Accessing Community

QUALITATIVE Sixsmith et al. / HEALTH ACCESSING RESEARCH THE COMMUNITY / April 2003

10.1177/1049732302250759 ARTICLE

Accessing the Community: Gaining Insider Perspectives From the Outside Judith Sixsmith Margaret Boneham John E. Goldring

Accessing participants for research projects is often treated as unproblematic. However the experience outlined here of negotiating access to participants within a community setting illustrates the inherent difficulties of recruitment. The authors describe the techniques used and practical challenges faced when accessing participants within a socially deprived community for a qualitative research project on social capital. They used a number of different strategies to generate a diverse sample including advertising, snowballing, accessing gatekeepers, and street surveys. The value of a stakeholder analysis is described alongside issues surrounding the use of gatekeepers. Rather than acting as outsiders seeking participants at every available opportunity, a more fortuitous strategy involved the ethnographic approach of “being there” as active contributors to community life. Here, the cornerstones of credibility and trust were addressed in a process of continually negotiating access from a semiinsider position. Keywords: recruitment; community; research process; researcher role; stakeholder analysis

O

ur aim in this article is to describe the inherent difficulties, but also the great satisfactions, of negotiating access into a community for the purposes of conducting research. We explored the relationship between social capital, health, and gender in a socially deprived community in the Northwest of England. This was a Health Development Agency–funded project that took an explicitly multimethod qualitative approach and necessitated the negotiation of access into a tightly knit community to generate a sample of male and female, younger and older participants. Published accounts of such research tend either to present access as a fait accompli or to sanitize the whole process. There are few existing guidelines concerning the ways in which research access to socially deprived communities can be negotiated. Indeed, Campbell and McLean (2001) have called for further methodological work to explore the processes involved in successful recruitment of diverse community members. Our experience of gaining access into a close-knit

AUTHORS’ NOTE: This research was funded by the Health Development Agency, London, UK. The authors wish to thank all the participants who gave their time when participating in this study. Thanks also go to the anonymous reviewers, whose comments on an earlier draft of this paper were most helpful. QUALITATIVE HEALTH RESEARCH, Vol. 13 No. 4, April 2003 578-589 DOI: 10.1177/1049732302250759 © 2003 Sage Publications

578

Sixsmith et al. / ACCESSING THE COMMUNITY

579

community was certainly problematic and felt at times like a tightrope-walking act. We will outline here the recruitment techniques employed and examine the practical issues that occurred as a product of undertaking community health research. In addition, strategies for maintaining credibility and trust within the community will be explored. As the research process unfolded, it became evident that obtaining a sample was not simply a matter of recruiting people into the research but, rather, a complex social process of gaining access into the community itself. As “outsiders” to a community, researchers need to address the psychosocial distance between themselves and the researched to gain valuable insights into the everyday lives of community residents. Here, the notion of outsiders and insiders is brought to the fore. Jewkes and Letherby (2001) have argued for an understanding of the complexities of the insider-outsider relationship. They suggest that this is a continually negotiated relationship where outsiders sometimes occupy social positions as insiders, and vice versa. It is in this context that the research team was at varying times both outsider and insider to the particular community under research, making access to the community a complex process of managing social relationships and personal impressions. To place the research in context, the concept of social capital will be discussed. Indeed, the fact that social capital provided the conceptual framework for the research had important implications for gaining access to community members.

THE NOTION OF SOCIAL CAPITAL Putnam’s (1995) notion of social capital refers to “features of social organisation such as networks, norms and trust, that facilitate coordination and cooperation for mutual benefit” (p. 66). He argues in his more recent book, Bowling Alone (2000), that community-based social decline is a product of lack of social networks, social trust, and reciprocity. Rather than joining in with group activities, Putnam suggests that people are increasingly “bowling alone,” reducing their access to social capital as a resource in times of need. Moreover, it has been argued that there is a link between social capital and health. As Putnam (2000) says, If you belong to no groups but decide to join one, you cut your risk of dying over the next year in half. If you smoke and belong to no groups, it’s a toss up statistically whether you should stop smoking or start joining. (p. 331)

