Abstract
Peer research has the potential to empower young people to participate in research by minimising power imbalances between researchers and participants; this may reduce bias and promote improved understanding to inform policy and practice. However, these benefits are not automatic; the relative inexperience of peer researchers adds layers of complexity to the research process. Moreover, the validity of findings from research adopting less traditional methods may be questioned and policy makers may be cautious about accepting this evidence, thus limiting its contribution and impact. This paper explores the advancement of participatory peer researcher methodology in research with children in and leaving care and ethical, practical and data quality issues that arose in two studies exploring young people’s transitions from care to adulthood. It concludes that the peer research methodology can yield rich data but that adequate resources and effective research management are crucial. The authors also caution against a reductionist approach that privileges peer research methodology above other methods of inquiry in the study of transitions from care to adulthood.
Introduction
A critical concern in the design and delivery of applied research to inform social work policy and practice is that it meets what Aldridge (2014) describes as the ‘top down’ demands of the academy and funders for scientifically robust research evidence, whilst upholding the ethical values of social work and trying to ‘give voice to clients and thereby attempting to democratise both research and social work itself’ (Parton and Kirk, 2010: 24). In practice, dilemmas and challenges can arise in meeting these demands. In research with vulnerable groups less conventional and more participatory research designs, which are tailored to meet the needs of participants, may empower them, facilitate their active engagement in research and give them a voice (Fleming et al., 2009; Kilpatrick et al., 2007; Murray, 2006). However, the validity of the findings may be questioned and policy makers can be cautious about accepting this research evidence, which can serve to undermine its transformative potential (i.e. contribution to policy and practice developments) (Aldridge, 2014; Walker et al., 2008).
This paper explores the approach employed by the authors’ to try and negotiate these tensions in two government pilot programmes (Right2BCared4 and Staying Put 18 Plus Family Placement Programme) aimed at improving outcomes for young people making the transition from care to adulthood (Edwards, 2011; Munro et al., 2010a, 2010b, 2011, 2012; National Care Advisory Service, 2012). It explores the foundations that are needed to protect and empower both peer researchers and research participants (in this context looked after children and care leavers) and examines whether this approach can yield high quality data to inform the development of policy and practice to improve outcomes for young people navigating the transition from care to adulthood.
Rationale for adopting a peer research methodology
Past research demonstrates that there are a number of barriers to securing the views of looked after children to inform decisions about policy and practice (Cunningham and Diversi, 2012; Daly, 2009; Munro, 2001; Munro, 2008a, 2008b; Munro et al., 2005; Ward et al., 2005). Lack of motivation, low self-esteem and power imbalances between adult researchers and young people mean that some young adults do not feel comfortable participating in studies. Power relations and the professional status of interviewers may also influence the responses of those who agree to participate (Cunningham and Diversi, 2012; Curtis et al., 2004; Kilpatrick et al., 2007; Moore et al., 2011; Shier, 2001; Ward et al., 2005). Fine and Sandstrom (1998) identify that children and young people may be reluctant to express dissent or dissatisfaction during interviews conducted by adult researchers. These issues mean that samples of looked after children or care leavers may not be representative and findings may be biased towards certain ‘sub-groups’ which is not an ideal foundation for policy development (Kirby, 1999; Ward et al., 2005).
In recent years, there has been increased recognition of the value of obtaining young people’s views, involving young people in decision-making and policy development (Checkoway and Richard-Schuster, 2003; Cunningham and Diversi, 2012; Hojer and Sjoblom, 2010; Munro, 2008a; Prout, 2002; Roche, 1999; Shier, 2001) and adopting peer research methodology, in studies involving looked after children and care leavers (Bureau, 2007; Children in Scotland, Daly, 2009; Scottish Executive, 2006; Stein and Verweijen-Slamnescu, 2012; WMTD/Rainer and National Children’s). However, a key criticism of the participatory research approach with children and young people is that it has the potential to be tokenistic. Peer researchers may be tasked with collecting data for ‘qualified and experienced academics’ to analyse, thereby excluding young people from the analysis and dissemination stages of the research (Clarke, 2004; Coad and Evans, 2008; Kellett, 2011; McLaughlin, 2005). In recognition of these issues the authors sought to move through Shier’s (2001; 2006) participatory framework. Shier (2001, 2006) identifies five levels of participation in relation to the involvement of children in decision-making: (1) children are listened to, (2) children are supported in giving their views, (3) children’s views are taken into account, (4) children are involved in decision-making processes, (5) children share power and responsibility for decision-making. Approaches one and two have been widely adopted in research with looked after children (see, e.g. Skuse and Ward, 2003). Yet, it is from level four that young people are directly involved at the point where decisions are made. Prior to this young people may be empowered, i.e. ‘strengthened’ or ‘supported’ but decision-making remains with the adults (Shier, 2001: 113–114).
