Working with vulnerable groups in social research: dilemmas by default and design

Involving vulnerable participants in research – concepts and challenges

When embarking on research studies with vulnerable participants, a number of dilemmas are inevitable and are thus there by default. These relate to the research governance and ethical matters as well as to issues of interpretation and representation. Others are there by design, both philosophically and literally, in the ways in which researchers are purposive in their approach to conducting research with vulnerable participants, despite the challenges involved, and also in the ways in which less-conventional participatory methodological strategies are employed (purposively) in order to promote and even embrace mutuality and subjectivity in terms of explicating lived experience.

However, Walker and colleagues (2008: 164) have argued that policy makers and practitioners tend to be ‘more cautious’ in their response to research evidence that is based on less-conventional, participatory approaches. Yet, such methods are often both apposite and necessary when working with vulnerable research participants in whichever field or discipline. For the researcher involved in designing and conducting research with vulnerable individuals and groups, dilemmas arise from the outset that first necessitate particular and careful consideration of notions of ‘vulnerability’, both definitionally and conceptually. Larkin (2009) recognises that a precise definition of vulnerability ‘remains elusive’ and also that ‘its meaning also varies according to the context in which it is used’ (p. 1). This is certainly true – in many respects, conceptually, vulnerability is context dependent; in health and social care discourses, for example, it is mainly used to denote susceptibility to harm or risk, or as an indicator of enhanced need.

Furthermore, while various health and social care descriptors define vulnerability in terms of a person’s capacity for self-care (Department of Health, 2000: section 2.3), as well as define vulnerable people as those in receipt of health or social care services, reference is also made more generally to those who ‘require assistance in the conduct of their affairs’ (Safeguarding Vulnerable Groups Act, 2006: 3). Importantly, Larkin (2009) makes distinctions between people who are individually, uniquely and innately vulnerable and those who are vulnerable because of their circumstances, because of the environment or as a result of the structural factors or influences. These are important distinctions because they also point to the reasons why some individuals can be susceptible to multiple vulnerabilities – innately, circumstantially and also potentially by research processes themselves. As Moore and Miller (1999) propose, ‘Combinations of factors, such as physical disability experienced during older adulthood or altered physiological status experienced during childhood and adolescence may render individuals doubly vulnerable’ (p. 1035). When such individuals are also involved as participants in research studies, it is also possible that the research processes themselves may exacerbate susceptibility. Research that focuses on ‘sensitive topics’ is recognised in Economic and Social Research Council (ESRC; 2010) classifications, for example, as having the potential for creating or enhancing (existing) vulnerability among research participants.

For the researcher embarking on empirical investigations with vulnerable people, definitional and conceptual dilemmas can be further underscored by the fact that their own perceptions, as researchers, of participant vulnerability, may change over time. Prior to, during or even after working closely with such participants, researchers’ perspectives and opinions about vulnerability – its innate or circumstantial characteristics – may shift or change. Furthermore, when working with vulnerable groups in research and practice, it is important to recognise that participants’ self-perceptions may not always accord with the perceptions of others, or fit easily with external identifiers or classifications, particularly when participants see themselves as resilient rather than vulnerable in certain contexts. Steel (2004) acknowledges both the number of individuals and groups who are described as vulnerable by service providers, for example, as well as the fact that ‘some of these people would not describe themselves as vulnerable or marginalised at all’ (p. 1).

While philosophical and conceptual questions relating to vulnerability and resilience are inevitable consequences of working with marginalised or socially excluded participants in research, from a research governance and ethical perspective, considerations such as these, as well as individual self-perceptions of vulnerability, are unimportant; what is important is that research participants themselves are not put at further risk of harm or their vulnerability is not exacerbated by research processes, and that researchers and institutions are equally protected. Thus, strict adherence to ethical regulations and procedures must, by default, form the necessary preconditions for any research project or investigation that involves working with vulnerable participants (see, Department of Health (DH), 2005; ESRC, 2010). ¹

Although the identifiers and classifications for vulnerable individuals and groups are, once again, somewhat broad in terms of DH and ESRC research governance criteria, ethical protocols are intentionally rigid in terms of the requirements of researchers embarking on studies with vulnerable groups. While it is clear that some researchers welcome this rigidity and the formal regulation of ethical procedures because of the protection it affords participants, researchers and institutions (Munro, 2008), others have argued that stringent ethical procedures and requirements such as these can serve to exclude vulnerable individuals and groups from being included in research studies (Hammersley, 2009; Hurdley, 2010). Once again, this places those researchers who wish to involve vulnerable people in research studies in somewhat of a predicament, with the result that sometimes researchers may choose the ‘easier’ option of often avoiding lengthy and challenging ethical procedures as well as complex research design issues because vulnerable participants often require methodologies and approaches that are less conventional or straightforward.

