Accessing Socially Excluded People — Trust and the Gatekeeper in the Researcher-Participant Relationship

by Nick Emmel, Kahryn Hughes, Joanne Greenhalgh and Adam Sales
University of Leeds; University of Leeds; University of Leeds; University of Sheffield

Sociological Research Online, Volume 12, Issue 2,
<http://www.socresonline.org.uk/12/2/emmel.html>
doi:10.5153/sro.1512

Received: 1 Sep 2006     Accepted: 5 Feb 2007    Published: 30 Mar 2007

Abstract

This paper describes methodological findings from research to recruit and research hard-to-reach socially excluded people. We review the ways in which researchers have used particular strategies to access hard-to-reach individuals and groups and note that little attention has been given to understanding the implications of the nature of the trust relationship between researcher and participant. Gatekeepers invariably play a role in accessing socially excluded people in research, yet discussion to date invariably focuses on the instrumental role gatekeepers play in facilitating researchers' access. In this paper we explore the possibilities for analysing relationships in terms of trust and distrust between gatekeeper and socially excluded participant. Our analysis considers the different kinds of relationships that exist between gatekeepers and socially excluded people and, in particular, the relationships of power between gatekeepers and socially excluded people. Insights into the nature of trust among socially excluded people will also be considered. Finally, we discuss how size and use of social networks among socially excluded groups and perceptions of risk in interactions with gatekeepers are important to understanding the possibilities for trustful relationships, and for meaningful and successful access for researchers to socially excluded individuals and groups.

Keywords: Social Exclusion, Access, Research, Gatekeepers, Trust, Distrust, Risk

Introduction

1.1 This paper aims to describe methodological findings from research to recruit and research hard-to-reach socially excluded people on a low-income estate in the north of England. In previous research[1] on this estate we were unable to recruit some individuals and groups. The challenge we therefore faced in the methodological research presented here was to access those socially excluded people who were extremely difficult to access previously.

1.2 The approach we employed in the research reported in this paper was to seek access through those whom we able to access, namely health and social care providers from the voluntary and public sectors and enterprising individuals in the community with no formal role. We particularly sought access through those working with socially excluded people in the low-income estate in which we were conducting the research. This paper reports on the kinds of relationships that exist between these people, who acted as gatekeepers for this study, and the socially excluded people with whom they work. Our analyses draw on the perspectives of these gatekeepers and also on the perceptions of the 27 socially excluded individuals we accessed, and were able to interview, through some of these gatekeepers.

1.3 The focus on social exclusion in this paper provides the context for the discussion of trust and distrust that we develop here. We draw on several perspectives generated through the research. First, the perceptions of social exclusion among socially excluded people. Secondly, the understandings of the different gatekeepers working on the low-income estate of those who are socially excluded. And thirdly, the appropriateness of the methods of access as understood by the socially excluded people we successfully accessed. Our reasons for adopting a methodological strategy of access through gatekeepers is briefly contextualised in an analysis of the policy framework within which social exclusion is being addressed in England.

1.4 Our paper begins with a review of the strategies researchers have used to access hard-to-reach individuals and groups. We suggest that these strategies are predicated on the building of a relationship of trust yet consider that, so far, insufficient attention has been given to explicating the nature of the trust relationship between researcher and participant or how the participant's relationship with the gatekeeper informs, or provides the possibility for, the successful formation of a relationship of trust between participant and researcher. We observe that gatekeepers invariably play a role in accessing socially excluded people in research but discussions to date commonly focuses on the instrumental role gatekeepers play in facilitating researchers' access.

1.5 Finally, we explore the possibilities of understanding relationships in terms of trust and distrust between gatekeeper and socially excluded participant. Our analysis considers the different kinds of relationships that exist between gatekeepers and socially excluded people and how these relationships facilitate or impede access. We argue that such an exploration provides insights into the nature of trust among socially excluded people. Our paper concludes by suggesting ways in which more appropriate relationships can be built for meaningful and successful access to conduct research with groups who, hitherto, have rarely been given a voice in research.

Trust and access

2.1 Trust has been established as important by researchers in facilitating access to hard-to-reach vulnerable groups to the extent that much methodological literature is largely descriptive of strategies of trustful access. However, the reasons why these strategies may be effective are not considered. To address these lacunae we draw on sociological literature on trust to make sense of these strategies of trust- building. This review provides a context for the discussion in the rest of the paper about methodologies to access socially excluded individuals and groups.

2.2 The position of the researcher as an outsider is considered important in devising methods to build trust in the researcher participant relationship because the degree of social separation between researchers and their study population is characterised by distrust. While distrust is not the antithesis of trust (see below), nevertheless it raises a range of concerns research has sought to address. Elliott et al. (2002), for instance, employ a strategy of peer interviewers to access drug users. They observe that the value of these peer interviewers is their ability to introduce and vouch for outside researchers because they are trusted by those the researchers wish to access. Similarly, Kuebler and Hausser (1997) emphasise that peer interviewers not only vouch for the researchers but build relationships of empathy and mutual trust with groups that are difficult to access, such as illicit drug users.

2.3 While Elliott et al. (2002) and Kuebler and Hauser (1997) consider the importance of empathy to generate a trustful relationship between researchers and participants, Sixsmith et al. (2003) consider the importance of the credibility of the researchers in accessing hard-to-reach groups. They describe methods used to both make the researchers credible in the eyes of a close-knit low-income community with whom they wish to undertake research and also consider how this credibility is maintained throughout the research process. These relationships of credibility and trust are built through immersion at the research site. Researchers spent considerable amounts of time in the community involving themselves in community activities, walking the site, and chatting with groups of residents in a strategy of being there and being seen. These authors note that this immersion not only builds credibility, but also rapport that allows for trust relationships to be established with groups in the community that would be otherwise invisible and difficult to access.