The Right2BCared4 and Staying Put evaluations actively involved the peer researchers in all stages of the research process including decisions about research questions, design of the research tools, undertaking of data collection, analysis of findings, and write up and dissemination – meeting the fourth level on Shier’s (2001; 2006) participatory framework (due to the timescale for submission of the research proposals it did not prove possible to involve young people in designing the studies). Young people aged 18–25 who had spent time in local authority care were trained as peer researchers and engaged in the evaluation of the Right2BCared4 and the Staying Put 18 Plus Family Placement Programme (see Edwards, 2011; Munro et al., 2010a, 2010b, 2011, 2012; National Care Advisory Service, 2012).
Peer research methodology was employed because the research team recognised that the ‘insider’ knowledge of the peer researchers and their assistance in framing the research questions and in interpreting the data may facilitate enhanced understanding of the subject (Cleaver, 2001; Moore et al., 2011; Smith et al., 2002). The approach also has the potential to empower participants by giving them a voice and minimise power imbalances between the researcher and participant which may reduce bias and enhance the quality of data collected (Clarke, 2004; Cleaver, 2001; Fleming et al., 2009; Kilpatrick et al., 2007; Murray, 2006). Mutual understanding of the research topic and a shared language also help build rapport and support effective communication (Alderson, 2001 cited in Fleming et al., 2009; Greene et al., 2009; Kilpatrick et al., 2007; Kirby, 1999, 2004; Murray, 2006). Evidence suggests that young people may be willing to discuss issues with peer researchers that they would not be willing to raise with academic researchers, thus improving our understanding of the topic and securing accounts that are not tailored for adult ears (Burns and Schubotz, 2009; Fleming et al., 2009; Kirby, 1999; Smith et al., 2002).
Smith and colleagues (2002) conclude that ‘participatory research is beneficial both because of its implicit values (such as empowerment and inclusion) and also because it improves our level of understanding of the substantive subject area’ (p. 2). However it is also important to acknowledge that such benefits are not automatic. Cleaver (2001) argues that these two strands do not intrinsically go hand in hand and that a key challenge is ensuring that those involved in research are ‘empowered’ to make a difference whilst at the same time ensuring that studies remain academically robust. The relative inexperience of peer researchers does raise important questions about whether the methodology can secure high quality data. Moreover, the use of peer research methodology raises ethical challenges and dilemmas. This paper explores these issues and outlines the peer research approach developed by the authors, based at the Centre for Child and Family Research (CCFR), Loughborough University in collaboration with the National Care Advisory Service (NCAS).
Application of the peer research methodology
Ethical issues
Recruiting suitable peer researchers for the task in hand is an important foundation for the effective conduct of research employing a peer research methodology. It is important that the desire to be ‘inclusive’ does not take precedence over the selection of peer researchers who have the skill and capability (with appropriate support and training) to contribute to the process and interview young people. As highlighted by Kellett (2011), researchers have a responsibility to conduct high quality, reliable and valid research. Poor recruitment and selection denies participants optimum conditions to tell their story. Proactive management of this part of the process is also important because otherwise there is a danger that peer researchers are ‘set up to fail’, which is contrary to the aim of empowering and equipping them with transferrable skills. In recognition of these issues, CCFR and NCAS developed a job description clearly outlining the essential skills and qualities (for example, good communication, literacy and organisational skills) that were needed for the task. This information was distributed to lead contacts (leaving care personal advisers or team managers) in participating authorities so that they could identify young people who may be suitable. They were also supplied with written information on the purpose of the evaluation and research plans, the nature and scope of the peer research role, time commitment required and training expectations. This provided sufficient information for team managers to assess which care experienced young adults had the capacity to fulfil the role. It also meant that potential peer researchers were provided with written information on what involvement would entail so they could make an informed decision about whether they were interested in taking part. It was also made clear to potential peer researchers that they would be fully supported by staff from the local authorities, NCAS and CCFR throughout the process.