It is certainly the case that vulnerable people can, potentially, be excluded from conventional social research studies and processes if methods used are not appropriate or sufficiently adaptable to take account of both innate and structurally related vulnerabilities. Thus, if researchers are not willing to step outside the boundaries of their conventional methodological fields, there remains the real possibility that vulnerable individuals and groups will be overlooked or considered too ‘difficult’ to include in research studies more generally. In many cases, utilising less-conventional methodological approaches means adopting participatory or inclusive strategies that emphasise mutuality and, at times, a certain degree of subjectivity in the way in which lived experience is explicated and represented.

Resolving dilemmas through participation and ‘voice’

Booth (1996) has argued that conventional, or ‘textbook’ social research methods can often discriminate against vulnerable respondents, ‘Methods that rely on reading or writing or abstract reasoning of verbal fluency may effectively exclude them from the role of respondent or informant in ways that mirror their exclusion from wider society’ (p. 252). It is clear that qualitative methods generally are often much more appropriate to use with participants who are vulnerable in some way because such methods (e.g. in-depth interviews, focus groups, case study, life history research, etc.) offer the degree of flexibility and sensitivity that is required and that the other methods do not provide (such as questionnaires, online surveys etc. – see Bryman, 2002). However, more conventional qualitative methods, such as interviews and so on, may require some adaptation, or should be used alongside other participatory approaches as part of a multi- or mixed-method design when used among vulnerable participants. This is especially important if such methods are to fulfil the principles and objectives of democracy and mutuality and if messages from these studies are to be transformative in broader social and political contexts (see discussion below).

Undoubtedly, less-conventional participative qualitative methods lend themselves more readily to the flexibility and adaptability of approaches necessary when working with vulnerable participants, including children and young people, disabled people and so on (see Aldridge, 2007, in press; Etherington, 2000). The intention of participatory, and participatory action research (PAR) methods in particular, has been to promote inclusion and participation and, importantly, the recognition and value of the voices of individuals and communities in social research (e.g. Borda, 1988; Goodley and Moore, 2000; O’Neill et al., 2002; Thomson, 2008). Furthermore, PAR draws on the theoretical principles and ideas that relate to mutuality and understanding in the research process and, ‘is designed to promote active involvement in every stage of the research process by those who are conventionally the focus of research’ (Chataway, 1997: 457; see also Borda, 1988; Bourdieu, 1996; McTaggart, 1997; Reason, 2006).

While the types of methods used in participatory and PAR studies can be diverse and embrace techniques adopted and adapted from other disciplines and practices, the stories and ‘voices’ of participants are placed centre stage both in the design and objectives of participatory approaches. Walmsley and Johnson (2003: 10) recognise this methodological diversity in studies that have been described variously as ‘participatory, action or emancipatory’, but that all have a common objective and intention to engage more effectively with participants and in ways that facilitate more empathic and democratising approaches to research participation. Ultimately, such studies and approaches may serve to emphasise and promote the resilience of participants rather than their susceptibility. Walmsley and Johnson describe this type of approach, as well as their own participatory research with people with learning disabilities, as ‘inclusive’ research, where participants ‘are active participants, not only as subjects but also as initiators, doers, writers and disseminators of research’ (Walmsley and Johnson, 2003: 9); in this sense, such approaches also emphasise proficiency and potentiality of research participants.