2.4 Not only was rapport built through being there and being seen, Sixsmith et al. (2003) also sought to generate trust through employing researchers who have similar experiences to those who live in the low-income community. Two of the researchers were from working-class backgrounds and had experience of working in socially deprived communities. Similarly, Standing (1998) drew on her own newly acquired identity as a mother during her research and used the contacts she made through baby groups, toddler groups and the like to recruit lone-mothers through a snowballing method. Standing (1998) observes that this meant that an introduction to a lone-mother in the research was always through someone known and a part of the participant's network. Standing (1998) goes on to note that using a strategy of snowballing allowed for some of the power relations between the researcher and the lone-mothers to be broken down and for trustful relationships to be built between researcher and participant.

2.5 Once access to a participant has been achieved building trust continues to be an important part of the researcher participant relationship. Rist (1981) suggests that trust is maintained through acts of reciprocity. Access to participants in research is a privilege cemented through informal exchange between researcher and participant. Feminist researchers pay particular attention to such acts of reciprocity. These acts are characterised as attempts to equalise the relationships of power that exist in the researcher participant relationship. Oakley (1981), for instance, in an investigation of first-time motherhood records the number of questions these mothers asked of her. Three quarters of the 878 questions she was asked related directly to information about pregnancy and caring for a new-born baby because she was known by participants to have particular, and from their perspective valuable, expertise in this area. Oakley understood this to be an illustration of how a researcher offers something in exchange for the information she collects in the interview thereby exchanging one valuable resource for another. In this way, this relationship may also be interpreted as a relationship of reciprocity in which women, experiencing the transition to motherhood and the new identity that entails (Standing, 1998), seek something from the researcher in exchange for giving of their time and answering emotionally taxing questions. Similarly, Goode (2000) reports that she sought to build relationships with participants such as tracking down phone numbers of appropriate agencies, taking a respondent and her child to the circus, and accompanying another respondent to meetings with social services, solicitors, and a court hearing. These Goode (2000) interprets as a courtesy and support to vulnerable people with low self-esteem and a need for support. Researchers involve themselves in activities through which reciprocal relationships are developed. These are necessary to maintain trustful access.

2.6 However, while researchers have identified strategies to maintain trust during research through undertaking extra, frequently unfunded reciprocal work, the potential for these trust relationships to break down when data collection is complete is recognised as a particular problem. Standing (1998) observes that the researcher continues to hold power in the researcher participant relationship. Researchers translate and interpret the private lives of participants and produce public representation of their lives in academic work. Familiarity with research and participants' understanding of the use of the findings from research has the potential to create distrust between researchers and participants (Berg, 1999). Participants, who give of their time and emotional resources in contributing to research activities, may perceive that no tangible benefits accrue from their involvement (Hammersley and Atkinson, 1983). Further, social science researchers may interpret findings about the values and behaviours of low-income groups in ways that contribute to the perpetuation of inequalities of already oppressed groups. As Rist (1981:272 emphasis in the original) observes, 'allowing outsiders to come in is risky'.

2.7 These strategies for generating trust and access to researchers—using peer interviewers, immersion in communities, researchers from similar backgrounds and snowballing—implicitly consider ways to reduce risk to participants in gaining access. We suggest that reciprocity is a strategy necessary to maintain the relationship between researcher and participant during the research process once successful access has been gained. Further, each of the methods discussed here seeks, though empathy, credibility, rapport, and breaking down power relationships, to increase researcher trustworthiness by demonstrating their probity and commitment to those with whom they are doing research.

2.8 In considering how trust relationships are formed Lee Treweek (2002) describes a typology of trust that is increasingly grounded in the person's reflection on the experience of a particular interaction with a person to be trusted. First, trust is built upon the credible accounts of others who have an experience of those to be trusted. Second, trust is produced through connecting new experiences with that which is already familiar. And third, trust building continues and is reinforced through its location within other forms of knowledge. Trust is generated through increasing reflection in which experiences of the unfamiliar are connected and grounded in broader knowledge and understandings. This continuum of action suggests that trust is built through experience. Sztompka (1999) observes that trust is embedded in social action. In part, building trust is anticipatory. Trust is developed in situations where we trust that individuals or institutions will commit actions that will be favourable to our needs and interests. Trust may be betrayed however (Misztal, 1996), or the experiences of the actions of others may not fit meaningfully with one's own pattern of life (Luhmann, 1979:72) suggesting that risk and trust intertwine (Giddens, 1990). These experiences of increased risk lead to distrust, which should not be seen as the antithesis of trust. Distrust arises out of the failure of risk environments to address felt need and enhanced feeling of relative deprivation (Sztompka, 1996) that lead to 'a state of mind that could best be summed up as a feelings of angst and dread' (Giddens, 1990:100).

2.9 It is these feelings of distrust arising out of perceived risk that researchers seeking to access hard to reach groups develop strategies to overcome. In this paper we are particularly interested to understand the ways in which trust relationships are built between researchers and socially excluded individuals and groups, rather than a putative hard-to-reach group. Our concern is to understand the relationships that allow for trustful relationships. In the next section we outline particular features of social exclusion that are likely to lead to distrustful relationships between socially excluded groups and researchers that may impeded access. Furthermore, we explain why gatekeepers may provide a route of potential access and also an opportunity to evaluate the nature of trust relationships important for access.

Considering social exclusion and the importance of gatekeepers in access

3.1 In contemporary life we are increasingly called upon to trust those we do not know (Giddens, 1990;Misztal, 1996). As Sztompka (1999) observes, there is a growing anonymity of those who are responsible for our wellbeing. This is particularly the case for socially excluded individuals and groups who Cattell (2004) characterises as embedded in dense networks among people like themselves and, at the same time, having limited networks with those unlike themselves. This definition of social exclusion makes an important contribution to normative understanding of social exclusion (SEU, 2004). It draws attention to the relationships as well as the outcome indicators that characterise socially excluded populations.

