22.214.171.124 Focus groups A focus group is a particular form of group interview. It usually consists of group of people in a room who are asked to draw on their personal experience to discuss a specific topic selected by the researcher. The structure of the discussion is often loose, lacking specific questions and inviting any views on the general topic of the focus group.
When groups consist of up to six people they usually discuss the topic in the group as a whole. When focus groups are larger, the process often involves the group breaking into smaller clusters to explore the topic in detail followed by a plenary where the key points from the subgroup are discussed by the whole group.
A co-ordinator or chair, often referred to as a moderator, is necessary to ensure that the discussion is orderly and inclusive. The discussion and its outcomes need to be recorded in some fashion.
There are various definitions of focus groups but the essence is that discussion is organised, albeit not controlled by the researcher, to address a research topic and the contributions of the group constitute the scientific data for subsequent analysis.
Issues are raised by the members of the group through collective activity and the interaction within the group leads to modifications or enhancements of the issues raised.
AnitaGibbs (1997) suggested that it is difficult to get a representative sample in a focus group but that is rarely the aim of focus groups. (See CASE STUDY: Generalisation in qualitative research) Usually, the researcher targets particular types of individuals for the group and hopes to recruit the desired number of people that fit the typification, representativeness rarely comes into the equation in practice. For example, Mary Ferguson et al. (2011, p. 29) suggested, in their study of prisoners, that
prisoners and ex-prisoners who have particular issues and concerns may have been overrepresented in their focus groups as the participants were self-selected.
This may limit the generalizability of their comments to the broader prison/juvenile detention population in New South Wales. Despite this, the views of opportunities and barriers to accessing health care within prison were remarkably consistent across all participants and groups. There were no demonstrable differences in the views expressed by males and females across age groups within prison or in the community.
In many cases, the research is guided by phenomenological and sometimes critical dialectical epistemological concerns and thus positivist concerns about statistical representativeness are not relevant. Indeed, focus groups are a useful tool to get at people's meanings and as Gibbs (1997, np) suggested:
focus groups elicit information in a way which allows researchers to find out why an issue is salient, as well as what is salient about it (Morgan, 1988). As a result, the gap between what people say and what they do can be better understood (Lankshear 1993). If multiple understandings and meanings are revealed by participants, multiple explanations of their behaviour and attitudes will be more readily articulated.
126.96.36.199.1 History of focus groups Focus groups as a research tool are usually regarded as having started with Paul Lazarsfeld's research in the 1940's on the audiences emotional reactions to radio programmes. Robert Merton, a colleague of Lazarsfeld, also used the method to explore radio audiences' reception to persuasive messages encouraging war bond pledges and, later, the impact of training films on World War II soldiers. In these early days, the method came to be called focussed interviewing (Merton et al., 1956). Merton et al. explicitly drew on Rogers' (1945) concept of non-directive interviewing in formulating their concept of the focused interview. There was something a hiatus in the use of focus groups in social research, not least as the originators became more and more preoccupied with quantitative, statistical, survey-based study.
Market researchers used focus groups as 'a means of uncovering consumers' psychological motivations; (Suter, 2000, np). Indeed, one might argue that the standard focus group methodology groups began as a market research tool (Morgan, 1988) and continues to be used in that context. For example, a group might be asked what makes a good supermarket?
Focus groups in social researchers seem to have gone in and out of fashion throughout the last 70 years. Focus groups have also come to be used in medical research over the last twenty years (see, for example,Powell and Single, 1996).
188.8.131.52.2 Focus groups and other methods Focus groups can be used alone as a freestanding research methodology, or in conjunction with other methods, especially for triangulation (Morgan, 1988) and validity checking.
Focus groups can be used at the preliminary or exploratory stages of a study (Krueger, 1988); during a study, perhaps to get feedback as part of an evaluation or a method to develop a particular programme of activities (Race et al. 1994).
When used at the start of a project, focus group can help to explore or generate hypotheses (Powell and Single, 1996) or provide insight into participants' worldview, the language they use and their values and beliefs about a particular issue or topic, which is useful in design of the study. For example, in developing a student satisfaction questionnaire, Lee Harvey and Associates (1997) used focus groups with students to determine the questions that should be in the satisfaction questionnaire. This meant the survey was relevant to and addressed the issues that actually concerned students rather than one being constructed by staff or management on the issues students should be concerned about.