To give an example here, it has been argued that one of the key reasons children and young people are often both vulnerable and marginalised in society is because of the absence of their ‘authentic’ voices in public discourses about childhood (Hill et al., 2004). From an empirical perspective, children and young people have traditionally been left out of research studies that take as their focus macro themes and dimensions such as politics, citizenship and human rights. As Alderson (1995: 40) has argued, ‘Adult-centred research is conducted about the public world of the environment, politics, economics and other social affairs with little reference to children’ (p. 40).

However, in the last decade or so, considerable advances in research, theory and practice have meant that children’s ‘voices’ have been heard in a range of different contexts – in social research studies, for example, as well as, increasingly, in health and social care consultations and in the formulation of policy (see Hill et al., 2004; Redsell and Hastings, 2010; The Children Act, 2004). These revitalised perspectives and practices have been underpinned by a participation agenda with an international mandate (United Nations Convention on the Rights of the Child (UNCRC), 1989) and epistemological advances that have witnessed real shifts in the discourse on childhood and in the (‘new’) sociology of childhood specifically (see Wyness, 2006). In addition, more recent emphasis in areas of health and social care practice has been on children’s resilience which, according to Parrott and colleagues (2008) is reinforced by children’s own ‘sense of security, a recognition of self-worth and the experience of control over one’s immediate environment’ (p. 4).

In these contexts, children’s voices in social research studies have been at the forefront of new approaches and new ways of working collaboratively, and, in more genuinely participatory ways, with children and young people. Increasingly, research studies that involve children – and particularly those who experience multiple vulnerabilities – have been more adaptive and creative in their approaches to research design and collaboration (see Aldridge and Sharpe, 2007; Joanou, 2009; Thomson, 2008). In some cases, of course, such adaptive strategies ² are necessary, not least in order for researchers themselves to resolve a number of methodological challenges presented when working with vulnerable children and, specifically, in designing research methods and approaches that are more suitable to participants who, for example, may not want to take part in studies that use conventional qualitative research techniques.

To give a further example here from my own work with vulnerable children who provide care for parents ³ who are chronically sick or disabled, while numerous investigations had been conducted on and with these children over a number of years (including those that used both qualitative and quantitative methods – see Aldridge and Becker, 1993, 1994, 2003; Dearden and Becker, 1995, 2004), it became increasingly clear that some children and young people either did not want to, or felt unable to, take part in studies that relied on their verbal contributions. Vulnerable children such as these – as well as, potentially, any research participants who are unwilling or feel unable to verbalise their experiences and needs – are more likely to be excluded from research studies if they present as reluctant or challenging in research design and recruitment terms, and this can only serve to further emphasise their vulnerability. These same issues and challenges are reflected in disability and disability action research that seek to engage with disabled research participants in more democratic and inclusive ways (Goodley and Moore, 2000; Walmsley and Johnson, 2003) but at the same time aims to be transformative – in terms of translating the messages from research into practice in order to improve lives. The ethical and methodological predicaments facing the researcher in these contexts are obvious, particularly when participatory or inclusive objectives can so easily be compromised by broader issues and dynamics – as I have said, the ‘top down’ demands of the academy for ‘scientific’ rigour, for example (see Walmsley and Johnson, 2003), the needs of funders, the need for more genuinely innovative methodological approaches (Wiles et al., 2011) and so on.

However, the notion of ‘voice’ in participatory research methods places both the narratives and the rights of participants centre stage in research agendas and processes by enabling them to speak or ‘tell’ their stories in their own ways. Acknowledging Britzman’s (1989) multi-conceptual understanding of ‘voice’ (literal, metaphorical and political), Thomson’s (2008) participatory visual research with children and young people, for example, attempts to ‘find ways to bring previously unheard voices into scholarly and associated professional conversations’ (p. 3). Significantly, Thomson recognises the importance of conferring competency and agency on vulnerable research participants – of ‘giving voice to the voiceless’ (Thomson, 2008: 3) – by engaging with them in more direct and inclusive ways using a range of different, less-conventional methods.For Thomson and others engaged in participatory visual research, this means involving children as the co-producers of research – by giving them cameras, video equipment, art materials and so on in order to collect their own evidence (see also Kaplan, 2008; Leitch, 2008).

