The first text on focus groups was written by Merton and Kendall (1946), as a result of these writers working with groups during and shortly after the Second World War. In 1956 they expanded their knowledge into a book (Merton et al., 1956). Business and market researchers had used this type of in-depth group interview since the 1920s. It became especially popular in market research in order to gather information about customers’ thoughts and feelings about a product, though initially this type of research was not rooted in the qualitative tradition.

Today the focus group interview is used by a wide variety of researchers in the area of communications, policy, marketing and advertising. Focus groups in the social sciences and health professions have become fashionable since the growth of qualitative research methods in the 1980s. This approach does not rely merely on the ideas of the researcher and a single participant; instead, the members of the group generate new questions and answers through verbal interaction. Through these group interviews, researchers are able to discover the needs and feelings of their clients, the perceptions and attitudes of their colleagues, and they can examine the thoughts of decision makers. The cultural values and beliefs of people can also be explored this way.

Focus groups produce thoughts and opinions about a topic relevant to health care, treatment evaluation and illness experiences. Many examples are reported in nursing and social science journals.

Focus groups are characterised by interaction between the participants from which researchers discover how people think and feel about particular issues. It is not the intention to examine a wide variety of issues in one study; these groups are set up to explore a specific issue rather than general topics which are more often investigated in marketing or political focus groups.

Focus group members respond to the interviewer and to each other. The questions might start with eliciting knowledge about a specific condition, the use of a drug, a method of intervention, or by putting the members at ease but should soon go on to a discussion of feelings or thoughts. Different reactions stimulate debate about the topic because group members respond to each other. Discussions in groups might help not only in the development of ideas about problems and questions which researchers have not thought about before but also by finding answers to some of these questions and solutions to problems.

In nursing and other healthcare arenas, focus groups are used to

These are just some of the functions of focus group interviewing. The ultimate goal for the researcher is to understand the reality of the participants, and not to make decisions about a specific issue or problem, although future actions may be based on the findings of the focus group interviews. Focus groups differ from individual interviews in that they depend on the stimulus that participants gain from each other, and that they discuss both unique and shared perceptions and experiences. Focus groups can be used as a single source of interview data or in conjunction with one-to-one interviews.

The sample is linked closely to the research topic. The people who are interviewed in a focus group usually have similar roles or experiences. They may be colleagues who share the same speciality, use the same technical equipment or nursing procedures or patients who suffer from the same condition. The purpose of the focus group generally determines its composition and size. Morgan (1998b) claims that a small group is better for controversial or complex topics, while larger groups tend to have lower levels of involvement with less highly intense topic areas. We have also found that in smaller groups individuals can be heard more clearly, though groups with larger numbers of participants might generate more ideas.

Morgan (1998a) suggests that well-defined criteria are needed for this selection. These might include demographic factors, gender, ethnic group membership and specific experiences or conditions. Participants in focus groups will have had common experiences, have the same condition or receive the same treatment. For instance, if a doctor wishes to interview a group of people with diabetes, she or he obviously involves individuals with this condition in the focus groups. A midwife might obtain the feelings and thoughts of pregnant women or new mothers by small focus groups. Colleagues who are interviewed generally share common interests, work in similar settings or perform similar tasks. If the interviewer wants the thoughts of colleagues from a psychiatric setting, for example, then the sample has to be composed of nurses with psychiatric experience. Students too can be interviewed in focus groups about perspectives on their education. Health promotion often is a topic for research.

The choice of the members of focus groups depends on a condition or experiences that potential participants have in common. Although group members share these, it does not mean that they all have the same views, or that they come from the same background or organisation. It might be useful to recruit members from naturally occurring groups such as antenatal classes, patient support groups or carers. While they have similar experiences, they are nevertheless heterogeneous in other ways, and so could illuminate the topic from all sides.

The number of focus groups depends on the needs of the researcher and the demands of the topic area. For one research project, the usual number is about three or four, but the actual number depends on the complexity of the research topic. If the sample of participants is heterogeneous, more groups are needed.