In two local authorities, semi-formal interviews were held to select peer researchers but in the majority of cases social workers or managers identified specific young adults who they judged would be willing and able to actively engage in and meaningfully contribute to the research. In total, 28 peer researchers were recruited and all but seven remained involved in the research projects until their completion (approximately 18 months later). In two cases young adults opted out after the first training event, deciding that they did not wish to remain involved as peer researchers; one of these young people felt that they did not possess the literacy skills required to fulfil the role. In one case, a Criminal Records Bureau disclosure revealed recent convictions which were assessed to be sufficiently serious to prevent a young person undertaking interviews. In another, recent allegations had been made and the support worker [leaving care worker] from that local authority decided that this young person could not undertake interviews until the allegations had been resolved. In the remainder of cases, peer researchers’ circumstances changed during the course of the evaluation. It is important to recognise that most young people will be in education, employment or training which will take priority and situations often change as young people negotiate multiple transitions during early adulthood (Kilpatrick et al., 2007). In this context, it is valuable to recruit extra peer researchers and/or factor in time and resources to train additional peer researchers during the course of the study.
Power differences between children and adults represent a considerable ethical challenge for researchers seeking to secure children and young people’s views (Alderson, 1995; Morrow and Richards, 1996; Thomas and O’Kane, 1998). Peer research can be understood as an approach that aims to resolve this challenge but raises other ethical dilemmas and adds new layers of complexity to the process. These are associated with the subject group, research topic and the use of peer researchers. Both the studies involved researching older children in and leaving care and whilst the research did not focus upon experiences of abuse or neglect it did involve discussion of potential sensitive topics including, for example, the quality of young people’s relationships with their foster carers and how they coped with the transition from care to independence. The peer researchers conducting the interviews were inexperienced in fulfilling this role. There was also the potential for the interviews to reawaken issues associated with their own experiences in and leaving care. In this context, it is evident that robust systems and processes need to be put in place to protect both the participants and the peer researchers from harm. The mechanisms put in place also need to be subject to independent scrutiny. These studies were both approved by Loughborough University’s Ethics Committee and reviewed by a number of the participating local authorities so that they could assure themselves that they were compliant with the Research Governance Framework for Health and Social Care (Department of Health, 2005).
Training was provided to ensure that peer researchers were confident in obtaining informed consent and outlining the limits of confidentiality to potential research participants. To facilitate this, a clear protocol was established. In the event of disclosures, peer researchers were to notify their support worker who would in turn contact the research team at Loughborough University. Where there was a difference of opinion as to whether a disclosure was sufficiently serious to warrant breaching confidentiality, advice was sought from a senior manager from a local authority that was not directly involved in the evaluation.
Mechanisms were put in place to ensure researcher safety when undertaking fieldwork. Face-to-face interviews, undertaken by peer researchers with research participants, were carried out in the offices of the participant’s local authority. Peer researchers were also provided with on-going support from leaving care workers in their own local authorities; many already had established relationships with these workers prior to commencement of the evaluations. Participation workers at NCAS and research staff from CCFR, Loughborough University were also available throughout the process, to provide additional support.
Training events
Three training events were held over the course of each of the research studies. The first two-day training event served as important foundations for subsequent activities. They provided an opportunity to build relationships, discuss and agree respective roles and responsibilities and establish the principles of engagement, with the aim of ensuring that the peer researchers had a sense of ownership over the process (see also McLaughlin, 2005). At the event the research team provided an overview of the project and research methods to be adopted the timescale for submission of the research proposals meant that it did not prove possible to involve young people in designing the studies and outlined their hopes and expectations concerning the peer research component. Through interactive sessions, the peer researchers were actively engaged in designing accessible information about the project for potential research participants and developing the research tools (including a survey (Right2BCared4) and interview topic guides (Right2BCared4 and Staying Put)). The peer researchers suggested topics and questions to be included in the survey and interview guides and highlighted areas that they felt were important to explore in the context of the aims of the evaluations. There was full transparency with peer researchers over the likelihood that topics and questions might be re-framed and re-ordered to facilitate the flow of the interviews. This is the same iterative process used in other research whereby drafts are developed, reviewed and refined to ensure that the data collection tools created obtain information that will answer the main research questions of studies and meet the expectations of commissioners. Alterations made to the survey and interview questions were sent to the peer researchers for review and comments before final versions were created and administered.