Using photographic participation methods with vulnerable participants

It is certainly the case that there has been something of a sea change in recent years in the ways in which methodological conventions have been reconceptualized and reshaped and thus have allowed for degrees of creativity, subjectivity and even ‘playfulness’ (see Gaver et al., 2004) to be accepted and assimilated in research design and praxis terms as well as in a range of different disciplines and fields (see Plummer, 2001; Richardson and Adams St Pierre, 2005). This change has coincided with an increasing interest in presenting human, lived experience in research that, as Booth (1996) has recognised, can ‘restore [the] emotional content of human experience’ (p. 237). It is clear that a key objective of research studies that include vulnerable groups must be to provide evidence of meaningfulness for these groups and that thus allow insight into their ‘inner worlds’. The methods used must then be faithful, not simply in interpreting and representing experience but in illustrating it and testifying to it.

Participatory photographic methods (and other visually based methods of data collection) would seem to be appropriate ways of achieving these objectives. The combination of the visual and the personal in participatory photographic methods means the images command viewers’ attention, and, when photographs have been taken by the participants themselves, they are absorbing in a personal sense in that they provide direct insight into lived experiences. A number of researchers and commentators have proposed that visual methods, and Photovoice techniques in particular, can also provide insight into unseen or hard-to-reach locations and ‘hidden’ experiences and that they can help to ‘bridge gaps between researcher and researched’ (Joanou, 2009: 214). Pillow (2003) proposes that such methods can also help develop more reciprocal relationships between researchers and participants and that ‘hearing, listening and equalizing the research relationships – doing research “with” instead of “on”’ (p. 179) participants is a good starting point for such reciprocity.

With respect to vulnerable children and young people, the use of photographic methods can help challenge fixed perceptions and assumptions about children’s lack of competency as decision makers or active citizens, and can help to highlight children’s agency and resilience rather than their susceptibility. In her further discussion about using visual research methods with children and young people, Thomson (2008) argues that their ‘critical voice’ is essential in ensuring that, as speakers, they exercise ‘some agency and control of circumstances which previously felt beyond reach’ (p. 5). This is certainly true of vulnerable children in research, but it is also true of research participants more generally. Furthermore, visual methods can prove useful in a range of different empirical contexts and with diverse participant groups, depending on the subject matter, the needs and willingness of respondents to participate and the type of research questions addressed. At the very least, however, visual methods may be more relevant for participants who are unwilling or unable to verbalise their experiences simply because in most cases, and from a purely sensory perspective, seeing (and the visual world) comes before saying (the spoken world).

The inclusive and democratising aspects of photographic participation research can be illustrated in two studies that used these approaches successfully with vulnerable participants (and in which I was involved in working closely with the participants themselves). Uncovering the silent and hidden aspects of children’s lives as carers through the use of participatory research methods was one of the main objectives in the Young Carers Research Group’s (YCRG) photographic participatory study with children who care for parents with serious mental health problems ⁴ (for a full discussion of the findings see Aldridge and Sharpe, 2007). Two further objectives were also outlined: to gain greater insight, through the use of visual methods, into the needs and resilience of young carers, and to gauge the usefulness of the methodology itself. Similar objectives were at the forefront of Sempik and colleagues’ photographic participation study involving people with learning disabilities who attended Social and Therapeutic Horticulture (STH) projects in the United Kingdom (for a full discussion of the findings see Sempik et al., 2005). Here, photographic participation methods were used among participants with learning disabilities in order to promote a shared sensory approach (see Radley and Taylor, 2003) that would accommodate and reflect participants’ different abilities and needs because they could not contribute verbally via interviews. The main objective in both studies was to understand the relevance and efficacy of the photographic method with participants who either could not or would not participate in interviews, thus, in methods that relied on their verbal contributions.

It is important to note that in considering these studies together, the intention is to illustrate the efficacy of participatory visual methods when used collaboratively with vulnerable groups and not to suggest that young carers and adults with learning disabilities are vulnerable in precisely the same ways. ⁵ Furthermore, there were notable contextual differences between the two studies in terms of the participant communities studied. In the study of young carers, participants were drawn from diverse geographical locations and had no community-based connections; quite the opposite was true in fact because, in many respects, children’s lives as carers tend to be home based, isolated and hidden both as a result of circumstance (the demands of caring) and choice – children often do not want to disclose the fact that they live with and help to care for a parent who is ill or disabled and choose not to share these experiences with other children in their local and school communities, for example (see Aldridge and Becker, 2003). Thus, the young carers’ study did not collate and present community-based evidence although the photographic images were analysed and presented collectively to raise awareness of young caring more generally in health and social care contexts and education services (through the use of the online photographic ‘albums’ and also via an exhibition of the photographs made available to schools, etc. – see YCRG, 2012).

