Service user involvement in research has triggered debate about epistemology, power relationships between researchers and participants, ethics and the validity of research findings. The social work profession encourages respect for service users and promotion of their rights—values which arguably should be reflected in social work research. This paper presents a case study of service user involvement in a government-funded adoption research project. The research team included a birth relative consultant group and the paper discusses both the process and outcomes of their involvement. The benefits and costs of service user involvement are highlighted. Reflections from the consultants about their experiences of being involved in the research are included. The discussion positions the project in relation to existing theories and models of service user involvement.
The development of user involvement in research is relevant to applied social research and research on public services. There has been an international interest in service user involvement in research (Callard and Rose, 2012). The social work profession encourages respect for service users, promotion of their rights and promoting equal opportunities. An ethical argument can be made for the promotion of service user involvement in social work research, echoing the commitment to service user involvement in social work practice and training. This paper considers the benefits and challenges of service user involvement in relation to a specific social work research project.
The ‘Researching Adoption Support’ project was government-funded and aimed to map, cost and evaluate services to support the birth relatives of children adopted from care, and services to support post-adoption contact in complex cases (www.adoptionresearchinitiative.org.uk). Service user consultants were recruited to the project. They included both birth relatives of adopted children and adoptive parents. The paper focuses on the role of the birth relative consultants and discusses both the process of establishing and working with the service user consultant group and the impact of the consultants on the research. Whilst the current paper is written by academic researchers, it includes reflections from the consultants about their experiences of involvement in the research. The discussion will position the project in relation to existing theories and models of service user involvement.
The context for service user involvement in research
In the UK, service user involvement is a social policy development reflecting a participation agenda in public services which was part of Labour's modernisation agenda in the 1990s (Warren and Cook, 2005). Research Governance in the National Health Service calls for active involvement of service users and carers in research (Department of Health, 2001). From 1996, there has been an advisory group on consumer involvement in NHS research, whose remit was widened to take in social care research in 2003 when it was renamed ‘INVOLVE’ (Telford and Faulkner, 2004). User involvement has been referred to as the ‘new mantra’ in public services (Cowden and Singh, 2007). Cowden and Singh suggest that there are two opposing stories to be told—one involving a ground-up critique of welfare services that marginalised and excluded certain groups, the other a top-down managerial co-optation of such critiques involving minimal attention to power relationships.
In addition to the policy mandate, social work researchers are arguably bound by a professional commitment to abide by the values of the social work profession (Butler, 2002). Dominelli (2005, p. 230) argues that social work research can distinguish itself from other research in similar areas by a more egalitarian relationship between researcher and researched and a holistic engagement with the problems they are investigating. The empowerment of service users and pursuit of social justice are reflected in the code of ethics for social work research of the Joint University Council Social Work Education Committee (JUC SWEC website, www.juc.ac.uk/swec-research-code.html code of ethics of JUC SWEC). Service users' and carers' perspectives are a requirement in UK social work education (Department of Health, 2002). Given this context, it may seem surprising that much of the existing empirical literature on service user involvement in research comes from a health perspective (McLaughlin, 2010).
Beresford (2009) distinguishes between democratic and a managerialist/consumerist approaches to service user involvement. The former stems from the work of service user organisations and has explicitly ethical and political aims, to reshape the research process (Turner and Beresford, 2005; Beresford, 2009). The latter is associated with providers and involves consultation without challenging existing power relationships between academic researchers and service users. There are similarities between the democratic model and the long tradition of participatory research, such as in community development. Participatory research methodologies place an emphasis on asking who determines the research questions and who generates, analyses, represents and acts on the findings (Cornwall and Jewkes, 1995). Within the participatory action research tradition, the researcher is seen as a facilitator who can empower people to develop their own knowledge and take action. As participatory research has developed, challenges to this model have been identified, including: the ‘community’ not wishing to take part; the time-consuming nature of research; task exhaustion; and the realisation that ‘communities’ are not homogenous and integrated, but differentiated in terms such as wealth, age, gender, ethnicity and hence power (Cornwall and Jewkes, 1995).
