Qualitative Research: Introducing focus groupsBMJ 1995; 311 doi: https://doi.org/10.1136/bmj.311.7000.299 (Published 29 July 1995) Cite this as: BMJ 1995;311:299
- Jenny Kitzinger, research fellowa
This paper introduces focus group methodology, gives advice on group composition, running the groups, and analysing the results. Focus groups have advantages for researchers in the field of health and medicine: they do not discriminate against people who cannot read or write and they can encourage participation from people reluctant to be interviewed on their own or who feel they have nothing to say.
This is the fifth in a series of seven articles describing non-quantitative techniques and showing their value in health research
Rationale and uses of focus groups
Focus groups are a form of group interview that capitalises on communication between research participants in order to generate data. Although group interviews are often used simply as a quick and convenient way to collect data from several people simultaneously, focus groups explicitly use group interaction as part of the method. This means that instead of the researcher asking each person to respond to a question in turn, people are encouraged to talk to one another: asking questions, exchanging anecdotes and commenting on each other's experiences and points of view.1 The method is particularly useful for exploring people's knowledge and experiences and can be used to examine not only what people think but how they think and why they think that way.
Focus groups were originally used within communication studies to explore the effects of films and television programmes,2 and are a popular method for assessing health education messages and examining public understandings of illness and of health behaviours.3 4 5 6 7 They are widely used to examine people's experiences of disease and of health services.8 9 and are an effective technique for exploring the attitudes and needs of staff.10 11
The idea behind the focus group method is that group processes can help people to explore and clarify their views in ways that would be less easily accessible in a one to one interview. Group discussion is particularly appropriate when the interviewer has a series of open ended questions and wishes to encourage research participants to explore the issues of importance to them, in their own vocabulary, generating their own questions and pursuing their own priorities. When group dynamics work well the participants work alongside the researcher, taking the research in new and often unexpected directions.
Group work also helps researchers tap into the many different forms of communication that people use in day to day interaction, including jokes, anecdotes, teasing, and arguing. Gaining access to such variety of communication is useful because people's knowledge and attitudes are not entirely encapsulated in reasoned responses to direct questions. Everyday forms of communication may tell us as much, if not more, about what people know or experience. In this sense focus groups reach the parts that other methods cannot reach, revealing dimensions of understanding that often remain untapped by more conventional data collection techniques.
Some potential sampling advantages with focus groups
Do not discriminate against people who cannot read or write
Can encourage participation from those who are reluctant to be interviewed on their own (such as those intimidated by the formality and isolation of a one to one interview)
Can encourage contributions from people who feel they have nothing to say or who are deemed “unresponsive patients” (but engage in the discussion generated by other group members)
Tapping into such interpersonal communication is also important because this can highlight (sub)cultural values or group norms. Through analysing the operation of humour, consensus, and dissent and examining different types of narrative used within the group, the researcher can identify shared and common knowledge.12 This makes focus groups a data collection technique particularly sensitive to cultural variables—which is why it is so often used in cross cultural research and work with ethnic minorities. It also makes them useful in studies examining why different sections of the population make differential use of health services.13 14 For similar reasons focus groups are useful for studying dominant cultural values (for example, exposing dominant narratives about sexuality15) and for examining work place cultures—the ways in which, for example, staff cope with working with terminally ill patients or deal with the stresses of an accident and emergency department.
The downside of such group dynamics is that the articulation of group norms may silence individual voices of dissent. The presence of other research participants also compromises the confidentiality of the research session. For example, in group discussion with old people in long term residential care I found that some residents tried to prevent others from criticising staff—becoming agitated and repeatedly interrupting with cries of “you can't complain”; “the staff couldn't possibly be nicer.” On the one hand, such interactions highlighted certain aspects of these people's experiences. In this case, it showed some resident's fear of being “punished” by staff for, in the words of one woman, “being cheeky.” On the other hand, such group dynamics raise ethical issues (especially when the work is with “captive” populations) and may limit the usefulness of the data for certain purposes (Scottish Health Feedback, unpublished report).
However, it should not be assumed that groups are, by definition, inhibiting relative to the supposed privacy of an interview situation or that focus groups are inappropriate when researching sensitive topics. Quite the opposite may be true. Group work can actively facilitate the discussion of taboo topics because the less inhibited members of the group break the ice for shyer participants. Participants can also provide mutual support in expressing feelings that are common to their group but which they consider to deviate from mainstream culture (or the assumed culture of the researcher). This is particularly important when researching stigmatised or taboo experiences (for example, bereavement or sexual violence).