The STH study was more community-based and generated visual evidence of the communal STH experience for people with learning disabilities. ⁶ While the intention was not to engage the participants collectively in the analysis of the images – this was made more difficult because of the limited verbal capacities of the participants involved – some of the participants were able to indicate during one-to-one work the images that had some meaning for them. In both studies, the data produced by the participants were subject to content and thematic analyses, which were also informed by the initial selection and editing of the data by the participants themselves to a greater or lesser extent, depending on their willingness and ability to be involved at that stage.

With respect to the outcomes from both of these studies, photographic participation methods proved useful in a number of ways. First, they helped overcome the inherent difficulties involved in accurately representing the ‘reality’ or lived experiences of participants as interpreted through the researcher as photographer. As Radley and Taylor (2003) commented, photographs that have been taken by participants themselves provide

a direct entry into [respondents’] point of view … the technique of photography is a culturally fashioned extension of the senses … so that it provides a potential to question, arouse curiosity, tell in different voices, or see through different eyes from beyond. (p. 79)

This is not to say, however, that both the data produced and the role of the photographer should not be viewed critically when participants adopt photographer–researcher roles nor that we should be unquestioning about the relationship between the photographer’s visual ‘story’ and the viewer’s reception of this (for a full discussion of this issue, see Aldridge and Sharpe, 2007).

Second, the use of photographic participation methods with vulnerable children and adults in these studies emphasised their capacities and abilities as opposed to their vulnerabilities. With respect to the STH study, specifically, the photographic approach shifted the focus away from the seemingly pervasive aspects of a purely pathological perspective that underlines only the deficits of people with learning disabilities and emphasised their capabilities – as photographers, as gardeners and so on. As Lakin (1997) has observed, ‘It has been assumed that “cognitive impairments” – which diagnosticians determine based on performance in vocabulary, memory, mathematics and abstract reasoning – are total impairments, pervasively diminishing everything those so “afflicted” can do’ (p. 4).

Furthermore, using photographic methods among vulnerable children, and adults with learning disabilities, provides an effective showcase for these groups of participants to demonstrate their skills and ‘strong present-orientation’ (Booth, 1996) – if the participants in research cannot tell us, but are able to show us, their experiences through photographic participation methods, then this can serve to emphasise the positive, inclusive advantages of using such a technique, particularly from the point of view of the participants themselves. Such approaches then can help address and resolve, to some degree, the somewhat thorny dilemma of a third-party representation. In these types of studies, the data were generated and, importantly, also owned by the participants themselves. In both the studies, participants were asked to select or show the researchers their ‘favourite’ images (and to attach meaning to these verbally only if they wished or were able to do so), and so they were, as far as possible, also involved collaboratively in the selection and analysis of the visual data. Walker and Colleagues (2008: 170) also note the importance of ownership and participation in participatory visual research and, in particular, in photographic studies that involve vulnerable groups. In their four-country study of abandoned and orphaned children, for example, the authors ‘learnt the importance of children being deeply involved in the process of interpretation and analysis’, and that the children gained a sense of ownership and contextual understanding of the research studies themselves through, the ‘control they had over the process of selection and interpretation’ (Walker and Colleagues, 2008: 170).