The democratic model is explicit in the importance it gives to experiential knowledge and regards knowledge as inherently political (Beresford, 2007). Instead of espousing a model of the researcher as impartial and neutral, emancipatory research argues that, where there is less distance between experience and its interpretation, the resulting research is likely to be more accurate (Beresford, 2003). Emancipatory research shares features with feminist standpoint theory (Telford and Faulkner, 2004; Beresford, 2007; McLaughlin, 2010). Feminism has a tradition of using women's experience to combat the supposed neutrality of knowledge claims. Debates about essentialism within feminism questioned whether there was a universal feminist voice which could encompass the plurality of experiences of women, in particular questioning the dominance of white middle-class Western women representing all women (hooks, 1982; Narayan, 1989). Debates about service user involvement in research mirror these concerns. In the context of service user involvement, experience is a crucial concept, as suggested by the term ‘expert by experience’ adopted by some mental health user organisations (Telford and Faulkner, 2004). Being clear about what experience is shared draws attention not only to similarities, but also to diversity. There has been discussion about how representative service users taking part in research may be of the larger constituency of service users. It is important not to assume that service user research is inherently inclusive (McLaughlin, 2010; Essien, 2009; Fenge, 2010).
Frameworks for assessing the extent of user participation in research offer a way of moving beyond a polarised debate to assess the degree to which individual projects involve services users at different stages of the research process. In an influential report for INVOLVE, Hanley et al. (2003) distinguish between three levels of service user involvement: consultation, collaboration and control, derived from Arnstein's ladder of participation (Arnstein, 1969). Consultation involves asking service users for their views, but with no requirement to act upon them. Collaboration involves an on-going partnership between service users and academic researchers, where control over the research is shared. User control occurs where service users make decisions and direct the research, although professional researchers may still be involved. An additional level of involvement has been added to the original typology by Sweeney and Morgan (2009). They propose a fourth level: ‘contribution’, to sit between that of consultation and collaboration and to reflect existing practices. Contribution refers to a process where researchers do more than asking service users' opinions, actively involving them throughout the research process, but without sharing decision-making powers (Sweeney and Morgan, 2009, p. 28). There is a debate (see Nolan et al. (2007) for a useful summary) about whether such typologies should be viewed as a hierarchy with user-controlled research being considered the pinnacle (Turner and Beresford, 2005) or whether different approaches are suitable for different projects (Hanley et al., 2003).
Epistemological and ethical arguments can be made for service user involvement. It is also important to ask what difference service user involvement makes, if any, to research design and findings. It has been argued that service users can help to ensure that research questions are relevant to service users, and hence pertinent for service development (Rhodes et al., 2002; Staley and Minogue, 2006; Wallcraft and Nettle, 2009; Morrow et al., 2010). The involvement of service users can be argued to offer greater validity to research by improving research design. Recruitment strategies can be optimised—particularly valuable where researchers are working with marginalised groups (Fisher, 2002; McLaughlin, 2010; Wallcraft and Nettle, 2009). Service users may be more sensitive to ethical issues in research (Barber et al., 2011). In relation to data collection, it has been suggested that participants may be more open in their responses to service user interviewers (Fisher, 2002). Some studies comment on the role of service users in analysing data, such as suggesting that service users have identified themes in data that the academic researcher would have overlooked (Fisher, 2002), and highlighted findings which are particularly relevant to service users (Ross et al., 2005). Advantages of involving service users in dissemination include making the research more accessible (Barber et al., 2011).
The challenges of service user involvement
Service user involvement, whilst potentially offering benefits, also presents many challenges. One such challenge is that of possible differences between the academics and service users involved in the project, about the aims and methods of the research and the roles of various stakeholders in the research process, including the funders (Telford and Faulkner, 2004). Some researchers may feel anxious about losing control over the research and it is important that there is negotiation and agreement about roles and research processes from the outset (Faulkner, 2009). Even where roles are clear, tensions may emerge as the research progresses. For example, Rhodes et al. (2002) noted a temptation to see their service user advisory group as another source of data rather than as having a different function to the research participants. Where service users are motivated to improve services, there may be frustration about the realistic possibilities and timescales for change (Ross et al., 2005).
Comprehensive good-practice guidelines for involving service users have been suggested by Schrank and Wallcraft (2009). Key principles include consultation about roles and involvement of service users from the start of the project; careful planning that takes account of individual needs; the provision of adequate support including emotional, practical and research-related support; clear communication avoiding jargon; realistic assessment of time and budget required for meaningful involvement; and a clear method of payment that adheres to employment law. The importance of nurturing and sustaining trusting relationships has been underlined (Fenge, 2010; Barber et al., 2011).