3.2 The population we sought to access experiences multiple deprivations such as high unemployment, low educational attainment, high rates of teenage pregnancy, and high rates of drug addiction. A large part of the population of this low-income estate can be characterised as deprived. Over two-thirds of households in the area live in council accommodation; about one-third of households with dependent children are headed by a single adult. There are high levels of unemployment, particularly amongst young men. This is compounded by large sections of the potentially economically active workforce who are either permanently sick or classified as 'other inactive'. Other indicators, such as the level of car ownership, are characteristic of a low income population. For social theorists (Giddens, 1999) and social policy makers (SEU, 2006), these multiple deprivations conspire together to increase the likelihood of being socially excluded. These indices of social exclusion also, as Macleavy (2006) has argued, provide the framework through which welfare policy for socially excluded people is constructed. Welfare policies are delivered by a range of service providers predominantly in the public and voluntary sectors. These service providers do have access to the networks of socially excluded people in various ways. In the research we report here we sought access through these service providers, acting as gatekeepers, to the networks of socially excluded people on the low-income estate.

3.3 The focus of this paper are the trustful or distrustful relationships between the different gatekeepers and socially excluded participants and the implications of these relationships for access. Little has been written about these relationships in the literature. Miller (1998) and Miller and Bell (2002), observe that gatekeepers may exert leverage on participants in research, inhibiting participants' continued engagement through the power they hold over participants. However, the literature almost exclusively addresses the instrumental role of gatekeepers in mediating access for researchers (Hammersley and Atkinson, 1983). As Cree et al. (2002: 50) observe, '(t)he bottom line is that researchers can get access … only with the co-operation of a number of different 'gatekeepers'; without this, there can be no research'. Consideration is given to a gatekeepers ability to place obstacles and barriers in the way of researchers wishing to access participants (Sixsmith, 2003) or the ways gatekeepers might regulate access. This regulation may include insisting on particular methodological approaches, ethical pre-conditions, or the inclusion of particular participants in the research which accord with the gatekeeper's reality (rather than the researchers') for the research to proceed (George, 1996; Lee, 1993). Lee (1993) considers that gatekeepers insist upon these regulatory activities in physical access to participants to ensure that their interests and activities are not undermined through acting as a gatekeeper. Attention, therefore, centres on how gatekeepers may facilitate or impede research rather than the relationship between gatekeeper and participant.

3.4 Despite these instrumental limitations of gatekeepers, they invariably play a role in accessing members of low-income communities to research (Sixsmith et al, 2003; Cree, 2002; George, 1996). Our understanding of the networks among socially excluded individuals and groups as dense and limited explains why this is so. There are few similarities of experience and no network connections between us, as researchers, and the socially excluded people in the low-income neighbourhood we wished to access. Our access strategy was through the many gatekeepers, most of whom play a role in addressing the welfare of socially excluded people in this neighbourhood. One of the important features of these gatekeepers is that we, as researchers do share a similar educational and experiential position. We suggest that because we have similar levels of education and experience, a common linguistic and cultural capital (Bourdieu, 2002), these facilitate an understanding of the research, its value, and its potential outputs between ourselves and these gatekeepers. These affinities of position allow for a negotiation and recruitment of potential gatekeepers at the research site and through these gatekeepers the most probable route of access to the networks of socially excluded people. This methodological strategy also allowed us to explore the trustful and distrustful relationships between gatekeepers and socially excluded participants and the implications of these for access. Our field work started, therefore, with access to gatekeepers on the low-income estate.

Entering the field

4.1 Snowballing from contacts made in previous research[1] led to recruitment of 39 gatekeepers. Of these, 34 were delivering services in the statutory sector, a further 4 worked for voluntary organisations. We also identified a gatekeeper who was living and working in the community but was not employed by any organisation. We conducted initial interviews with all these gatekeepers in which we discussed their understanding of social exclusion, their relationship with those they perceive to be socially excluded, how they access socially excluded people and the effectiveness of these methods, and how we might access socially excluded people through their gatekeeping. From analyses of these interviews we developed a continuum of gatekeepers from formal to informal. From the 39 gatekeepers interviewed in the first stage of the research we purposively selected 6 gatekeepers that were representative of gatekeepers along this formal—informal continuum (Patton, 1990), which fell into three groups, formal, comprehensive, and informal gatekeepers (Figure 1 provides an overview of the key features of these gatekeepers). We continued to negotiate potential access to socially excluded individuals or groups with these gatekeepers.

Figure 1: A typology of the key features of the gatekeepers—a continuum from formal to informal

Formal gatekeepers

Comprehensive gatekeepers

Informal gatekeepers

4.2 In the formal—informal continuum we characterised formal gatekeepers as those working with the community to achieve a particular end. The role of formal gatekeepers is to implement statutory measures to address social exclusion. Analyses of their interview data suggested that all formal gatekeepers interviewed in this research (n=10) work in multi-disciplinary teams and much of this multi-disciplinary work is focussed on enforcement. Their community involvement is restricted to punitive relationships with those with whom they work. Analyses of the interviews with these gatekeepers demonstrated that they understand their role to be to control, supervise and rehabilitate their 'clients' (their term). Contact with clients rarely happens in the community and more usually occurs in the organisation's offices, the courts, police cells, and enforced community control and service.

4.3 At the informal end of the continuum, we characterised informal gatekeepers as those with limited links to service providers and no mechanisms for referral. They use their own resources to address what they perceive to be the needs of those they work with, in relationships which are an end in themselves. These resources may be time, such as attending court hearings or opening up their house as a safe haven. These gatekeepers are embedded in the community and are seen by many to be working for the good of the community. Through interviews with socially excluded individuals and groups we learnt of other informal gatekeepers within the community who shared similar characteristics to the informal gatekeeper discussed in this paper. However, we did not attempt access through them as we only discovered their existence later on in data collection. Analyses of interactions with one of these gatekeepers, and other gatekeepers who spoke of them, described these informal gatekeepers' relationship with those with whom they work as one based on befriending, supporting, protecting, and even parenting those that they see as vulnerable and frequently misunderstood by service providers.