The benefits of adopting visual participatory methods and collaborating with vulnerable participants in these ways are demonstrated in the positive outcomes as described in the studies above, particularly in the fact that both groups of participants in these studies (the children and the adults with learning disabilities) enjoyed the act of photography itself and the opportunity to demonstrate visually their experiences in each case. Furthermore, what was also notable was the active role the participants played in constructing their visual narratives and the ways in which they located themselves, variously, as photographers – and thus as directors and producers of their own stories – and as subjects within their own narratives (as photographed subjects) (see Aldridge and Sharpe, 2007 for a full discussion of this point). Notably, a number of STH and young carers’ projects across the United Kingdom have since adopted photographic techniques as ways of working more effectively with service users who find it difficult to talk about their experiences and needs for a number of different reasons. Furthermore, evidence from the young carers’ study has been used to highlight the importance of utilising different ways of engaging with children in health-care consultations (see Aldridge, 2010). In this sense then, the evidence the children collected and produced has been used not only to contribute new knowledge but also to show practitioners alternative and more creative ways of working with children and their families in a range of different contexts. ⁷

Participatory approaches with vulnerable groups – some issues and considerations

Resolving dilemmas in participatory research – towards a multi-method approach

In many respects, where participants are included as co-researchers in participatory studies, this democratising of roles means participants can become the joint or sole ‘owners’ of the research data produced. Thus, distinctions between the researcher and researched and the observer and observed, for example, become less pronounced if not overturned completely. Banks (1995) argues that this has resulted in more studies that promote ‘collaborative representation’ between researchers and participants and others have pointed to the resulting possible collapse of conventional ethical mechanisms in research (see Walker et al., 2008: 173). This then suggests the need to reassess the role and location of the research participant in social research investigations which, in essence, is precisely what participatory and PAR methods have attempted to do. In this context, participation ‘includes the right to have a voice in terms of research ethics’ (Walker et al., 2008: 173). In which case, working collaboratively means working much more closely and inclusively with research participants at all stages of research design and production, including working with them to find ways of resolving various ethical and transformative dilemmas. Of course, working with people who have cognitive impairments, including those with learning disabilities, presents further challenges here in that, as the participants in research studies, they may have little understanding of the processes involved as well as what genuine participation and collaboration actually entails. Nevertheless, it is important to ensure that participants who are vulnerable for whatever reason are not excluded from studies simply because they are perceived as too ‘difficult’ to recruit or include in research investigations that are intended to be about them, for them and with them.

However, with respect to providing evidence that is not only worthwhile to research participants themselves but also has value and credibility in the broader contexts of academia and social and political arenas, we have to look more carefully at the role and efficacy of less-conventional participatory methods in research. Drawing on my own experiences of using these kinds of techniques and participatory visual methods specifically, it is perhaps more appropriate to consider their usefulness in terms of a mixed or multi-method approach. With respect to the photographic participation methods, Donaldson (2001) has argued that, ‘in that company, they look pretty good’ (p. 6). Thus, used in an integrative way, it may be possible to resolve more readily some of the default and design dilemmas in research – between conflicting ‘top down’ pressures and the need for creative, individualistic, ‘bottom up’ approaches to working with vulnerable groups – by rethinking and redesigning research practices along more inclusive, collaborative lines and as part of mixed or multi-method research approaches.

In empirical terms more broadly, there are a number of dilemmas that must remain unresolved to some extent – this is simply the nature of research, and particularly of research that is conducted in different and more challenging contexts. As long as researchers are committed to working with vulnerable research participants in more inclusive and democratic ways, certain personal and philosophical tensions will, then, simply have to be countenanced. It is only if researchers attempt to reconcile these through ‘safe’ or easy option research designs and intentions that real problems will arise, particularly for vulnerable individuals or groups who might be deemed too ‘difficult’ to recruit and are thus excluded from social research studies altogether. It is the role and responsibility of researchers themselves to consider ways of securing funding for and conducting research studies that don’t fit easily into recognised agendas or the demands of the academy, and how to make inroads into dominant modes of research design and production when using methods and approaches that often conflict with recognised ‘ways of doing’ research.

In the context of applied or social action research, where the intention is for research to be transformative in some way, researchers must also negotiate ways of remaining faithful to the experiences and needs of vulnerable research participants – and how to represent their needs appropriately and accurately – and strategies for making sure the messages from research can help to change lives. This can be particularly difficult in the context of straightened and thus more competitive research funding environments and when different types and styles of governments shift the focus (and funding) away from vulnerable populations. For the researcher engaged and committed to working with vulnerable groups in research, the way forward then is far from straightforward, and this is reflected in the methodological, philosophical and professional dilemmas that are inevitable and that need to be addressed if such research is going to change lives.