In support of Putnam, Kawachi, Kennedy, Lochner, & Prothrow-Stith (1997) linked declining social capital to higher levels of morbidity and mortality. Kawachi’s study, along with much of the empirical work in this area, has taken a quantitative survey-based approach, often based on secondary analysis of data (Campbell, Wood, & Kelly, 1999; Kennedy, Kawachi, Prothrow-Stith, Lochner, & Gupta, 1998). These researchers have struggled to identify the mechanisms and processes through which social capital might operate. They also tend to assume that social capital exists as a wholly positive force in people’s everyday lives. In our research, we sought to question such assumptions by examining the meanings and lived experiences of family and community life and how these came to play a part in the way people experienced and managed their health. By taking a qualitative perspective, we were able to examine the detailed narrative accounts of

580

QUALITATIVE HEALTH RESEARCH / April 2003

people’s health experiences and interrogate these for evidence of social capital. Moreover, we hoped to be able to disentangle the critical role that living in a socially deprived community might play in influencing health inequalities.

THE COMMUNITY The community in which the research was conducted was a place of strong contrasts. On the surface, it appeared to be pleasant and safe. However, this belied an area that was multiply disadvantaged. The population was predominantly White British living in terraced council housing. This was a stable community, with people moving within local streets but little migration in and out of the area. A recent local government report into the area (Bolton Metropolitan Borough Council [MBC], 1994) highlighted unemployment at approximately 10% and reported that 50% of households were on housing benefit. Thus, the community was composed of lowincome households, with a high percentage of single-parent families. Crime rates were above average, especially break-ins and drug-related crime. The area had few recreational facilities. A local park was available but tended to be the locus of activity for drug dealing and youth gangs. There were three main community centers in the area, which provided child playgroups and bingo. Retail facilities were restricted to a few shops, such as fish and chips and newsagents. There were no local pubs on the estate and no designated venues for older people to meet. In terms of health statistics, coronary heart disease and cerebrovascular diseases within the community stood at 26% and 27%, respectively (Bolton MBC, 1994), and perinatal mortality was particularly high. Dog dirt and litter were a particular health and environmental problem. Neither residents nor council workers took responsibility or initiative for improving the environment. This illustrated the physical manifestation of declining social capital in the area, which had immediate health implications for community members.

METHOD We adopted a multimethods approach to our research designed to access as many different sorts of data as possible. Consequently, a case study methodology was adopted (Yin, 1984). The study included the following methods: • In-depth interviews: Interview schedules were designed to reveal meanings, experiences, and participant narratives concerning their health status and history, and the role of social networks (friendships and family) and community life in supporting people in coping with stressful life transitions. • Organizational interviews: These established what sorts of facilities/resources existed in the area and their use. • Focus groups: The focus group agendas explored shared understandings and social talk about health, local health services, ideas about community belonging, and health promotion. The groups were structured by age and gender to allow contrasts and similarities in discourse to emerge. • Street interviews and a community center questionnaire: The purpose of the street interviews and questionnaires was to investigate qualitatively and quantitatively key

Sixsmith et al. / ACCESSING THE COMMUNITY

581

issues around perceptions of community and notions of belonging, which had arisen during the in-depth interviews and group discussions.

In this article, we explore issues surrounding the generation of a sample of community participants for the interviews and focus groups. Although the interviews and focus groups were conducted in two intensive periods between February to April and July to August 2000, building links within the community to recruit participants was a much lengthier process, which began in November 1999 and continued until March 2001. Forging links with the community and building a platform of credibility for the project were essential first steps toward recruiting participants. This began at the outset of project planning and continued through to community feedback sessions at the end of the project.

