Doing Focus Groups. Rosaline Barbour
– carried out focus groups with people with Parkinson’s disease, their family carers and professionals in order to establish their views on quality of life measures. The researchers reported that this exercise identified important new themes.
However, focus groups have added potential – particularly for the practitioner– researcher – for use in overtly action-research-oriented projects. Crabtree et al. argue that ‘it is possible to use focus groups as a data collection tool and an intervention simultaneously’ (Crabtree et al., 1993, p. 146).
Claims and challenges in focus
The varying terms employed by different writers (or disciplinary representatives) who label their method as involving ‘group interviews’, ‘focus group interviews’ or ‘focus group discussions’ reflect different assumptions as to the purpose of the study and the nature and status of the data that are generated. Some usages emphasize the capacity of focus groups to access ‘views’ and may, at times, implicitly reference the quantitative tradition of data collection, privileging the notion of fixed ‘attitudes’. Others focus, either implicitly or explicitly, on a social constructionist model that emphasizes the potential of focus groups to elicit exchanges between participants as they co-construct perspectives and responses. (This is discussed in more detail in Chapter 2 in relation to ‘epistemological and ontological’ underpinnings of focus group research.) The definition that I continue to apply (first outlined in Chapter 1) is suitably broad to afford space for both applied and more theoretical usages of focus groups: ‘Any group discussion may be called a focus group as long as the researcher is actively encouraging of, and attentive to, the group interaction’ (Kitzinger and Barbour, 1999, p. 20).
Being actively encouraging of group interaction relates, most obviously, to running the focus group discussion and ensuring that participants talk amongst themselves rather than interacting only with the researcher, or ‘moderator’. However, it also relates to the preparation required in developing a topic guide and selecting stimulus materials that will encourage interaction, as well as to decisions made with regard to group composition, in order to ensure that participants will have enough in common with each other to make discussion seem appropriate, yet have sufficiently varying experiences or perspectives in order to allow for some debate or differences of opinion. Likewise, although being attentive to group interaction refers to the process of moderating discussions, with the researcher picking up on differences in views or emphasis of participants and exploring these, it also relates to the importance of paying attention to group interaction: to group dynamics and to the activities engaged in by the group – whether this be forming a consensus, developing an explanatory framework, interpreting health promotion messages, or weighing up competing priorities. Later chapters in this book are concerned with providing advice on all of these aspects of research design, the running of focus groups and analyzing the data generated.
Focus groups, in common with other qualitative methods, excel at providing insights into process rather than outcome. This, however, is sometimes overlooked by researchers who employ focus groups as a method. A common usage is the so-called ‘nominal group technique’, which has proved so popular in health services research. Literally meaning ‘a researcher-convened rather than naturally occurring group’ – a group in name only – the most common variant of ‘nominal groups’ involves employing a ranking exercise in order to encourage participants to determine their priorities. While I would contend that important insights can be gained by paying detailed attention to the discussion generated during the process of debating and weighing up competing priorities, many proponents of this approach concentrate their efforts, instead, on the outcome of such deliberations. A notable exception is provided by Demant and Ravn (2010) who capitalized on the data-generating potential afforded by a ranking exercise to allow young Danish people to articulate, explore, debate and refine their ideas about drugs and risk. (This study is discussed in more detail in Chapter 6 on generating data.)
Although some commentators have bemoaned the tendency of focus groups to produce consensus, others, such as Myers and Macnaghten (1999), have pointed out that, in the event, many groups do not develop a consensus, and that, moreover, it is the interchanges between participants that constitute the most valuable data for the researcher attempting to gain insight into group processes; not the outcome of the discussion.
All comments made during focus groups are highly dependent upon context and are contingent upon group members’ responses to others’ contributions and the dynamics of that particular group. Lehoux et al. (2006), for example, have questioned the emerging consensus about what constitutes the patient’s view, highlighting the extent to which this is context-dependent and a product of the focus group discussion setting. (This issue is re-visited in Chapters 8 and 9 which discuss the use of group interaction in the process of analysis of interaction in groups.) As Billig (1991) reminds us, views expressed in focus groups are highly specific and are ‘bound up in the argument [that is] happening’. It is misguided to attempt to extrapolate from focus group discussion to attempt to measure individuals’ attitudes. Although not explicitly utilizing focus groups as a ‘back-door’ route to survey data, some researchers, nevertheless, may express frustration regarding the perceived ‘slipperiness’, or elusiveness, of views throughout focus group discussions. Participants frequently change their minds about issues in the course of discussion, particularly where focus groups address a topic to which they had not previously paid a great deal of attention. Researchers are in danger of treating views as if they exist independently of our focus group discussions, when it would be more helpful to regard the research encounter itself as a ‘site of performance’ (Brannen and Pattman, 2005, p. 53). Virtually without fail, close analysis of focus group discussions highlights inconsistencies and contradictions. This is a problem only if one views attitudes as fixed. As David Morgan (1988) has observed: ‘Focus groups are useful when it comes to investigating what participants think, but they excel at uncovering why participants think as they do’ (1988, p. 25).
If looked at through a different lens, these so-called ‘slippery’ views can be perceived as a resource rather than as a problem. Focus groups excel at allowing us to study the processes of attitude formation and the mechanisms involved in interrogating and modifying views. Some researchers have used this advantage to illuminate, for example, the influences on public attitudes to distribution of donor liver grafts (Wilmot and Ratcliffe, 2002). Focus group researchers are also well-placed to interrogate shifts in views over time – for example, Järvinen and Demant (2011), who traced the evolution of the views of young Danish people about drug use, through comparing data generated via focus groups held, respectively, when participants were 14–15 years, 15–16 years and 18–19 years of age. (This study is discussed in more detail in Chapter 4 in relation to research design.)
Involving semi-structured topic guides (see Chapter 6) and allowing for in-depth consideration of open-ended questions and stimulus materials, focus groups have the capacity to reflect issues and concerns salient to participants rather than closely following the researcher’s agenda. This means that the resulting data can yield surprises. For example, participants may take factors into account in their deliberations that researchers have not anticipated and this may highlight the relevance for the researcher of alternative explanations for perceptions or behaviour – or even of new theoretical frameworks that can usefully be brought to bear in analysis.
Focus groups are also especially well-suited to uncovering participants’ misconceptions and how these can arise. It is for this reason that focus groups have been used so frequently, to good advantage, in gauging the impact of health promotion campaigns. Keane et al. (1996) carried out research into African-American beliefs about immunization for infants, conceptualization of illness and efficacy of vaccines. Interestingly, the focus group discussions