Lack of evaluation is a limitation in assessing the value of service user involvement (Minogue, 2009), although guidelines for assessment have been suggested (Wright et al., 2010). A distinction can be made between the quality of involvement with a focus on process and the impact of involvement with a focus on outcomes (Morrow et al., 2010). It would seem that both are important, indeed interlinked, since, however good the process of involvement, it would be in danger of being tokenistic if there were no influence on outcomes. This paper aims to give an account of both process and outcome of service user involvement in one project offered as a case study, in order to consider the costs and benefits of service user involvement. By trying to provide a transparent account of both process and impact, the twin dangers that McLaughlin (2010) argues may befall service user involvement—that of being reduced to a tokenistic exercise or viewed as a panacea—can be minimised.
Service user involvement in the researching adoption support studies: process
The research was commissioned by the Department for Children, Schools and Families (now Department for Education) in response to a government tender. Part of the application involved a commitment to service user participation in the research process. Service user involvement purely motivated by policy and funding considerations could be potentially tokenistic (Wallcraft and Nettle, 2009). The involvement of service users was driven by the requirements of the tender, but also by the view that service users offered a unique standpoint that could improve the research, making it more valid and relevant to practice. From an ethical perspective, it was important to the research team that social work research should be undertaken with similar ethical standards to those governing social work practice. The final plan involved two service user advisory groups (adoptive parents and birth relatives) comprising twenty-four service users (fourteen birth relatives and ten adoptive parents). The service user consultant groups met separately from the project advisory group. This allowed researchers to adapt activities according to each group's needs and created an informal atmosphere that helped service users to feel confident to contribute (Rhodes et al., 2002).
The term ‘service user consultants’ was used to refer to the advisory group members. There is not a consensus in the literature over terminology, with various terms being used: ‘public involvement’, ‘consumer’, ‘client’, ‘service user’, ‘survivor’. Terminology positions people in different ways. ‘Consumer’ tends to be used in a health context and has been met with some resistance by those who may be forced to use services against their will (Telford and Faulkner, 2004). ‘Service user’ tends to mean people who use, or are eligible to use, health and social care services (Beresford, 2005), although, for some, ‘user’ is associated with substance misuse (Boote et al., 2002).
The term ‘service user consultant’ recognised that the experience of using or being eligible to use adoption support services was the relevant experience that researchers wished to access for the purposes of the project. The term ‘consultant’ aimed to respect their specialist expertise in relation to the project. This paper is focused on the birth relative advisory group, since the birth relative consultants were a particularly disempowered group with many diverse needs and the work required a particular degree of commitment, creativity and sensitivity from all involved.
Setting up the birth relative advisory group
Birth relatives whose children are compulsorily adopted are a vulnerable group, with high levels of difficulties including mental health problems, substance misuse, learning disabilities, criminal records, poverty and unemployment (Neil, 2000). It was anticipated that, just as it would be difficult to recruit birth relatives as participants in the study, it would also be difficult to recruit them as service user consultants to the project. The aim was to recruit a diverse group of birth relatives, including those without any previous experience of research. The key criterion was motivation and interest.
An adoption support agency offered to bring together a group of birth parents to talk about becoming birth relative consultants. The academic researchers did not have contact with them beforehand. The meeting proved difficult; most of the birth relatives talked very emotionally about their experiences of having a child adopted. When further meetings were discussed, some people thought a support group was on offer and one birth mother said she would not want to meet in a group again, because she felt uncomfortable with the strong feelings of others—feelings which she did not share. The meeting had failed to clarify the role of service user consultant. To plan a way forward, the researchers liaised with a birth parent support worker, who was also experienced in group work and was a birth mother herself, and also with a worker experienced in advocacy with people with learning difficulties. It was decided that, to progress, it was necessary to recognise: As a result, the birth parent support worker was recruited to have individual meetings with interested birth relatives who had attended that first meeting to clarify mutual expectations. Consultants were offered the chance to be ‘corresponding’ members of the group, who could take part via written comments, phone calls or home visits by the researchers. Without such flexibility, there was a danger of recruiting only the most confident and articulate service users.
that a large group meeting might not work for all who were interested in taking part—that a range of options should be negotiated with service users;
that the tasks were many and complex, and needed to be broken down as much as possible, not expecting everyone to work on every task;
that it was necessary to allow service users to talk about their own experiences but to manage that, and support them in moving on to think about other people's experiences.