The Researchers It is recognized that the interview process is a dynamic relational interaction between people and that the values, attitudes, experiences, and role of the researcher, the interview schedule, and the aims of the project are all part of a vibrant process influencing the final narrative (Sixsmith, 1999). As Phoenix (1991) has highlighted, Researchers are not objective observers of social contexts and interactions, but are members of society who have specific social locations and who bring particular orientations to bear on their research. (p. 91)

It is thus important to come to terms with who the researchers were, as their involvement in the narrative production is crucial. At times, their social background and gendered identities helped to create an insider empathy with the experiences of the community participants. At other times, a more distanced researcher role encouraged people to divulge personal reflections on their lives, as they knew that the researcher (as an outsider) was unlikely to exchange such confidences within the community setting. In view of this, a word on the research team is presented here. The team, all of whom were White (reflecting the researched community), consisted of two female research directors, one male research assistant, and one female interviewer (ages ranged between 36 and 52 years). The research directors had experience in working with issues of health and community living, particularly around problems of disadvantage and inequality. The research assistant and interviewer held the major responsibility for generating community links and constructing the project sample. Both originated from working-class backgrounds and had prior experience of living and working in socially deprived communities. This was crucial to establishing empathy and trust between researcher and researched. In this sense, it was also important that the team consisted of both males and females to reflect the gender balance of the participants. Throughout the life of the project, there was continual discussion about the ways in which a research presence in the community was both ethically acceptable and ideologically appropriate. In this way, the varied backgrounds of the team enabled a rich cross-fertilization of ideas, which enhanced understanding of the chosen community.

582

QUALITATIVE HEALTH RESEARCH / April 2003

The Sample To address the aims of the study, a structured sample was required, in which gender and age differences were represented. Thus, the sample included both men and women aged between 18 and 79 years old (detailed below in Table 1). We deemed it inappropriate to select a “representative” sample of community members, as people varied in many ways from each other. Alternatively, a strategy of “theoretical sampling” (Glaser & Strauss, 1967) was used. Theoretical sampling works with the notion of participant diversity. The aim is to recruit participants into the research as long as each person’s thoughts and experiences add to theoretical or conceptual insight. This continues until saturation is reached and additional recruits do not generate new information (see Taylor & Bogdan, 1998). Hence, in this study, theoretical sampling ensured that the universe of content pertaining to the particular research area of social capital and health was adequately covered. In total, 146 individual community voices were included in the research. In Table 1, we have highlighted the size of the sample, structured by gender and age, for the different methodologies used. In addition, we conducted 23 interviews with people representing 17 organizations, and a further 70 people completed community questionnaires. Although it is acknowledged that research can only ever offer a partial insight into the lives of others (Jossellson, 1996), the sample generated gave access to a wide variety of experiences and opinions. The strategies used to access the sample will be detailed in the remaining sections of this article.

ACCESSING THE COMMUNITY The first problem we encountered was that none of the research team had any personal contacts within the community. Given this lack of prior knowledge, we decided that the generation of an initial sample should involve the strategy of conducting a stakeholder analysis.

Stakeholder Analysis Grimble, Chan, Aglionby, & Quan (1995/1999) have defined stakeholder analysis as “an approach for understanding a system by identifying the key actors or stakeholders in the system, and assessing their respective interest in that system” (pp. 3-4). When considering who might be eligible as a stakeholder, it is useful to think in terms of the social contexts within which people are located. Thus, local community residents, formal and informal group leaders, service providers, and business people working within the area, indeed, anyone with an interest in the community can be defined as a stakeholder, and all offer links into the community. These links are extremely useful in gaining an understanding of both the complex interactions underlying community living and the very different needs of community groups. It is from this deeper understanding of the dynamics of community that specific community gatekeepers can be identified and approached. As part of this analysis for the current study, a diverse range of stakeholders were contacted, including community center committee members and users,

Sixsmith et al. / ACCESSING THE COMMUNITY

TABLE 1:

583

Sample Size and Composition Semistructured Interviews (n = 46)

Younger men Older men Younger women Older women

Number

Mean Age

Street Interviews (n = 69)

13 11 11 11

31 65 28 66

11 25 9 24

Focus Groups (n = 31) 8 7 8 8

tenants, council officials, police, shop keepers, youth workers, religious leaders, educational providers, bus company representatives, and local project leaders. Talking to such stakeholders was invaluable in four respects: 1. It revealed the values, ideologies, and linguistic diversity prevalent within the community, including hitherto concealed community groups such as youth subcultures (locally known as “alchies,” “bagrats,” and “smackheads”). 2. It underscored the interpersonal conflicts between community members making it possible to minimize their impact on the project. 3. It enabled an evaluation of the stakeholder’s potential to broaden the diversity of the sample. 4. It pointed to valuable community activists and gatekeepers who had the potential for introducing the research team to the wider community.