4.4 Between these two poles are a group of gatekeepers we describe as comprehensive gatekeepers. These gatekeepers are employed within the statutory sector and have a specific remit to address health and social care. In addition to delivering the services for which they are employed, these gatekeepers also implement more comprehensive services, including what they describe as innovative service delivery. They see their role as including wide-ranging referral across service provision. Analyses of the interview data with this group of gatekeepers demonstrated that, invariably, these providers have long-standing relationships with individuals and groups in the community, often spanning generations. They spend considerable amounts of their working lives on the estate meeting and talking with individuals and groups. In this research these comprehensive gatekeepers include health visitors, drugs workers, and voluntary organisations. In the context of health visitors, de la Cuesta (1993) has observed that these workers carry out fringe work, particularly in low-income areas. This fringe work meets deficiencies in the service they deliver and is intended to make service relevant to perceived need through mobilising existing and new resources. These are bespoke services tailored to each person's needs. The examples of this fringe work—setting up services like pregnancy testing, filling out forms, writing letters to the municipal authority—are similar to the activities the gatekeepers deliver on the low income estate in this research.

4.5 Typical of this approach to service delivery, a drug worker who acted as a gatekeeper reflects on the limitations in how many drug services are designed and delivered, and goes on to emphasise the importance of responding to felt need as part of the delivery of the particular service he is responsible for:

'There's a lot of emphasis put on your professional approach. Your corporate image. The way you implement things, the way you actually design your services. Forget who you're actually designing it for, it's more about how you do that.' (T:31)

4.6 This drug worker was attached to a decentralised drug team surveying the extent and kinds of drug use on the low-income estate. Alongside this survey activity, this team were concerned with empowering drug users to take up services. They saw themselves as distinct from therapeutic services. In subsequent interviews with socially excluded individuals we learned that they were seen by these participants as a most likely route to appropriate services to address their health needs and provide support in interactions with housing, social security agencies, and other statutory sector bodies. In common with other gatekeepers who fell into this group these gatekeepers were likely to address particular issues in people's lives such as child care, family support, and drug abuse, yet considered their professional responsibility as being far wider than just addressing selective problems. These gatekeepers referred often to a social model and holistic model in implementing their role and were particularly critical of service delivery approaches that did not take account of the multiple needs of those with whom they worked.

Gaining access to socially excluded people through comprehensive gatekeepers—a relationship of trust

5.1 It was through this third group of comprehensive gatekeepers that we were able to access socially excluded individuals and groups. The interviews with socially excluded individuals and groups contain numerous examples of the value of, and indeed peoples' dependency on the services of those who are willing to address their needs in a more comprehensive way. In addition to traditional, or formal services, comprehensive services range far wider to include, for example, rubbish clearance, trips, repairs, transport provision, support in interactions with, and referral to other health and social care providers, and assistance with form filling. S, a mother of eight and life-time resident of the estate, for instance, talks about getting rubbish cleared from her back garden:
'Yeah, I mean it went up into a point where all gardens backed on to each other so the top part used to be like a little rubbish tip where everybody threw. We got told to bin-bag it. I mean how do you bin-bag an old washer and double mattresses and stuff like that. And it was actually Sa, well not actually Sa but the other, it'll be our Lo, with the baby, the other, yeah, my other, my other health visitor sent a letter in (to the Council) and got them moved.' (S:27)

5.2 This exemplar of the activities of these gatekeepers not only points to the comprehensive nature of their activities, but also to a long-term relationship built up between gatekeeper and socially excluded individuals. These gatekeepers talked about knowing the individuals we wanted to access over generations, drawing on their own and their colleagues' experiences of working with them. Similarly, S, like other participants, recalls interactions that are consistent and comprehensive over time. The accounts of socially excluded people emphasise moments where professionals have mediated access to particular needs.

5.3 It is through these long-term relationships, in which such needs in people's day-to-day lives are addressed, that trust is developed. Ha, a long-term resident of the estate, describes the characteristics of her relationship with the gatekeeper which provides the possibility for researchers to achieve access to her and, at the same time, explains when access can not be gained:

'I mean, even Sa's got a relationship with me cos she was my kids' health visitor, she's got a rapport, a good rapport with me. So yeah, that's the only way you could do it, someone who's got very close contact, very close knit with somebody, otherwise you can't just sort of out right. Advertise yourself and nobody'd put their name down like. I certainly wouldn't have done. It's cos Sa, I know Sa, I said yes. Its because I knew Sa and I trust Sa, it's the trust you see. If Sa put it forward, yeah, I trust Sa, I will let somebody else, somebody into my home. I mean, you could be anybody, but Sa put you forward. So yeah, I'll let you in my home. I volunteered basically yeah, I would do it for Sa, it's the relationship people have. It's the trust and the faith you have in someone that, that you know allow you to say, determined my yes, determined my answer.' (Ha:73-74)

5.4 Ha emphasises trust and faith as important in the relationship with the gatekeeper Sa that facilitated the researcher's access. Indeed, from Ha's account, it is clear she would not be accessible to researchers in the absence of the relationship with this comprehensive gatekeeper. Across the interviews respondents repeatedly emphasise the trustworthiness of the comprehensive gatekeeper through whom we gained access. In addition, like Ha, they emphasise the 'relationship' they have with this gatekeeper. This is a relationship where 'they would do it (be interviewed) for …'. The relationship between socially excluded individual and comprehensive gatekeeper is described as building upon a long-term exchange during which they have experienced actions by this gatekeeper that lead to relationships of trust because they have directly gained benefits from their interaction with the service provider. Although a long-term relationship in which the comprehensive gatekeeper was well known, participants did not, however, consider these relationships to be friendships, as this exchange between husband (B) and wife (D) suggests:

B: I think if you've known them a while you can talk more, you know what I mean, you can say what you feel you know what I mean, because, well I suppose it's like having a mate in't house, you know what I mean, like having a friend. Because I think they have become good mates over the years haven't you?
D: Mmm.
B: Well not mates, but something there, because she knows D, in and out and she knows her, so I suppose it's easier to talk to somebody like that. (B&D:81)

5.5 From these accounts we concluded that participants granted access because we were introduced by a gatekeeper with whom they had a relationship based on trust and faith. This relationship was not one of friendship; rather, the accounts from socially excluded people highlight the ways in which these gatekeepers are effective in mediating access to services to address day-to-day needs. These gatekeepers see this role as integral to their work. One consequence of their comprehensive service provision is that these gatekeepers have an increasingly trustful relationship built up over time in the networks of socially excluded individuals and groups.

Not gaining access through formal gatekeepers—a relationship of distrust

6.1 Our data suggested that participants use their experiences of gatekeepers and those of their limited networks to carefully evaluate risk when making a decision to involve themselves in our study. Our analyses of the role of the comprehensive gatekeeper as a facilitator of access were supported through interrogation of the features of gatekeepers through whom we did not gain access. Gatekeepers who could not provide access also recognised the importance of the character of the relationship between gatekeeper and participant in facilitating a researcher's access. Field notes from an unrecorded interview with a formal gatekeeper, the head of the Anti-Social Behaviour Unit, recorded:
'Pa suggested that because of the Anti-Social Behaviour Unit's enforcing role with the young men who have anti-social behaviour orders, they would not receive a positive response from them if they acted as gatekeepers. He talked about his organisation having the wrong sort of relationship with them to introduce us to them and that if they tried they would just tell them to "get lost". He suggested that an agency like the Youth Offending Team, that had more of a supportive role might produce better results. He did say that he could introduce us to residents who had for example complained to the Anti-Social Behaviour Unit or people who were "regular members of the community".' (FN:116)

6.2 The Anti-Social Behaviour Unit deals with particular deviant behaviours of social exclusion such as vandalism, gang activities, bullying, and nuisance through treatments such as the ASBO. Repeatedly, socially excluded participants who we did access in the research, including young men with ASBO’s, described the activities of this agency as not addressing their needs. There is a deep rooted distrust and perception of increased risk among those we interviewed. Those who worked for these and similar agencies recognised this as well. As H, the head of the Youth Offending Team observes:

'…there are other youngsters, and it doesn't matter if you're visiting them at home, you wouldn't find them. They won't engage with you, until they're forced to because you have to take them back to court and they end up back in court again, and 'eventually' (emphasised) they do wear down and they do engage with us.' (H:43)

6.3 Both Pa and H recognise the impossibility of access through their organisations to those they consider to be socially excluded. The difficulties in accessing such people through these formal gatekeepers are, in part, explained by the role of these organisations in addressing social exclusion. Access is not possible because these gatekeepers are distrusted by socially excluded people with whom we were trying to gain access. Furthermore, and emphasising the observation made earlier, these gatekeepers do not have long-term trustful relationships with those we wished to access. As H notes, to gain access, 'you have somebody who is community based who has got the trust of some of the, err, the parents and the young people' (H:65).

6.4 The socially excluded participants in the research not only made observations about the inappropriateness of the services managed and delivered by these gatekeepers but were deeply suspicious of the methods they use, considering these methods to support the interests of the organisation rather than their own. This perception was further supported by a number of our own experiences with these gatekeepers. For example, in the context of ethical procedures, formal gatekeepers insisted on formal ethical procedures in the research, 'for the safety of their organisations, their workers, and the participants' (JW:47).

6.5 In a revealing example early in the research we accessed a participant through a comprehensive gatekeeper. We were refused access however while the researcher was in the participant's home, after the researcher had discussed the information sheet and asked the participant to sign a consent form. Analysis of the interaction between the researcher and participant suggested that the participant's refusal was based on the implied link between the researcher and other (unidentified) statutory organisations. Our approach to gaining written ethical approval from participants echoed procedures used by organisations that control and supervise, where the experience of the filling out of forms and, subsequently, such association between us and these organisations resulted in a perception that participation in the research process may lead to punitive outcomes. Research can mimic access to service provision that is perceived to be risky. Despite the lengthy and extremely sensitive process through which the fieldwork had sought to establish trust, such association removed any trust and closed the possibility for access. Our approach to gaining consent in the research was subsequently modified in the light of this experience of failed access (Hughes et al., 2004). Further, this example emphasises the fragile relationships of trust that exist between researcher and participant, even where a trustworthy gatekeeper has mediated access.

Not gaining access through informal gatekeepers—reinforcing distrustful networks

7.1 The fragility of the relationship of trust between researcher and participant is a feature highlighted in our inability to gain access through the one informal gatekeeper we recruited to the research. In common with other informal gatekeepers identified late in the research, this community based worker has no formal ties with statutory services. L's role is one of an unpaid community worker. She is widely known to service providers, but, although L does work alongside these service providers her approach is often considered inappropriate by service providers working in the statutory and voluntary sectors. L sent a proposal for the future use of a community centre in the area to the head of youth services but received a response saying 'they could not entertain the proposal because it was not written up as a business proposal' (10:35). This L sees as a good example of her position outside the system, not as a social worker, without a qualification, and not following text book methods in an approach which seeks to act to protect young people in the community, whom she sees as particularly vulnerable. As L notes:
'I treat them kids out there like my own kids. I show them what I would show my own kids. I stick by them with all their wrongs and downfalls…It's the worst kids I want. I want to show people what they're really like. They have a right to express themselves (10:78)'.