The second events focused on training the peer researchers to undertake interviews with older looked after children and care leavers. Sessions included short formal presentations from the research team, role play, group discussion and feedback sessions. This combination was intended to suit different learning styles. The content was designed to ensure that the peer researchers were trained in interview techniques, including building rapport, listening skills, and questioning and probing to elicit high quality data. During the course of the training, attention was drawn to the importance of being empathetic without putting forward personal reflections, opinions or views on care that could influence participant responses and undermine the reliability of the data. Considerable attention was also given to ethical issues (as outlined earlier). The final events involved training and supporting the peer researchers to analyse the young people’s interview transcripts, identify key messages for policy and practice and design a peer research report to provide research participants and other children and young people with an accessible summary of the findings from the interviews (see Edwards, 2011; National Care Advisory Service, 2012).
Data collection arrangements
Around 9990 young people make the transition from care to adulthood each year, i.e. an average of 65 per local authority (Department for Education, 2013). The relatively small population means that potential research participants may be friends or acquaintances of peer researchers living in their area. Pre-existing relationships may influence participant responses to the interview questions and heighten anxieties regarding confidentiality. On this basis a decision was taken to establish partnership arrangements between local authorities involved in the pilot. Peer researchers then travelled to their partner local authority to undertake interviews there.
The arrangements meant that peer researchers had to travel some distance to undertake interviews, were conducting interviews in unfamiliar locations and did not have a member of staff who they knew well on hand. However, leaving care personal advisers were committed and proactive in supporting peer researchers from their partner authority throughout the process. This included, for example, contacting the peer researchers to organise the fieldwork, and collecting them from the bus or train station and taking them to local authority premises in which the interviews were being conducted. They were also on hand while interviews were underway so that they could offer advice and guidance if necessary. Additional support was also available from participation workers at NCAS throughout the process. On the whole the approach worked well and the target number of interviews for each study were completed; 65 in total. However, the reciprocal arrangements did make the logistics more complicated and the timetable for completing the work had to be extended by 1 month for Right2BCared4 and 3 months for Staying Put. This is not unique to the authors’ research, others have also noted that the timescale for completing studies involving participatory approaches is longer than those adopting traditional methods (Kilpatrick et al., 2007; McLaughlin, 2005). Peer researchers also found it disheartening when, having travelled some distance, young people failed to turn up for the interview. While this is not necessarily uncommon in research it highlights the importance of communicating the fact that this might happen to the peer researchers and local authority personnel. It also means it is necessary to develop payment structures for peer researchers which are not based solely on the completion of interviews.
Data analysis
Concerns have been raised that peer researchers are often tasked with collecting data for ‘qualified and experienced’ academics who then exclude them from participating in the analysis phase of the research cycle (Clarke, 2004; Coad and Evans, 2008; Kellett, 2011; McLaughlin, 2005). Holland et al. (2010b) argue that ‘lay people’ may find analysing data too ‘emotionally or intellectually’ challenging. Yet others have questioned this view and suggest that lay people (including young people) understand the principles of thematic analysis and are able to use this technique to make sense of the data (Coffey and Atkinson, 1996; Fleming, 2011).
The research team was committed to ensuring that the peer researchers were actively involved in the process from the development of the research tools through to the write up of the findings and therefore chose to include them during the analysis stage. A two-day analysis event was held which included training on the thematic analysis of qualitative data. The peer researchers then focused upon coding transcripts and identifying the key findings. As one of the peer researchers articulated ‘the team headed back to Loughborough [University] for an evaluative meeting in which the findings of the interviews were assessed. These findings consequently influenced the topics included in the report’ (National Care Advisory Service, 2012: 9).