Gatekeepers It is important to understand both the benefits and the disadvantages of using community leaders for accessing participants. Berg (1999) highlighted the “value of obtaining the support of community leaders when conducting research in minority communities” (p. 239). However, community leaders can also act to block access in their role as community gatekeepers. Seidman (1998), commenting on gatekeepers, contended that they fall into two categories: “absolutely legitimate (to be respected) to self-declared (to be avoided)” (p. 37). Although this might be a valid point for much research, the social capital project, with its emphasis of community, concluded that all gatekeepers were stakeholders and thus worthy of respect. In addition to this, however, it was acknowledged that gatekeepers could be working to their own agenda regardless of legitimacy. They could attempt to influence the research process with their own version of “reality” by indicating only participants “approved of” by themselves. As the guiding principle of the research was one of inclusiveness, we sought to include as many views and narratives as possible. Consequently, all gatekeepers and their recommendations, although respected, were treated with some degree of caution. Dealing with gatekeepers can have great benefits. They have local influence and power to add credibility and validity to the project by their acceptance of it (Seidman, 1998). Alternatively, they can erect barriers, preventing access and effectively shutting the project down before it has begun (Berg, 1999). For example, one younger female gatekeeper refused access to her friends, saying, “No, they don’t do that sort of thing.” However, several of her friends were obtained as participants via

584

QUALITATIVE HEALTH RESEARCH / April 2003

different routes and were interviewed. Had her denial been effective, she would have denied access to a set of extremely rich and colorful narratives. As this example shows, negotiating obstacles and barriers was an integral part of accessing community samples requiring constant attention. In the current research, several gatekeepers were in conflict. They were jealous of their own part in the project and dismissive of each other. To keep them on board entailed a highly sensitive approach that balanced loyalties. For instance, one voluntary gatekeeper relinquished her role as a community leader due to the perceived lack of support from professionals. Both gatekeeper and professionals sought our approval as researchers to rally support for their position. This left us in a difficult situation, as neutrality was essential if we were to maintain full access to the community. In this case, we represented the views of each party to the other to engender more understanding of their different perspectives. Such conflicts (and strategies for resolving them) caused some consternation. It is at this point that the research team provided both emotional and practical support for each other in talking through ways to deal with emergent dilemmas. As Moore, Sixsmith, and Knowles (1996) have suggested, the research team members have a responsibility to each other when sensitive situations arise.

Snowballing Once gatekeepers had identified initial participants, a snowballing technique was used to enhance the variety of the sample. The advantages and disadvantages of this recruitment strategy are well documented (see Gilbert, 1993; Robson, 1993). For the current project, this technique proved useful in many respects. Seidman (1998) has commented that introductions from other community members can help to even out inherent power relations as a known person introduces the researcher to the participant. The researcher might be perceived as a “friend of a friend,” whereas the participant is viewed as an expert on his or her own experiences (e.g., Harré & Secord, 1972). To ensure that participants feel valued, it is important to explain fully the aims of the project and the participants’ role in the research process. It is in this way that the foundations of trust and rapport can become established (Standing, 1998). Moreover, the snowballing technique can sometimes be one of the few ways of accessing a vulnerable or inaccessible sample because security features are intrinsically built into the sample framework (Lee, 1993). Access to community members who were embedded in the local drug culture was achieved (in the current project) only via credibility established through their family and friendship networks. There are also drawbacks to the use of snowballing. It can result in a restricted sample of participants drawn from similar backgrounds. For research into social networks and social capital, the snowballing technique can be somewhat tautological and, indeed, could jeopardize the data. Because of such drawbacks, we implemented other strategies for accessing the community—advertising and questionnaires.