7.2 When asked about why she feels that access to socially excluded young men in our research might be difficult through the mediation of her gatekeeping, L emphasised the observation already made about the appropriateness of services to these socially excluded people. Drawing on her interactions with these young people and reflecting their perceptions of services provided by outside agencies, L observed that:

'These are things they are not used to. These are things they're not expecting someone to do. They're not expecting someone to help. They've gone beyond that.' (L:119).

7.3 L did not allow our conversations to be tape-recorded in this research. The field notes from the many interactions with L throughout the research record L's talk about her distrust of services and organisations. The following is an extract from the field notes after a telephone conversation with L:

'I was concerned before the phone-call that she had become slightly suspicious or 'cold' towards our study and me. She is suspicious of services and institutions in regards to their treatment of the young people she works with, but she also has a general distrust of organisations. …She has stated recently that institutional time and their time are very different and when service providers say they are going to do something then this means in the intangible future. I likewise am aware that our study will not be able to offer immediate and real benefits for those experiencing massive problems connected with poverty and social exclusion. I am therefore concerned that L would think that we are exploiting her and these young men who she works with. In a way I suppose we are. We are taking something from their lives which might not be of direct benefit' (10:95).

7.4 This informal gatekeeper was sceptical about the benefits of the research to this vulnerable group. Subsequent analysis of our relationship with L, and the consequences of her refusal to grant us access to these young men, led us to consider how L could be seen as reinforcing the inward-lookingness of the close-knit networks among these young men. In discussions with the researcher she clearly reframed our study as likely to play out in the same way as services the young men may have accessed which were unable to deliver services to meet appropriate needs.

7.5 For L, the ostensible value of this research for the young men with whom she worked was questioned during her conversations with the fieldworker and, finally, it became clear that she would not support or facilitate access to the young men with whom she worked. In this way, this informal gatekeeper blocked our access to the young men as an extension of her protective role towards them. Our field notes drawn from chance conversations with some of these young men, with other informal service providers working in this location who know these young men, and interviews with two of these young men whom we accessed through their family, who in turn was accessed through a comprehensive gatekeeper, suggest that they form and participate in very inward-looking groups. Indeed, they were described by others on this estate as 'having their own language'.

7.6 Towards the end of our field work we found two further informal gatekeepers in the low-income estate. Like L, these gatekeepers worked with close knit groups outside of formal structures. Also, in common with L, these gatekeepers continued to interact with us throughout the research and to question the potential value of the research to those with whom they work. Despite continued attempts that lasted until we brought our fieldwork phase to a planned close, access was denied through informal gatekeepers who described themselves as having the absolute trust of those with whom they work.

Social exclusion and trust in the gatekeeper participant relationship—understanding relationships of power

8.1 Access to socially excluded participants was closely allied to the fringe work (de la Cuesta, 1993) of the comprehensive gatekeepers. Fringe work is an activity that both addresses day-to-day needs among socially excluded people and, simultaneously, produces the conditions for particular kinds of social relationships that provide the possibility for researcher access. Participants' accounts of their social exclusion provide accounts of how the dynamics of social exclusion are produced, reproduced and occasionally disrupted. We learnt from those we interviewed that these processes of social exclusion include three important dimensions. First, all participants identified the importance of relational networks within the low-income estate that are used to address everyday problems and everyday needs. These might include child-care, both formal and informal economic activity, and even being able to walk down the street in safety. Secondly, participants' experiences of social exclusion are framed in a described social context in which relationships with each other and with service providers, predominantly in the public sector, facilitate or impede access to basic needs such as appropriate health care, housing maintenance, social-security benefits, and domestic waste clearance. The significance and centrality of these relationships to peoples' lives comprise the third important finding. The socially excluded people we interviewed described moments of powerlessness and constrained powerfulness in trying to address their day-to-day needs. Powerlessness was expressed in the accounts of those we interviewed as an inability to effect change of their life circumstances through direct mediation with (most often) statutory agencies such as health services, schools, and local authority and with employers. However, in some circumstances, individuals were able exert constrained powerfulness through a relationship with a service providers to effect change.

8.2 Unlike comprehensive gatekeepers, formal gatekeepers demonstrate discursive and institutional practices in which they objectify those who are socially excluded and whom they talk of as clients. Analyses of interview data suggests that socially excluded people are distrustful and hostile, even frightened of these formal gatekeepers and the power they have to control their lives. Indeed, our participants relate a number of anecdotes in which formal gatekeepers act to intervene punitively in these participants' lives during chance encounters. Formal gatekeepers recognise that their access is most likely to be secured through the authority of the court order and police. Throughout our study, socially excluded people related experiences, exclusively negatively and layered with a dark humour, of their interactions with these gatekeepers. We would describe the relationship of power between gatekeeper and those who are socially excluded as vertical, in which power is retained by the gatekeeper. The power exercised by those who are socially excluded is based on their ability to evade and deceive. We did not hear any accounts in which participants were able to successfully resist these gatekeepers in their service provider role over a long period of time. In terms of the discussion of access in this paper, an important consequence of this relationship of power is that researchers are unable to access participants through their referral because such relationships are considered risky and produce only distrust between gatekeeper and socially excluded individuals and groups.

8.3 Informal gatekeepers have a very different relationship with those we sought to access. They understood the aims of our research, indeed L, with whom we had a good relationship, presented an overview of the work she does with young men to some of our post-graduate students during the research. Nonetheless, these informal gatekeepers position themselves as invariably in opposition to formal gatekeepers, recognising and reacting to the inappropriateness of the services that are offered by these gatekeepers. Their relationships are that of protector, trying to shield those they care for from further experiences of a system perceived as damaging, even pathological in perpetuating the social exclusion of those with whom they work.