Following these individual meetings, a service user group was established in a northern city, with another London-based group following the same model. Fourteen birth relatives were recruited (two birth parent couples, nine mothers and one grandmother). Consultants appreciated being considered as experts and asked for their help and advice. In all the birth relative consultant meetings, researchers were joined by a practitioner experienced in group work. Having a practitioner present was important, because her sole task was to concentrate on the dynamics of the group and the well-being of the consultants.
Practical considerations and the importance of relationships
The birth relative consultants met for day-long meetings throughout the project. Before meetings, papers would be sent out, supplemented by individual phone conversations. The meetings were held at accessible venues and travel expenses reimbursed in cash on the day. Good refreshments were provided and each consultant was paid for their participation. After the meetings, written information was circulated detailing changes made to the research as a result of the consultation. Attention to practical details was a way of demonstrating respect and trying to work inclusively, to prevent anyone from being unable to take part because of financial constraints or impact on benefit claims. As the project developed, some consultants were in contact with each other outside of the project, sometimes socially and at other times to work on their own adoption-related projects. Peer support became an important by-product of the advisory group.
The impact of birth relative consultants on the research
Improving the recruitment strategy
Birth relative consultants made several suggestions that aided recruitment: first, to ensure the leaflet was professionally designed and made clear that a university was carrying out the research. This would demonstrate that the research was important and was independent of social services. Second, they suggested that there was too much written information on the leaflet and that this was off-putting. The leaflet should encourage contact with the team who could then explain the details about the research on the phone. Third, their advice was to emphasise that taking part might help other people in a similar situation. This would be a key motivator for participants. They also advised that a Freephone number and a website should be provided, and to ensure that people could communicate by text messages, which are cheaper. It was important to ensure that letters going out from adoption support agencies to potential participants had not gone through the social service franking machine, as the letter might be discarded without being opened. The consultants helped the research team to understand how hostile participants might feel towards statutory agencies and social work. The consultants also suggested that it was important to recruit not only those who had used services, but people who were eligible but were not using services. As a result of these suggestions, recruitment to the study was relatively unproblematic.
Challenging research design and measures
As the research was in response to a government tender, the research questions were decided before the involvement of the service user consultants. An abiding concern for the birth relative consultants was how social workers could better support families to keep their children, rather than providing support for them through and after the adoption process. This concern also emerged as an important theme in the interview data. This suggests the importance of involving service users in research commissioning to ensure that the research questions are relevant to service users.
The birth relative consultants were involved in the design of the interview schedules and information sheets. They made suggestions about the wording of the consent form, suggesting that the first draft, which was intended to reassure participants that they did not have to answer all the questions, sounded like a police caution. They also suggested that birth relatives who might feel that they had signed their children away would be wary of signing any documents, and might have difficulty in reading the form and understanding what it was they were signing. It would be better to get recorded verbal consent as an alternative. Several birth relatives felt that it might be difficult for participants to do a telephone interview, and face-to-face interviews were therefore offered. Since one-third of birth relative participants took up this option, this made a considerable difference to recruitment.
Changes were also made to the measures in the study. A proposed Likert scale about satisfaction with contact was negatively received. The birth relative consultants likened it to giving a tick-box questionnaire to someone who had just attended a funeral. They felt that the topic was far too emotive to be constrained by a scale and that an open question would be more sensitive. As well as being attuned to the ethical dimensions of the research interview, the consultants thus drew attention to the tension between standardised measures and individuals' accounts of their own situations (Fisher, 2002), recommending that we seek the latter through more open questions. Improvements were suggested to the interview schedule such as asking not only about formal support services, but also informal support from friends and family. In the interview data, this opened access to an important theme about stigma and rejection which might otherwise not have emerged as strongly.