Advertising: All That Glitters . . . Advertising for participants offers an explicit and dedicated avenue for recruitment. In the social capital project, participants were sought by posting notices in

Sixsmith et al. / ACCESSING THE COMMUNITY

585

church halls, community centers, and youth centers. Although this strategy has worked for other projects in terms of attracting a sample base (Leicester Social Services, 1986), its use in a socially deprived area might not be successful. Community members might feel that such appeals are not relevant to them and be unwilling or unable to proactively pursue the opportunity, perhaps due to lack of self-confidence and resources. This may have been the situation in the social capital project, as advertising generated no response. However, it might be that advertising contributed to heightening the profile of the project within the community, thereby adding a sense of legitimacy when participants were approached through other recruiting techniques (Berg, 1999).

Questionnaires A short, closed-response questionnaire was designed to elicit attitudes toward community living (including issues of crime and public safety, local facilities, and sense of belonging). These were placed in the neighborhood community centers advertised through poster displays. Residents were invited to complete a questionnaire anonymously. However, there was also an opportunity for people to volunteer to take part in the in-depth interviewing and group discussions. In fact, this strategy was rather unsuccessful in that, although 70 questionnaires were returned, only eight people elected to assist further in the project. It is possible that the lack of personal contact with researchers inhibited any willingness to continue participating, particularly when community members have had very little experience of research. This demonstrates the distance between the researcher and the researched and the need to bridge this gap in a more personal engagement.

Street Interviews One of the methods employed within the project was that of street interviewing. Here, community residents were approached in the street and requested to complete a 10-minute interview involving short answers to several open-ended questions. This gave community residents some insight into the aims of the project and personal contact with the researchers. Thus, it became relatively easy to recruit people for the more in-depth interviews. Having already contributed to the project and met the researcher in the street increased the likelihood that residents would agree to further, more intensive participation in the research. Indeed, many of the focus group participants were recruited in this way. If recruitment through street interviews is to be successful in generating a diverse sample, then attention needs to be paid to the sorts of people who will be available on the streets at particular places and times of day. Daytime and evening, weekday and weekend slots are required to access a diverse selection of community residents. In our own research, recruitment took place at a variety of locations, all of which provided centers for community life, such as local shops, medical clinics, and so on. However, this form of recruitment might exclude those people whose disabilities prevent active community participation. With this in mind, researchers need to become much more involved with community life to identify more marginalized and vulnerable people.

586

QUALITATIVE HEALTH RESEARCH / April 2003

Being There—Being Seen To access as diverse a community sample as possible and to ensure inclusion of more marginalized peoples, an ethnographic strategy of “being there” was adopted (see Whyte, 1955). This involved our researchers in immersing themselves in the various subcultures (drug users, youth gangs, old age forums, parent groups, etc.) that made up the community. Here, the principle was to get as close as possible to people’s lived experience. In particular, we used our own experience in materially and socially deprived situations to build a bridge to the lives of participants. For instance, knowing how to offer sympathy to a young depressed man and when to stand back and allow an older female to grieve for her lost family all had personal resonance in our own lives. The level of insider information about such experiences and how living in a working-class community contextualized these events was of paramount importance in deciding how to deal with very emotive situations. This empathy born of common understandings created a platform for the genuine exchange of sensitive and personal information, which enriched the quality of data received. In the social capital project, the research assistant spent many hours doing various community activities, such as being bandaged up in the first aid class, bouncing babies on his knee in the mother-and-toddler groups, helping out at Christmas parties, making tea at the luncheon club, or just walking around the estate chatting to groups of youths meeting on street corners. Being there and being seen in the community not only provided opportunities to generate leads to possible participants but also afforded occasions to talk to different people and to evaluate the way in which the research itself was perceived. Employing these strategies gained the research team a certain amount of respect, credibility, and subsequent access to a more inclusive sample that went well beyond gatekeepers and their social networks. These included, for example, people cohabiting while claiming benefits, drug users, and those involved in criminal activities. In addition to this, one participant accessed could neither hear nor communicate verbally. It was, therefore, necessary to make use of an interpreter for this interview. Although we recognized this could have constrained the data, it had the advantage of “giving voice” to an otherwise “invisible” community member. Finally, although this research was not explicitly a Participatory Action Research project, being there in the community and observing the environment, organization, and structure presented opportunities to put various organizations and countless community members in contact with each other for mutual growth and benefit. This constitutes what Burton and Kagan (1999) call “the edge effect,” where communities can benefit from the knowledge and contacts that researchers bring, thereby developing bridging capital at a local level. In addition, the social capital project employed other methods of “putting back” into the community, such as creating a library of educational material in the community center and conducting a cookery course. This added authenticity to the research team, as it was felt by many residents that the team was not just taking but also contributing to the community. Residents’ attitudes toward the research were revealed in community feedback sessions, in which they learned about the results of the project and the research team learned about the impacts they had had on the community.