8.4 We would suggest that the behaviour of informal gatekeepers serves to reinforce this inward-lookingness but recognise that this is in response to extreme levels of vulnerability, not only to the formal service providers in this area, but also to other residents on the estate, and members of their own and other groups. In terms of the discussion of access in this paper, we would suggest it is not possible to access these individuals and groups through informal gatekeepers with whom they have such protective relationships, which are used to convey distrust based on an assumption that researchers can do harm.

8.5 Analyses of participants' descriptions of the relationships they have with comprehensive gatekeepers, clearly show that trustworthiness and mutual valuing are central, and that these are often developed over a long period of sustained delivery of comprehensive service provision. However, as we have shown, comprehensive gatekeepers are not seen as friends or members of the networks of socially excluded people. Nonetheless trust characterises these relationships. The services provided by these gatekeepers are crucial to these participants' lives, and their attempts to address their day-to-day living. As described, these gatekeepers mediate between participants and service providers in delivering comprehensive services. In this way, it would be easy to describe the power differential between the service providers and socially excluded people as vertical. However, close analyses of our data from both comprehensive gatekeepers and participants suggest that these relationships become progressively horizontal. They are based on a mutuality of understanding of felt need. We have shown how the mediation by these comprehensive gatekeepers results in the delivery of services deemed to be of value to those who are socially excluded, which allows for trustful relationships between gatekeeper and participant that makes access by researchers possible.

Conclusion

9.1 Discussions of access through gatekeepers in the methodological literature have focussed on the instrumental relationship between researcher and gatekeeper. In this paper we have sought to illuminate the range of relationships between the gatekeepers accessed for the study and the participants and the implications of this relationship to allow access to socially excluded individuals and groups by researchers. Furthermore, rather than considering strategies of access as trust-building activities by researchers to facilitate access to vulnerable and marginalised groups, we have considered how relationships of trust are built between gatekeeper and participant in low-income communities, and that this trust flows out, to some extent, into the relationship between researcher and participant.

9.2 We have alluded to, but not discussed in detail the relationship between researcher and gatekeeper. Undoubtedly our relationship was strongest with comprehensive gatekeepers. Their social model of health delivery, with its concern to address broad need rather than merely the outcomes of social exclusion, was closest to our own. We complied with formal gatekeepers insistence that we use formal ethical procedures, such as written informed consent, and learnt from using these procedures that they can act as a strong barrier in access to socially excluded groups. Formal ethical procedures in the field reproduce organisational procedures that are distrusted by those vulnerable and excluded individuals we sought to access. Additionally, these formal procedures failed to address the need for ongoing consent necessitated by prolonged immersion in the field essential to furthering relationships of trust (Hughes, et al, 2004). We maintained a good relationship with informal gatekeepers throughout the research. These enterprising individuals working on this low-income estate were distrustful of our research and the benefits that would accrue to those they represent.

9.3 Importantly, a trustful relationship between gatekeeper and participant provides the necessary precondition for the researcher and participant in the research to come to know each other. An understanding of the conditions for a trustful relationship are closely interweaved with an understanding of the processes of social exclusion among socially excluded individuals. The gatekeepers who provide access are those that come into contact with the networks of socially excluded individuals and groups and build up a relationship of trust with members of these networks through addressing the broader social needs of socially excluded people with whom they work, invariably over the long-term. In the context of broader debates of power differential and inequities between researcher and participant in research we would argue that the gatekeeper's position between researcher and participant should not only be analysed as an exercise of power in the research process as Miller and Bell (2002) and Miller (1998) have done, but must be seen in a wider context in which gatekeepers are able to facilitate access, impede access, or are unable to arrange access because of the relationship of trust or distrust they have with those with whom we want to do research. Analyses which seek to do so are therefore able to recognise and understand the constraints and possibilities for those who are socially excluded to resist access and thereby provide a more nuanced picture of their lives and of the relationships in which they are embedded.

9.4 Our research emphasises the importance of social action to understand the possibilities for trustful relationships between socially excluded individuals and groups and researchers (Sztompka, 1999). As Giddens (1990) and Misztal (1996) observe, evaluations of risk are made and acted upon in decisions about building a trustful relationship. Comprehensive gatekeepers, through their fringe work, build trust because these activities fit meaningfully with the day-to-day needs of those who are socially excluded (Luhmann, 1979). Distrust is informed by the negative experiences of interacting with gatekeepers, who may be feared and dreaded (Sztompka, 1996; Giddens, 1990). However, in making evaluations of trustworthiness socially excluded individuals are constrained through the small and dense networks that characterise social exclusion. Building trust is based first and foremost on more proximal experience of the trustworthiness of the gatekeeper who is referring the researcher to the socially excluded individual or group. The activities of a gatekeeper and the organisation they represent might be considered too risky because the experiences of their activities are of heightened risk through control and punishment. By association the researcher is also considered to be risky. A gatekeeper may also reinforce perceptions of risk, reinforcing the inward-looking networks of vulnerable socially excluded individuals that also excludes researchers.

9.5 Trust is built through long-term reciprocal relationships between gatekeeper and socially excluded individual or group. The features of empathy, credibility, rapport, and the slow evening out of power relationships that have been identified by researchers as arising from using particular strategies to access hard-to-reach groups are all displayed in the reciprocal relationship between comprehensive gatekeepers and participants. Nonetheless, researchers who have gained access through the referral of a trusted gatekeeper may still be refused access because they use approaches similar to those used by formal and statutory organisations that are not trusted.

9.6 The familiar that can be trusted (Lee Treweek, 2002) emerged as proscribed within narrow boundaries for those we interviewed, while their experiences arising out of interactions with the formal gatekeepers operate as a strong countervailing force leading to an increased perception of risk in relationships with those they perceive to resemble formal gatekeepers, and therefore distrust. The experiential framework within which trust can be developed for our participants is constrained and limited through their experiences of vulnerability and social exclusion. These limits to trust mean that researcher access to socially excluded individuals and groups will always be extremely difficult, time consuming, and resource intensive. Researchers can, however, increase the likelihood of access through trusted comprehensive gatekeepers who have spent considerable time addressing the day-to-day needs of socially excluded individuals and groups they wish to access.