Data analysis: developing a model of coping with adoption
The academic researchers wanted to involve the consultants in the analysis rather than presenting them with the findings and getting their views. However, there was a great deal of data and much of it consisted of transcribed interviews—two factors which made it less accessible for the consultants. To resolve this difficulty, we made four ten-minute recordings of excerpts from interviews, selected as they varied on a particular dimension of interest. The use of edited recordings helped to respect the confidentiality of the data. The excerpts were read out by actors and recorded. This ensured that issues of literacy did not prevent people from contributing. Birth relatives were asked to rate the interview between 1 and 10 according to how well they thought the interviewee was coping with their child being adopted. This led to a broader discussion about what it meant to be coping.
In the analysis, the research team were trying to build on a previously developed model of ‘acceptance’ of adoption (Neil, 2007). For the researchers, the term ‘acceptance’ covered a wide range of reaction including the dictionary definition meaning ‘to tolerate’ or ‘submit to’, to know that the adoption had happened and could not be undone. Included in the overall acceptance model were categories of positive acceptance, passive and resistant reactions. The birth relatives said that the term ‘acceptance’ was not appropriate for the overall model, since, for them, acceptance could not be limited to the understanding that the adoption was final, but always carried with it the emotional connotation of agreement, which they felt was an unreasonable expectation. Instead of ‘acceptance’, the research team developed a rating scale about ‘coping with adoption’, which more adequately reflected the birth relatives' view that a person could never get over his or her child being adopted, but that it did make sense to think about how well or badly people were coping with the loss of the child. The consultants also differed from the researchers in their view of the psychological significance of anger as an indicator of coping, seeing intense anger as a response that would not necessarily diminish over time. They highlighted instead feelings of self-worth, self-destructive behaviours and lack of any positive feeling as being more important indicators of coping than the dissipation of anger over time. Furthermore, they emphasised that birth relatives could be rated not only according to how they were dealing with their emotions, but the extent to which they were able to re-engage with wider life activities such as having a job or a hobby. Thus, their emphasis was more akin to the recovery model in mental health, rather than solely a concentration on psychological symptoms and problems (Tew, 2011).
Making suggestions for practice
From the analysis of the qualitative interviews, we identified three stages to the adoption process, each with associated feelings. These were presented to the consultants in accessible form using flip charts and pictures. Consultants suggested services informed by their own experiences but grounded in the participants' data. The concrete suggestions were included in the reports and books which resulted from the research (Neil et al., 2010, 2011). Suggestions included using other birth relatives as peer supporters to people going through the adoption process; providing a phone line for out-of-hours support; considering how services are marketed to birth relatives to avoid creating unnecessary barriers; provision of online information and support via websites and discussion groups; and help and advice with activities birth relatives could undertake to help them restore/reorient their lives after adoption, such as encouragement regarding hobbies, interests, education and voluntary work. The service user consultant standpoint led to suggestions for practice that were specific, concrete and rooted in service user needs rather than current service provision.
Ensuring a service user perspective was central to the launch event
It was felt that it would be powerful to include the voices of service users in our launch conference, but both consultants and researchers were concerned about the emotional impact of the day. It was decided to invite the consultants to the television studio in the university to record interviews which could be edited and played during the conference. Several of the consultants would be present on the day and available for discussion in intervals and breaks. However, as the conference approached, the consultants suggested a Q&A panel of birth relatives, answering questions from the conference floor. At the conference, both the video interviews and the panel took place. A break-out room was booked solely for the use of birth relative consultants with a support worker on standby to provide support if necessary. It was essential that this safe space was available and it was utilised by all over the course of the day, for support and as somewhere to regroup and discuss how the day was progressing. The panel was a powerful opportunity for the consultants to be heard by practitioners and others who deliver services.
Feedback about the birth relative consultants' group
After the research concluded, one of the practitioners who assisted in the project interviewed consultants about their experience of being members of the advisory group. As the practitioner was familiar to some of the interviewees, this may have increased their willingness to be interviewed. However, it may also have impeded their willingness to be critical. Eight of the birth relatives were interviewed and one other birth mother has been interviewed independently about her experience of the project (TwoCan Associates, 2010).
There were many motivations for taking part in the research. The most common themes were wanting to help other birth relatives, wanting to get views across and wanting to improve the system. One person said that they were interested generally in learning more about the research topic. Few talked specifically about what difference they had made to the research, preferring to answer that question in terms of the difference taking part had made to themselves, although improving recruitment and making a difference to services were mentioned. For most consultants, the best part of being involved was meeting other people and sharing similar experiences. The group did offer peer group support as well as contributing to the research (Rhodes et al., 2002). On a personal level, several of the consultants felt that they had benefited from their involvement, which helped them to feel valued and important:
If it wasn't for them [the researchers] I wouldn't be here, I'd most likely be in the ground or something. They made me feel better, made me see more, the opportunities in life.