Sixsmith et al. / ACCESSING THE COMMUNITY

587

Sharing Knowledge We decided it was vital to share the knowledge and insights gained during the analysis of findings with the community residents who had been interviewed. The different methods and recruitment strategies employed gave rise to very rich and vivid data, which had direct relevance to the community involved. In particular, findings suggest that the community was characterized, in terms of social capital, by extensive bonding ties but not bridging ties. This meant that shared norms within the community could constrain efforts to deal with health problems. For instance, experience of “nerves” was commonplace, and such people were often advised by family and friends to calm nerves by smoking cigarettes. The advice of health professionals was rarely solicited or listened to. Moreover, community space was highly gendered and described by many as “female space.” Many men felt excluded from community activity and participation, resulting in poorly developed social networks that rarely operated to support them in times of need. Such findings emerged from the intensive interviews made possible by a sensitive and lengthy relationship built up not just with participants but also with the wider community. We invited all research participants to a meeting in the community center to discuss preliminary findings. A further meeting was held with representatives of the organizations involved and feedback given on their contributions. These events constituted a two-way process whereby findings were presented and explained by the researchers and a debate with residents was encouraged. These feedback sessions afforded the opportunity to share knowledge but also to ascertain areas of potential misunderstanding. As such, the relationship developed during the recruitment process was continued after data collection and analysis, underlining the importance of maintaining access, trust, and credibility through the life of the project. Indeed, as more funding was made available, we have been able to continue our work in the community, giving rise to further development of relationships and detailed information on community living.

CONCLUSION The aim of this article was threefold in terms of outlining recruitment strategies, examining practical issues and exploring strategies for maintaining credibility and trust.

Recruitment Techniques Perhaps one of the key issues to emerge here was that any one technique would not adequately provide a diverse sample. Rather, a combination of techniques that complemented each other was necessary to generate a sample, which gave voice to more vulnerable and marginalized members of the community. Furthermore, the application of such techniques could not be undertaken in a mechanical manner. A degree of sensitivity to the people of the community and an understanding of the potential impact the research process might have on the researched were essential components in the success of the project.

588

QUALITATIVE HEALTH RESEARCH / April 2003

Practical Issues Conducting a stakeholder analysis proved extremely useful in identifying the complex nature of community relationships and emphasized the deep family and friendship bonds apparent in this socially deprived community. It was only when this had been completed that a degree of understanding of community life was possible. Indeed, the sample could not have been constructed without a negotiation of the many conflicts and power struggles between community members made visible through the stakeholder analysis and contact with gatekeepers. As such, generating a sample could be characterized as a process of ongoing negotiation involving a huge investment of time (especially in terms of “being there”) and emotional energy on the part of the researchers and researched alike.

Strategies for Maintaining Credibility and Trust As the research progressed, it became clear that strategies for maintaining trust within the community were of paramount importance. As Lee (1993) has suggested, continually negotiating access with the community depends on the quality of interpersonal relationships between researchers and participants. In our research, strategies for maintaining credibility and trust included a reflexive stance, in which each contact with community members was thought through very carefully and included being there and being seen as a familiar face; actively seeking opportunities to contribute to community life, thereby establishing reciprocal relationships; explaining not only the aims of the project but also the process of research, and informing participants of their role in this process; and sharing the findings with the community to anticipate closure and to leave the community members in a positive frame of mind toward future research. Finally, this research has revealed that recruitment into projects needs to take into account the sociocultural context of the community (see Campbell and McLean, 2001). Without an understanding of the norms and mores of community life, access to participants could have been extremely difficult and the relationship between researcher and researched one of suspicion and noncooperation. The quality of data gained was a direct reflection of the ways in which researchers managed the difficult negotiation of “insider” and “outsider” status while tailoring recruitment strategies to the diverse population encountered.