Notes

1 This research was commissioned by a Primary Care Trust, the reference has been omitted to ensure anonymity

Acknowledgements

This paper is an output from Developing methodological strategies to recruit and research socially excluded groups ESRC Research Methods Programme Grant H333250001.

Our thanks to the participants in this study and to Graham Crow, Andrew Clark, Inna Kotchetkova, Susan Eley, and the anonymous reviewers for their comments on drafts of this paper. Earlier versions of the paper were presented at the Family Life and Kinship Group (FLaG) The University of Leeds, The Government Seminar Series. Her Majesty's Treasury London and The Innovative Methods Seminar Series at the Scottish Office. The usual disclaimers apply.

References

BERG, J.A. (1999) 'Gaining access to under researched populations in women's health research' Health Care for Women International 20: 237-43.

BOURDIEU, P. 'Understanding' in P. Bourdieu et al. The weight of the world: social suffering in contemporary society. Cambridge: Polity Press, 2004.

CATTELL, V. (2004) 'Social Networks As Mediators Between the Harsh Circumstances of People's Lives, and the Lived Experience of Health and Wellbeing' in B Phillipson, G. Allan, and D. Morgan (editors) Social networks and social exclusion: sociological and policy perspectives. Aldershot: Ashgate.

CREE, V. KAY, H. and TISDALL. K. (2002) Research with children: sharing the dilemmas. Child & Family Social Work 7.1:47-56

DE LA CUESTA, C. (1993) 'Fringe work: peripheral work in health visiting' Sociology of Health & Illness 15.5: 665-82.

ELLIOTT, E. WATSON, A.J. and HARRIES, U. (2002) 'Harnessing expertise: involving peer interviewers in qualitative research with hard-to-reach populations' Health Expectations 5: 172-78.

GEORGE, A. (1996) 'Methodological Issues in the Ethnography of Sexuality: Experiences From Bombay' in K de Koning and M Martin Participatory research in health: issues and experiences. London: Sage.

GIDDENS, A. (1991) The Consequences of Modernity. Cambridge: Polity.

GIDDENS, A. (1999) The Third Way : the Renewal of Social Democracy. Malden: Polity Press.

GOODE, S. (2000) 'Researching a Hard-To-Access and Vulnerable Population: Some Considerations On Researching Drug and Alcohol-Using Mothers'. Sociological Research Online 5[1], http://www.socresonline.org.uk/5-1/goode.html

HAMMERSLEY, M. ATKINSON, P. (1983) Ethnography: Principles and Practice. London: Tavistock.

HUGHES, K. EMMEL, N. GREENHALGH, G. and SALES, A. (2004) 'Access to the socially excluded: the constitutive power of ethical practices in the researcher/gate-keeper/participant encounter.' ESRC Methods Programme Research Methods Festival Oxford.2nd July 2004 (2004).

KUEBLER, D. and HAUSSER, D. (1997) 'The Swiss hidden population study: practical and methodological aspects of data collection by privileged access interviewers' Addiction 92.3: 325-34.

LEE, R. (1993) Doing Research on Sensitive Topics. London: Sage.

LEE-TREWEEK, G. (2002) 'Trust in complementary medicine: the case of cranial osteopathy' The Sociological Review 50.1: 48-68.

LUHMANN, N. (1979) Trust and Power Chichester: Wiley.

MACLEAVY, J. (2006) 'The language of politics and the politics of language: unpacking 'social exclusion' in New Labour policy' Space and Polity 10.1: 87-98.

MISZTAL, B.A. (1996) Trust in Modern Societies. Cambridge: Polity Press.

MILLER, T. (1998) 'Shifting Layers of Professional, Lay and Personal Narratives: Longitudinal Childbirth Research' in J. Ribbens and R. Edwards (editors) Feminist dilemmas in qualitative research: public knowledge and private lives. London: Sage.

MILLER, T. and BELL, L. (2002) 'Consenting to What? Issues of Access, Gate-Keeping and 'Informed' Consent' in Mauthner, M. et al. (editors) Ethics in qualitative research. London: Sage.

OAKLEY, A. (1981) 'Interviewing Women: a Contradiction in Terms' in H, Roberts. (editor) Doing feminist research. London: Routledge & Kegan Paul.

PATTON, M.Q. (2002) Qualitative Evaluation and Research Methods. London: Sage.

RIST, R.C. (1981) 'On What We Know (or Think We Do): Gatekeeping and the Social Control of Knowledge' in T.S. Popkewitz and B.R. Tabchnick (editors) The study of schooling: field based methodologies in educational research and evaluation. New York: Praeger.

SEU. (2004) Tackling Social Exclusion: Taking stock and looking to the future: Emerging findings. London, Office of the Deputy Prime Minister.

SEU. (2006) Transition: Young Adults With Complex Needs. Office of the Deputy Prime Minister. London.: A Social Exclusion Unit final report, 2006.

SIXSMITH, J. BONEHAM, M. and GOLDRING, J.E. (2003) 'Accessing the community: gaining insider perspectives from the outside' Qualitative Health Research 13.4: 578-89.

STANDING, K. (1998) 'Writing the Voices of the Less Powerful' in J. Ribbens and R. Edwards (editors) Feminist dilemmas in qualitative research: public knowledge and private lives. London: Sage.

SZTOMPKA, P. (1999) Trust: a Sociological Theory. Cambridge: Cambridge University Press.

SZTOMPKA, P. (1996) 'Trust and Emerging Democracy: Lessons from Poland' International Sociology 11.1: 37-62.