It's like getting over the wall, doing something different, getting back out there.
Two of the consultants felt that they had been able to channel their anger into something constructive. One commented:
… usually I couldn't be in a room of social workers. Even our local support group has noticed a difference in me. I'm still as angry, but not as violent as I used to be.
Some of the consultants talked about what they valued in the process. The practical elements were important in demonstrating respect and care which allowed trust to develop between consultants and researchers. The process of involving service users in the research process echoed themes in the research findings about how small gestures and basic compassion are valued in the provision of adoption support services. Being treated with respect was particularly important for the birth relative consultants whose experience of losing a child to adoption had been felt as disempowering:
We were being treated with consideration. We are used to being treated as lower than the low. Whereas here we were warmly greeted, fed and watered, pampered to an extent.
The project was held in such a manner i.e. compassionate, sympathetic and non-judgemental researchers, travel and other expenses made easy, that I don't think there was any way to have done better. The organisers adapted well to the diversity of the group, keeping the meetings professional and meeting the needs of each individual as and when required.
Suggested improvements included more time and more meetings, and having a dedicated counsellor available on meeting days to provide support if needed.
Discussion and implications for practice
We have argued that it is important for researchers to document both the process and the impact of service user involvement on research in order to evaluate its value. This paper has offered a case study of a service user consultant group in an adoption research project. In terms of a continuum of involvement, it best fits with that of ‘contribution’ as described by Sweeney and Morgan (2009), since decision-making power was retained by the researchers. Over and above consultation, ‘contribution’ reflects the significant input of service user consultants into the research and the longevity of the relationships established between researchers and consultants. It might be argued that the current project stemmed more from a managerialist approach to service user involvement (Beresford, 2009) than a democratic one, since decision-making power about the research was retained by the academic researchers. However, service user consultants' reflections on their involvement suggest that this was not a disempowering experience. The research team aimed to work in a way which ensured that social work research reflects the values of the social work profession (Dominelli, 2005). The specific aims and methods of service user involvement evolved throughout the project in discussion with the whole team. As relationships were established and trust grew, plans were adapted to reflect changing levels of confidence. For example, at the launch conference, the initial plan for filmed recordings was supplemented by a service user panel. The principle challenges of user involvement were that it was time-consuming, resource-intensive, at times emotionally draining, and these aspects were exacerbated by the fact that the academic researchers were themselves learning alongside the service user consultants.
The factors that contributed to the effectiveness of involvement included: As well as a consideration of the process of involvement, this paper has discussed the concrete impact on the research itself. Service user involvement can be justified on epistemological and methodological grounds. The service user consultants offered a unique standpoint which not only improved the research design, but also offered a unique perspective on the data analysis, grounded in the experience of having a child adopted from care. This standpoint supplemented, and at times challenged, the academic researchers' training in research methods and analysis in a valuable way, enriching the analysis of coping with adoption. The service users involved all shared the experience of having a child adopted from care; however, working with a group drew attention to diversity and different experiences, for example according to gender and relationship to the child adopted. The consultants made important suggestions for changes to practice as a result of the research, and have been involved in presentations to disseminate the findings. Ultimately, to assess the value of involvement, it was crucial that service user involvement could be demonstrated to make a concrete difference to the research itself, especially since the central motivation for service users to get involved was to make a difference and improve services for others going through the adoption system.
discussion to establish clarity about the role of service user consultant;
having a number of consultants with diverse experiences who developed confidence as a cohesive group over the course of the project;
working in an inclusive manner, focusing on processes and building trusting relationships;
the provision of practical and emotional support from the research team and practitioner co-workers;
breaking down the research into manageable tasks;
regular communication and feedback to service users about their impact on the research between meetings;
a willingness to learn on the part of academic researchers and service user consultants and a commitment to overcoming challenges;
a commitment to service user involvement on the part of funders.
The paper draws upon a study funded by the Department for Children, Schools and Families as part of the Adoption Research Initiative. Approval for the research was obtained from the Association of Directors of Social Services. Participants were given the choice of providing verbal or written consent.