Studies with large focus groups and many informants are more difficult. Group sessions can last from one to three hours. We must stress, however, that three-hour interviews with patients would be far too long and demanding. In market research, participants are paid for their time and effort but not usually in healthcare research, because this would coerce the informants and squander resources. Much new information is gained in initial groups as the researcher can follow up the ideas obtained in subsequent interviews. As in other qualitative research, important themes emerge often at an early stage, although some serendipitous results might be found in a later phase.

Each group might contain between four and twelve people, but six is probably the optimum number as it is large enough to provide a variety of perspectives and small enough not to become disorderly or fragmented. Indeed, one of our colleagues found that in her experience, even a group of six was too large and that the optimum number of members in the group was three, but the number could of course vary depending on the topic or the background of group members. Greenbaum (1998), a market researcher, however, claims that group dynamics work better if the group is not too small. The larger the group, however, the more difficult the transcription becomes. When several people start talking together and the group is lively and noisy, it can be difficult to distinguish voices.

There may well be a difference between groups who come together for market research purposes and those who gather for health research. The former will feel much less vulnerable because the area of discussion is rarely threatening or sensitive. The nature of the topic area is of importance: focus groups in which sensitive topics are discussed are more difficult to facilitate.

Members of the group, although sharing common experiences, do not have to know each other. In a group of immediate colleagues or friends, private thoughts or ideas might not be revealed, although occasionally the opposite could be true. One individual is more likely to dominate others and the past history of the group may inhibit or lead individuals in a particular direction. In healthcare research, familiarity between participants, or participants and researchers could be useful because the ‘warm-up’ time – the time where informants get to know each other to facilitate interaction – is shorter, and the researcher can focus on the topic immediately. Stewart and Shamdasani (2007), for instance, believe that compatibility among group members is more productive than conflict or polarisation, although this too depends on the topic; sometimes conflict can generate new and different ideas.

Gender and age of the group members affect the quality and level of interaction and through this the data. For instance, evidence shows greater diversity of ideas in single sex groups than in those of mixed gender according to Stewart and Shamdasani. Mixed gender groups tend to be more conforming because of the social interaction between males and females; both groups sometimes tend to ‘perform’ for each other.

Focus group interviews must be planned carefully. The informants are contacted well in advance of the interviews and reminded a few days before they start. As in other types of inquiry, ethical and access issues are considered. The environment for a focus group is important as the room must be big enough to contain the participants and the tape-recorder placed in an advantageous location, where they can all be heard and recorded. For focus group work, it is essential to have a top quality tape-recorder. Merton and King (1990) suggest a spatial arrangement of a circle or semi-circle, which seems the most successful seating arrangement.

The group interviews should have a clearly identified agenda otherwise they deteriorate into vague and chaotic discussions (Stewart and Shamdasani, 2007). Morgan (1997) believes in the importance of time management because both interviewer and informants have limited time. Time management is one of the tasks of the facilitator. Focus groups are more productive if the time for interchange is not too short. Usually focus group interviews last around to 2 hours but this might depend on age, vulnerability or power of concentration of participants.

From the beginning the researcher establishes ground rules, so that all group members know how to proceed. Researchers plan the initial questions and prompts. When the interviews start, the interviewer puts the group at ease and introduces the topic to be debated. Strategies such as showing a film or telling a story related to the topic sometimes stimulate interaction. Kitzinger and Barbour (1999) also suggest such stimulus material as vignettes or photographs. Researchers often adopt the strategy of asking stimulus questions and generally proceed from the more general to the specific, just as in other qualitative interviews. Involving all the participants, rather than letting a few individuals dominate the situation demands diplomacy and would be easier with a smaller group. Extreme views in a group of people are balanced out by the reactions of the majority when debating questions. As suggested before, focus groups can be combined with individual interviews, observation or other methods of data collection but this is not essential.