Focus groups are often used alongside individual interviews (Douglas et al., 2009). Jenny Hislop and Sara Arber (2003) used focus groups alongside audio diaries in their study of sleep deprivation. Hilde Stephansen and Nick Couldry (2014) used focus groups along side participant observation and interviews in their study of community building. The National Democratic Institute for International Affairs (Thornton, 2000) in Thailand used focus groups along with surveys and self-evaluation exercises as a precursor to their anti-corruption campaign.
Elizabeth Suter (2000) goes further in suggesting that focus groups can in some circumstances substitute for observation research, especially when some subjects are hard to observe. Her own area of work was the ethnography of communication (see Section 6.7), an area of work that was fairly uncompromising in expecting studies to use participant observation as the primary method of enquiry. Suter (2000) argued that the focus group method provides another means of looking at communicative phenomena without compromising the ethnographic principle: 'Unlike other methods of data collection, focus group interviewing created conversational groups that, in turn, facilitated participant observation-like understandings' (Suter, 2000, np). She urged other ethnography of communication researchers to step outside their traditional methodological practices, when necessary, and integrate the focus group method into their research protocols. Participant observation is unfortunately limited or impossible for projects where access to observation is restricted. Suter offered examples:
in an investigation of how people think about the causes and prevention of heart attacks, sociologists Morgan and Spanish (1984; 1985) used focus groups to simulate informal discussions about heart attacks. Morgan and Spanish employed focus groups because of their limited if not complete lack of access to such conversations. It was unreasonable to expect opportunities for participant observation of these informal discussions to occur over a reasonable time span…. Similarly, in a recent study I conducted, focus groups were the primary method by which I observed married women's informal discussions about their naming choices. Given that conversations on naming choices rarely occur, I suffered from similar problems of access.
Suter summed up the key differences between participant observation and focus groups as observation in different settings, the former in a naturalistic setting and the latter in an unnatural or contrived social setting. She asked:
Is the effect of focus groups' unnatural social setting so large that it detracts from the insights gleaned, or does it create new opportunities for inquiry?
The nature of focus groups necessitates that they are created and managed, although the degree of management varies greatly depending on the moderator. In contrast, participant observation, by theory, is a technique employed to allow observation of open and unfettered discussion of the research topic.
If the heart of ethnography is thick description (Geertz, 1973), yet there are topics whose limited access to observation leads to “thin” description, then as ethnographers we need to be there in a focused way. (Suter, 2000, np)
184.108.40.206.3 Purpose of focus groups Anita Gibbs (1997) provided a sound overview of the purpose of focus groups and how they compared with interviewing and observation methods.
The main purpose of focus group research is to draw upon respondents' attitudes, feelings, beliefs, experiences and reactions in a way in which would not be feasible using other methods, for example observation, one-to-one interviewing, or questionnaire surveys. These attitudes, feelings and beliefs may be partially independent of a group or its social setting, but are more likely to be revealed via the social gathering and the interaction which being in a focus group entails. Compared to individual interviews, which aim to obtain individual attitudes, beliefs and feelings, focus groups elicit a multiplicity of views and emotional processes within a group context. The individual interview is easier for the researcher to control than a focus group in which participants may take the initiative. Compared to observation, a focus group enables the researcher to gain a larger amount of information in a shorter period of time. Observational methods tend to depend on waiting for things to happen, whereas the researcher follows an interview guide in a focus group. In this sense focus groups are not natural but organised events. (Gibbs, 1997, np)
In essence, the strengths of focus groups come from a compromise between the strengths found in other qualitative methods. Like participant observation, they allow access to a process that qualitative researchers are often centrally interested in: interaction. Like in-depth interviewing, they allow access to...the attitudes and experiences of our informants. As a compromise, focus groups are neither as strong as participant observation on the naturalistic observation of interaction, nor as strong as interviewing on the direct probing of informant knowledge, but they do a better job of combining these two goals than either of the other two techniques.
220.127.116.11.4 Interaction within focus groups The interaction between participants is an important element of focus group methodology. It is the element that distinguished focus groups from group interviews. The latter involves interviewing several people at the same time, the researcher asks questions and the group members respond (Harvey et al. 1982). In focus groups, a topic is supplied by the researcher (Morgan, 1997, p. 12) and the group interact to discuss it and the data is the result of the interaction.