Dilemmas and tensions did arise for the authors as they sought to embed a participatory approach whilst trying to ensure that the analytical task was both manageable and robust. Taking into consideration the relative inexperience of the peer researchers, their other commitments (i.e. college/university and work) and resultant capacity to undertake this time consuming task, the research team opted to group data from the full transcripts thematically. It could be argued that this approach results in only partial participation in the analysis stage. However, this can also be understood as a pragmatic decision that took into account peer researchers’ availability (i.e. amount of time they could reasonably contribute to the task) and experience (see also Moore et al., 2011). In addition, the themes adopted (e.g. preparation for leaving care and support networks) were those that the peer researchers identified as important during the first training sessions, rather than themes identified and imposed by academic researchers during preliminary analysis.
The sections of the transcripts supplied from each interview, under each theme, were purposively lengthy to make sure that the interview data were not divorced from the context of the discussion. Information on the age, gender and current care status (looked after child or care leaver) of the participant were also supplied so that the peer researchers could examine similarities and differences according to these characteristics. The peer researchers were supplied with highlighter pens and read and manually coded the data. Key findings emerging from the data were written up on flip charts and work was undertaken with the peer researchers to identify recurrent issues and to explore what they perceived to be the most important messages. Although the peer researchers were not engaged in analysis to the same extent as full-time academic researchers, the approach employed served to facilitate communication and a joint process of knowledge production (Berghold and Thomas, 2012).
Data quality: Validity and reliability
Peer researchers are inexperienced compared to their academic counterparts, who at a minimum will have a degree in a relevant discipline, as well as up-to-date knowledge of children’s services roles and responsibilities, understanding of both quantitative and qualitative research methodology, good analytical skills and the ability to write to publication standard. Instead peer researchers bring their lived experiences and ‘insider’ understanding of (in this case) being looked after and making the transition to adulthood. They were also provided with training on core research skills and interview techniques. The interview data they collected demonstrates that the peer research methodology can secure rich data that powerfully conveys the experiences of young people navigating the transition from care to independence; as the peer research reports from the two studies show (Edwards, 2011; National Care Advisory Service, 2012). There was also no evidence to suggest that the peer researchers were too ‘close’ to the topic under discussion and that this influenced participants’ responses.
In general, findings from the Right2BCared4 and Staying Put evaluations aligned with other studies on transitions from care to adulthood, conducted by experienced researchers employing traditional approaches (Barns et al., 2005; Dixon et al., 2006; Goodkind et al., 2011; Hojer and Sjoblom, 2010; Holland et al., 2010a; McCoy et al., 2008; Ofsted, 2012; Stein and Munro, 2008). However, the lived experiences and ‘insider understanding’ of the peer researchers did contribute to the analysis and write up of the findings.
Their involvement led to a unique interpretation that may not have otherwise been obtained. The concept of ‘pathway planning syndrome’ was introduced by the peer researchers to convey what young people had told them about this social work process. The Children (Leaving Care) Act 2000 requires a Pathway Plan for all eligible, relevant and former relevant young people. The purpose of the plan is to assess the needs of young people and the actions and services required to respond to these needs and the type of support that will be provided during the transition to adulthood and independence. The peer researchers highlighted that a number of young people thought that leaving care personal advisers placed too much emphasis on bureaucratic processes and completing the pathway plan paperwork at the expense of spending time with young people discussing their individual support needs and reviewing changes in their circumstances (see also Munro et al., 2011). Without the inclusion of peer researchers in this study the data on pathway planning is unlikely to have been interpreted as ‘pathway planning syndrome’ reflecting the benefits of working with the researched group.
In addition to the above, the importance of foster carers maintaining contact with former looked after children, after care, was highlighted by the peer researchers: The peer researchers felt strongly that where possible that the relationship between foster carers and young people should be continued and maintained once they have moved on. (Munro et al., 2012: 72; National Care Advisory Service, 2012: 14).
Prompting and probing during interviews
Although the research served to identify important messages for policy and practice it is important to acknowledge that there were variations in the quality of the data. In some transcripts the peer researcher’s inexperience was apparent, for example, it was clear that some of the peer researchers found it difficult at times to process information provided by participants quickly enough and asked repetitive or inappropriate questions as a result.