In focus groups, as in all other research, ethical issues must be considered. Confidentiality, in particular, could be problematic in group interviews as members of the group might discuss the findings in other settings and situations. They should be reminded to keep the discussions confidential. Anonymity cannot be guaranteed, as members of the group might be able to identify other participants even when researchers only use first names. Participants may make remarks that are hurtful to others, or show prejudice, and the researcher has to find ways to deal with this.

The involvement of the interviewer

The interviewer becomes the facilitator or moderator in the group discussion although it could be useful to have another person who takes notes. In health research, the health professional is usually the interviewer (while in market research focus groups, professional moderators are employed). In a small project, a single interviewer usually facilitates the groups. The presence of a note-taker who can make fieldnotes, draw diagrams with the names of participants and generally help with practical matters, could be very useful. The researcher should have the particular qualities of the in-depth interviewer: flexibility, open-mindedness and skill in eliciting information. The creation of an open and non-threatening group climate is one of their initial important tasks.

Researchers must be able to stimulate discussion and have insight and interest in the ideas of the informants. The leadership role of the moderators demands abilities above that of the one-to-one interviewer. They must have the social and refereeing skills to guide the members towards effective interaction and sometimes be able to exert control over informants and topic without directing the debate or coercing the participants. If the group feels at ease with the interviewer, the interaction will be open and productive, and the participants will be comfortable about disclosing their perceptions and feelings. Researchers might experience difficulties with particular groups such as teenagers, while getting together groups of disabled people may present practical problems in the available space.

Morgan (1997) advises that the interviewers hold back on questioning if they want to examine the real feelings of participants; much of the discussion evolves from the dynamics of group interaction. Indeed Kitzinger (2005: 57) claims that ‘a defining feature of focus group research is using the interaction between research participants to generate data and giving attention to that interaction as part of the analysis’. This non-directive approach has particular importance in exploratory research where perceptions are examined. High involvement of the interviewer leads more quickly to the core of the topic, but special facilitation skills are needed if the focus groups are going to be successful. The interviewers should not express their own biases or assumptions in the focus groups. A special relationship with a specific individual, an affirmative nod at something of which the interviewer approves, or a lack of encouragement for unexpected or unwelcome answers may bias the interviews too. Again, group behaviour is an important factor. Polarisation of views may generate a difficult group climate. Although conflicts of opinion can produce valuable data, the interviewer must defuse personal hostility between members, which demands good facilitating skills. Gestures and facial expressions have to be controlled to show members of the group that the interviewer is non-judgemental and values the views of all participants. Streubert Speziale and Rinaldi Carpenter (2007) argue that a good facilitator can help the group to avoid ‘group think’ and offence to some participants.

The principles of qualitative data analysis are similar to those of other non-structured or semi-structured interviews. Most often the interviews are recorded, and initially the researcher listens several times to each tape before making transcripts. Although this method has been used in market research, it is difficult to identify individuals’ voices on a tape. The problem of identification might be overcome with videotaping, but Sim (1998) suggests that this might inhibit participants, particularly when they discuss a sensitive issue.

All tapes, fieldnotes and memos are dated and labelled. A wide margin is left on the transcript for coding and categorising. The transcription should include laughter, notes about pauses and emphasis, and the researcher makes fieldnotes on anything unusual, interesting or contradictory and writes memos about theoretical ideas while listening, transcribing and reading. It is important to be clear about who says what, because this can identify those individuals who try to dominate the discussion. The interviewer could note this while listening to the tape. At the listening stage, major themes and patterns can already be found. It is important, however, that researchers focus on the context of group interaction not just on the comments of particular individuals but on all of them (Asbury, 1995). This interaction might stimulate thought in the participants but it could also intimidate some or encourage others.