Focus group methods are also popular with those conducting action research and those concerned to “empower” research participants because the participants can become an active part of the process of analysis. Indeed, group participants may actually develop particular perspectives as a consequence of talking with other people who have similar experiences. For example, group dynamics can allow for a shift from personal, self blaming psychological explanations (“I'm stupid not to have understood what the doctor was telling me”; “I should have been stronger—I should have asked the right questions”) to the exploration of structural solutions (“If we've all felt confused about what we've been told maybe having a leaflet would help, or what about being able to take away a tape recording of the consultation?”).
Some researchers have also noted that group discussions can generate more critical comments than interviews.16 For example, Geis et al, in their study of the lovers of people with AIDS, found that there were more angry comments about the medical community in the group discussions than in the individual interviews: “perhaps the synergism of the group ‘kept the anger going’ and allowed each participant to reinforce another's vented feelings of frustration and rage.17 A method that facilitates the expression of criticism and the exploration of different types of solutions is invaluable if the aim of research is to improve services. Such a method is especially appropriate when working with particular disempowered patient populations who are often reluctant to give negative feedback or may feel that any problems result from their own inadequacies.19
Conducting a focus group study
SAMPLING AND GROUP COMPOSITION
Focus group studies can consist of anything between half a dozen to over fifty groups, depending on the aims of the project and the resources available. Most studies involve just a few groups, and some combine this method with other data collection techniques. Focus group discussion of a questionnaire is ideal for testing the phrasing of questions and is also useful in explaining or exploring survey results.19 20
Although it may be possible to work with a representative sample of a small population, most focus group studies use a theoretical sampling model (explained earlier in this series21) whereby participants are selected to reflect a range of the total study population or to test particular hypotheses. Imaginative sampling is crucial. Most people now recognise class or ethnicity as important variables, and it is also worth considering other variables. For example, when exploring women's experiences of maternity care or cervical smears it may be advisable to include groups of lesbians or women who were sexually abused as children.22
Most researchers recommend aiming for homogeneity within each group in order to capitalise on people's shared experiences. However, it can also be advantageous to bring together a diverse group (for example, from a range of professions) to maximise exploration of different perspectives within a group setting. However, it is important to be aware of how hierarchy within the group may affect the data (a nursing auxiliary, for example, is likely to be inhibited by the presence of a consultant from the same hospital).
The groups can be “naturally occurring” (for example, people who work together) or may be drawn together specifically for the research. Using preexisting groups allows observation of fragments of interactions that approximate to naturally occurring data (such as might have been collected by participant observation). An additional advantage is that friends and colleagues can relate each other's comments to incidents in their shared daily lives. They may challenge each other on contradictions between what they profess to believe and how they actually behave (for example, “how about that time you didn't use a glove while taking blood from a patient?”).
It would be naive to assume that group data are by definition “natural” in the sense that such interactions would have occurred without the group being convened for this purpose. Rather than assuming that sessions inevitably reflect everyday interactions (although sometimes they will), the group should be used to encourage people to engage with one another, formulate their ideas, and draw out the cognitive structures which previously have not been articulated.
Finally, it is important to consider the appropriateness of group work for different study populations and to think about how to overcome potential difficulties. Group work can facilitate collecting information from people who cannot read or write. The “safety in numbers factor” may also encourage the participation of those who are wary of an interviewer or who are anxious about talking.23 However, group work can compound difficulties in communication if each person has a different disability. In the study assessing residential care for the elderly, I conducted a focus group that included one person who had impaired hearing, another with senile dementia, and a third with partial paralysis affecting her speech. This severely restricted interaction between research participants and confirmed some of the staff's predictions about the limitations of group work with this population. However, such problems could be resolved by thinking more carefully about the composition of the group, and sometimes group participants could help to translate for each other. It should also be noted that some of the old people who might have been unable to sustain a one to one interview were able to take part in the group, contributing intermittently. Even some residents who staff had suggested should be excluded from the research because they were “unresponsive” eventually responded to the lively conversations generated by their coresidents and were able to contribute their point of view. Communication difficulties should not rule out group work, but must be considered as a factor.
RUNNING THE GROUPS
Sessions should be relaxed: a comfortable setting, refreshments, and sitting round in a circle will help to establish the right atmosphere. The ideal group size is between four and eight people. Sessions may last one to two hours (or extend into a whole afternoon or a series of meetings). The facilitator should explain that the aim of focus groups is to encourage people to talk to each other rather than to address themselves to the researcher. The researcher may take a back seat at first, allowing for a type of “structured eavesdropping.”24 Later on in the session, however, the researcher can adopt a more interventionist style: urging debate to continue beyond the stage it might otherwise have ended and encouraging the group to discuss the inconsistencies both between participants and within their own thinking. Disagreements within groups can be used to encourage participants to elucidate their point of view and to clarify why they think as they do. Differences between individual one off interviews have to be analysed by the researchers through armchair theorising; differences between members of focus groups should be explored in situ with the help of the research participants.