Peer researcher: Is there anything else that you would change about moving from care into independent living?
Interviewee: Yes. I’d rather be back in care.
Peer researcher: If you had a chance would you go back into care and if so, please tell me why?
Peer researcher: Okay. How would you describe your relationship now with your previous foster carers?
Interviewee: I haven’t spoken to them in about a year.
Peer researcher: That’s fine. Right. Is that what you hoped for, when you left your placement?
Interviewee: Not completely but for the most part.
Peer researcher: Erm…what is your relationship like with your Social Worker, Leaving Care Worker?
Influence on policy and practice
As discussed previously policy makers tend to be more cautious about using findings from research adopting the less traditional participatory method (Aldridge, 2014; Walker et al., 2008). Despite this tendency (and without inferring causal link) it appears that findings from the evaluations of Right2BCared4 and Staying Put have, along with other forces and evidence, influenced developments. Following the Right2BCared4 evaluation new statutory guidance on the role of the Independent Reviewing Officer (IRO) was issued (Department for Education, 2010). More recently, ‘Staying Put’ arrangements were placed on a statutory footing (Children and Families Act 2014, section 98). The new duty means that local authorities are legally required to advise, assist and support looked after children reaching legal adulthood to remain with their current foster carers beyond their 18th birthday and up to the age of 21 years, providing the foster carers are willing and able to and the young person wishes to stay. These amendments were made following a campaign by the Fostering Network, which drew upon findings from the Staying Put evaluation, alongside other studies, to lobby for reform to facilitate transitions from care to adulthood that are more akin to those experienced by young people in the general population (Cann, 2013).
Reflections on the expansion of peer research to obtain the views of looked after children and care leavers
In recent years there has been an expansion in the use of peer research methodology in studies exploring the views and experiences of looked after children and care leavers. A major EU project on transitions to adulthood in Finland, Albania, Poland and the Czech Republic adopted a participatory approach (Stein and Verweijn-Slamnescu, 2012). Peer research studies on corporate parenting and social pedagogy are also underway (Dixon, 2014; McDermid, forthcoming). As reflected in this paper, the methodology can promote looked after children and care leavers’ participatory rights and inform the evidence base. However, it also highlights the challenges that were encountered in the design and conduct of semi-structured interviews and thematic analysis. This raises questions about whether it is feasible or desirable to fully engage peer researchers in studies employing complex interview techniques and analytical approaches to answer specific research questions. Arguably, applying a historical intergenerational lens and using a life course biographical approach to understand transitions to adulthood (see Nilsen and Brannen, 2014) would be beyond the scope of peer researchers. On this basis, the authors caution against a reductionist approach that privileges peer research methodology above other methods of inquiry.
Conclusion
It is important to elicit young people’s views to improve policy and practice to meet the needs of vulnerable groups. The peer research methodology serves as a tool to facilitate this because it has the potential to empower young people to participate in research by minimising power imbalances between researchers and participants; thus reducing bias and contributing to children and young people’s voices being heard, enhancing understanding. However these benefits are not automatic. When peer research methods are adopted it is important to have robust training and effective support mechanisms in place to safeguard both the peer researchers and research participants and in order to obtain high quality data.
The Right2BCared4 and Staying Put evaluations achieved the highest level of participatory research, that is, they trained the subject group, in this case care leavers, in research methods to enable them to undertake research related to topics they have identified as important (Holland, 2010b). The peer researchers were successfully engaged through to the completion of the evaluations including thematic analysis and write up of the findings. However, it is important to note the limitations. Peer researchers cannot be involved in research projects to the same extent as salaried full-time academic researchers. Their time will be limited due to other commitments (education and employment) and they will not be as skilled in research methods as their academic colleagues who will have spent many years training and working in the field of research. This is not to say that their participation is not meaningful or that they cannot be involved in all stages of a research project, rather that additional resources, flexibility and time are required to successfully carry out a research study using the peer research methodology. Findings also highlight that while peer research has a place in understanding young people’s transitions from care to adulthood this should not be at the expense of the conduct of robust research employing traditional methods.