REFERENCES Berg, J. A. (1999). Gaining access to underresearched populations in women’s health research. Health Care For Women International, 20, 237-243. Bolton Metropolitan Borough Council. (1994). Deprivation in Bolton. Bolton, UK: Corporate Deprivation Group. Burton, M., & Kagan, C. (2000, September). Edge effects, resource utilisation, and community psychology. Paper presented at the Third European Conference on Community Psychology, Bergen, Norway. Campbell, C., & McLean, C. (2001). Social capital, health and ethnicity: Final report. London: Health Development Agency. Campbell, C., Wood, R., & Kelly, M. (1999). Social capital and health. London: Health Education Authority. Gilbert, N. (Ed.). (1993). Researching social life. Thousand Oaks, CA: Sage. Glaser, B., & Strauss, A. (1967). The discovery of grounded theory. Chicago: Aldine.

Sixsmith et al. / ACCESSING THE COMMUNITY

589

Grimble, R., Chan, M. K., Aglionby, J., & Quan, J. (1999). Trees and trade-offs: A stakeholder approach to natural resource management. In D. Buckles (Ed.), Cultivating peace: Conflict and collaboration in natural resource management. Washington, DC: The World Bank, International Development Research Center. Retrieved October 1, 2000, from http://www.Idrc.Ca/Minga/Conflict/Cases_E.Html (Original work published 1995) Harré, R., & Secord, P. (1972). The explanation of social behaviour. London: Blackwell. Jewkes, Y., & Letherby, G. (2001). Insiders and outsiders: Complex issues of identification, difference and distance in social research. Auto/Biography Studies, 16(2), 41-50. Jossellson, R. (Ed.). (1996). Ethics and process in the narrative study of lives. London: Sage. Kawachi I., Kennedy B., Lochner K., & Prothrow-Stith, D. (1997). Social capital, income inequality and mortality. American Journal of Public Health, 87(9), 1491-1498. Kennedy, B. P., Kawachi, I., Prothrow-Stith, D., Lochner, K., & Gupta, V. (1998). Social capital, income inequality and firearm violent crime. Social Science & Medicine, 47, 7-17. Lee, R. M. (1993). Doing research on sensitive topics. London: Sage. Leicester Social Services (1986). Black elders in Leicester. Leicester, UK: Leicester County Council. Moore, M., Sixsmith, J., & Knowles, K. (1996). Children’s reflections on family life. London: Falmer. Phoenix, A. (1991). Social research in the context of feminist psychology. In E. Burman (Ed)., Feminist psychological practice (pp. 89-103). London: Sage. Putnam R. (1995). Bowling alone: America’s declining social capital. Journal of Democracy, 1, 65-79. Putnam, R. (2000). Bowling alone. New York: Simon & Schuster. Robson, C. (1993). Real world research: A resource for social scientists and practitioner-researchers. Oxford: Blackwell. Seidman, I. (1998). Interviewing as qualitative research: A guide for researchers in education and the social sciences. New York: Teachers’ College Press. Sixsmith, J. (1999). Working in the hidden economy: The experience of unemployed men in the UK. Community, Work and Family, 2(3), 257-277. Standing, K. (1998). Writing the voices of the less powerful: Research on lone mothers. In J. Ribbens & R. Edwards (Eds.), Feminist dilemmas in qualitative research: Public knowledge and private lives. London: Sage. Taylor, S. J., & Bogdan, R. (1998). Introduction to qualitative research methods: A guidebook and resource. Chichester, UK: Wiley. Whyte, W. F. (1955). Street corner society. Chicago: University of Chicago Press. Yin, R. (1984). Case study research: Design and methods. Beverly Hills, CA: Sage.

Judith Sixsmith, Ph.D., is a senior lecturer in psychology at the Manchester Metropolitan University in the UK. Margaret Boneham, Ph.D., is head of health, social, and community studies at Bolton Institute in the UK. John Goldring, M.A., is currently a research fellow at the Institute of Public Health, Research and Policy, Salford University, Manchester, UK.