. (1995) "The role of social relations in health promotion" Psychosomatic Medicine, 57:245-54.

Binney, E. and C. Estes (1988) "The retreat of the state and its transfer of responsibility: the intergenerational war" International Journal of Health Services, 18:83-96.

Bjaras, G. (1991) "Can a community development model be used for health programmes in an industrialized country?" International Journal of Health Planning and Management, 6:209-219.

Brehm, J. and W. Rahn (1997) "Individual-level evidence for the causes and consequences of social capital" American Journal of Political Science, 41:999-1023.

Bush, R. (1997) "Can we achieve sustainable reductions in harm in local communities?" (editorial) Drug and Alcohol Review, 16:109-111.

Carr, A. (2000) Community Project Workers Scheme Crime Prevention Projects: Evaluation Report, Research Services, Department of Internal Affairs, Wellington.

Casswell, S. (1999) "Evaluation research" in C. Davidson and M. Tolich (eds.) Social Science Research in New Zealand: Many Paths to Understanding, Longman, Auckland.

Casswell, S. (2000) "A decade of community action research" Substance Use & Misuse, 35:55-74.

Casswell, S. and L. Gilmore (1989) "An evaluated community action project on alcohol" Journal of Studies on Alcohol, 50:339-346.

Coggan, C., P. Patterson, M. Brewin, R. Hooper and E. Robinson (2000) "Evaluation of the Waitakere Community Injury Prevention Project" Injury Prevention, 6:130-134.

Conway, K. (1998) "Booze, behaviour, beaches and bans: turning the tide through community action" Alcohol Advisory Council "Working Together" Conference, Nelson.

Conway, K., M. Tunks, W. Henwood and S. Casswell (2000) "Te Whanau Cadillac - a waka for change" Health Education and Behaviour, 27:339-350.

Duignan, P., S. Casswell and L. Stewart (1993) "Evaluating community projects: conceptual and methodological issues illustrated from the Community Action Project and the Liquor Licensing Project in New Zealand" in T. Greenfield and R. Zimmerman (eds.) Experiences with Community Action Projects: New Research in the Prevention of Alcohol and Other Drug Problems, US Department of Health and Human Services, Rockville, Maryland.

Ehrhardt, P. (2000) Models of Community-Government Partnerships and their Effectiveness in Achieving Welfare Goals: A Review of the Literature, Ministry of Social Policy, Wellington.

Etzioni, A. (1996) The New Golden Rule: Community and Morality in a Democratic Society, Basic Books, New York.

Gillies, P. (1998) "The effectiveness of alliances or partnerships for health promotion" Health Promotion International, 13:99-120.

Hauritz, M., R. Homel, M. Townsley, T. Burrows and G. McIlwain (1998) An Evaluation of the Local Government Safety Action Projects in Cairns, Townsville and Mackay. Report to the Queensland Department of Health and the Criminology Research Council, Centre for Crime Policy and Public Safety, Griffith University and Queensland Department of Health.

Holder, H. and R. Reynolds (1998) "Science and alcohol policy at the local level: a respectful partnership" Addiction, 93:1467-1473.

Holder, H. R. Saltz, J. Grube, R. Voas, P. Gruenewald and A. Treno (1997) "A community prevention trial to reduce alcohol involved accidental injury and death: overview" Addiction, 92:S155-S171.

Homel, R., M. Hauritz, R. Wortley, G. McIlwain and R. Carvolth (1997) "Preventing alcohol-related crime through community action: the Surfers Paradise Safety Action Project" Crime Prevention Studies, 7:35-90.

Humphrey, G. (2001) "Presentation" Harold Holder Seminar on the Use of Local Alcohol Policies to Reduce Alcohol Harm: Evaluation of a Community Action Trial in the United States, Alcohol & Public Health Research Unit, Auckland.

Kennedy, L. (2001) "Community involvement at what cost? Local appraisal of a pan-European nutrition promotion programme in low-income neighbourhoods" Health Promotion International, 16:35-45.

Lindqvist, K., T. Timpka and L. Schelp (1996) "Ten years of experiences from a participatory community-based injury prevention program in Motala, Sweden" Public Health, 110:339-346.

MacMaster, M. (2001) "Presentation" Harold Holder Seminar on the Use of Local Alcohol Policies to Reduce Alcohol Harm: Evaluation of a Community Action Trial in the United States, Alcohol & Public Health Research Unit, Wellington.

Maskill, C., and I. Hodges (2001) Intersectoral Initiatives for Improving the Health of Local Communities: A Literature Review. Report to the Ministry of Health (draft), HealthSearch, Auckland.

McKnight, J. (1987) "Regenerating community" Social Policy, 54-58.

Moewaka Barnes, H. (2000) "Collaboration in community action: A successful partnership between indigenous communities and researchers" Health Promotion International, 15:17-25.

National Civic League (1996) The Community Visioning and Strategic Handbook, National Civic League, Denver, Colorado.

Perkins, D. and D. Douglas (1995) "Speaking truth to power: empowerment ideology as social intervention and policy" American Journal of Community Psychology, 23:765-794.

Petersen, A. (1994) "Community development in health promotion: empowerment or regulation?" Australian Journal of Public Health, 18:213-217.

Poole, D. (1997) "Building community capacity to promote social and public health: challenges for universities" Health and Social Work, 22:163-170.

Robertson, A. and M. Minkler (1994) "New health promotion movement: a critical examination" Health Education Quarterly, 21:295-312.

Robinson, D. (1997) Social Capital and Policy Development, Institute of Policy Studies, Victoria University of Wellington.

Robinson, D. (1999) Social Capital in Action, Institute of Policy Studies, Victoria University of Wellington.

Rosenau, P. (1994) "Health politics meets post-modernism: its meaning and implications for community health" Journal of Health Politics, Policy and Law, 19:303-333.

Secker-Walker, R., B. Flynn, L. Solomon, J. Skelly, A. Dorwaldt and T. Ashikaga (2000) "Helping women quit smoking: results of a community intervention program" American Journal of Public Health, 90:940-946.

Smith, B. and J. Herbert (1997) Community-based Initiatives: Gateways to Opportunities, A Report of the Community-based Action Research Project, Department of Social Security, Canberra.

Stewart, L., S. Casswell and P. Duignan (1993) "Using evaluation resources in a community action project: formative evaluation of public input into the implementation of the NZ Sale of Liquor Act" Contemporary Drug Problems, 20:681-704.

Stewart, L., S. Casswell and A. Thomson (1997) "Promoting public health in liquor licensing: perceptions of the role of alcohol community workers" Contemporary Drug Problems, 24:1-37.

Stewart, L. and K. Conway (2000) "Community action to reduce rural drink and drive crashes in New Zealand: adapting approaches in dynamic environments" Substance Use & Misuse, 35:141-155.

Tertiary Education Advisory Commission (2001) Shaping the Strategy: Third Report of the Tertiary Education Advisory Commission, Ministry of Education, Wellington.

Timpka, T., K. Lindqvist, L. Schelp and M. Ahlgren (1999) "Community-based injury prevention: effects on health care utilization" International Journal of Epidemiology, 28:502-508.

Wagenaar, A., D. Murray and T. Toomey (2000) "Communities Mobilizing for Change on Alcohol (CMCA): effects of a randomized trial on arrests and traffic crashes" Addiction, 95:209-217.

Witten, K., M. Parkes and L. Ramasubramanian (2000) "Participatory environmental health research in Aotearoa/New Zealand: constraints and opportunities" Health Education and Behavior, 27.

Yin, R. (1992) "The case study method as a tool for doing evaluation" Current Sociology, 40:121-137.