The facilitator may also use a range of group exercises. A common exercise consists of presenting the group with a series of statements on large cards. The group members are asked collectively to sort these cards into different piles depending on, for example, their degree of agreement or disagreement with that point of view or the importance they assign to that particular aspect of service. For example, I have used such cards to explore public understandings of HIV transmission (placing statements about “types” of people into different risk categories), old people's experiences of residential care (assigning degrees of importance to different statements about the quality of their care), and midwive's views of their professional responsibilities (placing a series of statements about midwive's roles along an agree-disagree continuum). Such exercises encourage participants to concentrate on one another (rather than on the group facilitator) and force them to explain their different perspectives. The final layout of the cards is less important than the discussion that it generates.25 Researchers may also use such exercises as a way of checking out their own assessment of what has emerged from the group. In this case it is best to take along a series of blank cards and fill them out only towards the end of the session, using statements generated during the course of the discussion. Finally, it may be beneficial to present research participants with a brief questionnaire, or the opportunity to speak to the researcher privately, giving each one the opportunity to record private comments after the group session has been completed.
Ideally the group discussions should be tape recorded and transcribed. If this is not possible then it is vital to take careful notes and researchers may find it useful to involve the group in recording key issues on a flip chart.
ANALYSIS AND WRITING UP
Analysing focus groups is basically the same as analysing any other qualitative self report data.21 26 At the very least, the researcher draws together and compares discussions of similar themes and examines how these relate to the variables within the sample population. In general, it is not appropriate to give percentages in reports of focus group data, and it is important to try to distinguish between individual opinions expressed in spite of the group from the actual group consensus. As in all qualitative analysis, deviant case analysis is important—that is, attention must be given to minority opinions and examples that do not fit with the researcher's overall theory.
The only distinct feature of working with focus group data is the need to indicate the impact of the group dynamic and analyse the sessions in ways that take full advantage of the interaction between research participants. In coding the script of a group discussion, it is worth using special categories for certain types of narrative, such as jokes and anecdotes, and types of interaction, such as “questions,” “deferring to the opinion of others,” “censorship,” or “changes of mind.” A focus group research report that is true to its data should also usually include at least some illustrations of the talk between participants, rather than simply presenting isolated quotations taken out of context.
Tapping into interpersonal communication can highlight cultural values or group norms
This paper has presented the factors to consider when designing or evaluating a focus group study. In particular, it has drawn attention to the overt exploitation and exploration of interactions in focus group discussion. Interaction between participants can be used to achieve seven main aims:
To highlight the respondent's attitudes, priorities, language, and framework of understanding;
To encourage research participants to generate and explore their own questions and develop their own analysis of common experiences;
To encourage a variety of communication from participants—tapping into a wide range and form of understanding;
To help to identify group norms and cultural values;
To provide insight into the operation of group social processes in the articulation of knowledge (for example, through the examination of what information is censured or muted within the group);
To encourage open conversation about embarrassing subjects and to permit the expression of criticism;
Generally to facilitate the expression of ideas and experiences that might be left underdeveloped in an interview and to illuminate the research participant's perspectives through the debate within the group.
Group data are neither more nor less authentic than data collected by other methods, but focus groups can be the most appropriate method for researching particular types of question. Direct observation may be more appropriate for studies of social roles and formal organisations27 but focus groups are particularly suited to the study of attitudes and experiences. Interviews may be more appropriate for tapping into individual biographies,27 but focus groups are more suitable for examining how knowledge, and more importantly, ideas, develop and operate within a given cultural context. Questionnaires are more appropriate for obtaining quantitative information and explaining how many people hold a certain (pre-defined) opinion; focus groups are better for exploring exactly how those opinions are constructed. Thus while surveys repeatedly identify gaps between health knowledge and health behaviour, only qualitative methods, such as focus groups, can actually fill these gaps and explain why these occur.
Focus groups are not an easy option. The data they generate can be as cumbersome as they are complex. Yet the method is basically straightforward and need not be intimidating for either the researcher or the researched. Perhaps the very best way of working out whether or not focus groups might be appropriate in any particular study is to try them out in practice.
Morgan D. Focus groups as qualitative research. London: Sage, 1988.
Kreuger R. Focus groups: a practical guide for applied research. London: Sage, 1988.