Interviewers code paragraphs and sentences by extracting the essence of ideas within them and using labels which they put into the margin of the transcript. Through a reduction of these codes into larger categories, themes and ideas will be found. As in other types of qualitative research, the frequency of themes that are found is not as important as their significance; some obviously have priority over others for the specific study. The method of analysis in focus groups is similar to those of other approaches; in fact, focus groups can be analysed by thematic analysis (see Chapter 17) or another form of qualitative analysis.

The analyst repeats the process with each focus group interview and compares the transcripts. The major themes arising from individual interviews are then connected with each other; topics in one interview will overlap with those of other focus groups. Once these themes have been formulated, the patterns described and their meaning interpreted, the literature connected with these ideas is discussed. The appropriate literature becomes confirmation or challenge to the researcher’s findings as in other qualitative research.

Researchers substantiate their work with relevant quotes from the participants, showing the data from which the patterns and constructs arise; excerpts from interactions are part of these quotes. Although patterns and consistencies are important for reporting, individual comments are also important as they might form an alternative response to the rest of the data. If there are many such deviant cases, it might be useful to add one-to-one interviews to explore these further.

To write up the study, the interviewers develop a storyline, that is, they must produce an account that is readable and clear. The main concerns of the participants have to emerge from the report as the most important parts of the story. The findings from the focus group interviews are often used as a basis for action.

In general the advantages and limitations in this approach are those of all qualitative interviews, but there are a number of strengths and weaknesses specific to focus groups (Stewart and Shamdasani, 2007). The main strength is the production of data through social interaction. The dynamic interaction stimulates the thoughts of participants and reminds them of their own feelings about the research topic. Informants build on the answers of others in the group. Second, on responding to each other’s comments, informants might generate new and spontaneous ideas, which researchers had not thought of before or during the interview. Through interaction informants remember forgotten feelings and thoughts. Third, all the participants, including the interviewer, have the opportunity to ask questions, and these will produce more ideas than individual interviews. Kitzinger (2005) suggests that group interaction gives courage to the informants to mention even sensitive topics. The interview might empower participants because as group members they often feel more able to express their views.

The researcher has the opportunity for prompts and questions for clarification just like the other members of the group. These probes will produce more ideas than individual interviews, and the answers show the participants’ feelings about a topic and the priorities in the situation under discussion. The researcher can clarify conflicts between participants and ask about the reasons for these differing views. Focus groups produce more data in the same space of time; this could make them cheaper and quicker than individual interviews. Some people dislike opening up their inner thoughts in public and may be reluctant to answer some questions – one of the reasons for careful selection of participants. Though the presence of others might inhibit disclosure, which is a disadvantage in these settings, it can also allow individuals to be quiet and obviate the need to respond if they do not wish to disclose something.

There are also some disadvantages. The researcher generally has more difficulty managing the debate and less control over the process than in one-to-one interviews. As group members interact throughout the interview, one or two individuals may dominate the discussion and influence the outcome or perhaps even introduce bias, as the other members may be merely compliant. The group effect may, as Carey and Smith (1994) suggest, lead to conformity or to convergent answers. They use the term ‘censoring’, by which they mean the critical stance of group members towards each other. The participants affect each other, while in individual interviews the ‘real’ feelings of the individual informant may be more readily revealed. A person who is unable to verbalise feelings and thoughts will not make a good informant in focus groups. Indeed, Merton and King (1990) stress the importance of educational homogeneity of the group. If group members have similar educational backgrounds, the chance for contribution from all members is greater. The status of a few well-educated individuals would inhibit the rest of the members in the group and might even silence them, and therefore similarity of social background is useful. The group members might know each other before the meeting, and it is important to take this into account. This means that sampling procedures which determine the composition of the group, are of paramount importance.

The group climate can inhibit or fail to stimulate an individual or it can, of course, be stimulating and lively and generate more data. Where a researcher feels certain that confrontation and conflict is likely to occur between potential group members, she or he has to be sensitive to group feelings and reconcile their ideas. Conflict can be destructive but can also generate rich data. In any conflict situation, ethical issues must be carefully considered. Sim (1998) identifies some problems with focus groups.