The distinction is not as clear cut as this might suggest as in any group interview, the response of one person is likely to be mediated, elaborated or even disputed by another group member. The difference is not just the interaction but the way the interaction is orchestrated in focus groups.
It is therefore important, in running focus groups, to be aware of the group dynamics, the role of the moderator, as well as the synthesis that emerges (Kitzinger 1994, 1995). Agreement between group members can help to build an elaborated picture of their views; disagreement may lead to participants defending their views and provide further explanation. As Anita Gibbs (1997) explained:
the interaction between participants highlights their view of the world, the language they use about an issue and their values and beliefs about a situation. Interaction also enables participants to ask questions of each other, as well as to re-evaluate and reconsider their own understandings of their specific experiences.
For example, in her study of women's name change (following marriage) ElizabethSuter (2000, np) interviewed some women individually and then included them in focus groups. She explained how the group helped modify the outcomes of the individual interviews:
My women who were both individually interviewed and then retold their stories in the focus groups did not alter their stories between the two venues. Sometimes the group discussion brought out a new aspect of their story that had not been shared in the individual interview, but I used such instances to enhance my understanding of a particular woman's name change narrative.
One of the main problems in analysing focus group material, according to Gibbs (1997) is identifying the individual view from the group view. However, that, to some extent, is the price to be paid for obtaining several perspectives on the same topic, for which focus groups are ideal. Focus groups provide insights into people's shared understandings of everyday life and, furthermore, provide the researcher with clues as to how individuals are influenced by others in a group situation.
However, there is another issue that requires sensitive handling and that is the potential for performativity. Janet Smithson (2000) discussed the public performance element of focus groups and the need for the moderator to guide and in some cases constrain group members and to be aware of performativity when analysing and interpreting the data. The context, Smithson argued, is a vital element to be taken into account in drawing meaning from focus groups.
Gibbs (1997) also warned against assuming that the individuals in a focus group are expressing their own definitive individual view. They are speaking in a specific context, within a specific culture, and so sometimes it may be difficult for the researcher to clearly identify an individual message. For example, Sarah Neal and Sue Walters (2008, p. 291) explored the nature of rural society using focus groups. They commented on 'the fragility of the data' and suspected they were being told only particular stories by the groups.
We have argued elsewhere [Neal and Walters, 2006] that there was an element of Goffmanesque theatre—a management and public presentation of the 'best narrative' of rural and village life. This is not to argue that such best narratives do not exist, but rather to suggest that the focus group interview tends to create a forum for collective conversations which reinforce consensus rather than allow space for more diverse or contradictory truths to be expressed...
18.104.22.168.5 Organisation of a focus group session The size of a focus group for social research purposes varies widely and depends on the research enquiry, the purpose of the focus group as part of the research, the available resources and the ease with which members can be recruited to the group.
Commentators suggest that the standard for focus groups is between six and ten people (MacIntosh, 1993) although some have used up to fifteen people (Goss & Leinbach, 1996) or as few as four (Kitzinger, 1995). Elizabeth Suter (2000, np) argued that for market research the group should preferably consist of seven-to-twelve strangers. Others would suggest that around 20 is better, enabling four 'break out groups' of about five people.
In one unusual situation LeeHarvey (2000) experienced a focus group of 120 people! This was far too many for a whole group discussion but as the group met in a large hall with movable chairs, 20 subgroups of six people were convened, each with someone recording the main points. The research team consisted of five people and they circulated among four groups each, collecting the key points being raised in the subgroups. No attempt was made to have a whole group discussion although the key points were fed back to the whole group. A huge amount of information came from this group of health degree students about the nature of their education and how it could be improved.
Some studies use only one meeting with each of several focus groups (Burgess, 1996), others meet the same group several times.
The construction of focus groups is guided by the topic of research and research questions. They could be, for example, people at the same, or different, levels in the organization under study; people of the same age, class, gender; people of varying ages, classes and genders depending on the issue under study; naturally occurring groups – for example, occupational, or members of specific groups as in a rowing club. So members of the group might know each other, as in the latter, or know some or none of the group as in JanetSmithson's (2000) groups who were single-sex groups of people at similar life stages, that is, university students, in vocational training, young unemployed, in semi-skilled or professional jobs.
AnitaGibbs (1997) suggested that one should avoid groups that are too homogenous as it may not lead to diverse opinions and experiences, conversely a group that is too heterogeneous people may be loathe to contribute being wary of people from different social backgrounds, for example, or it might lead to extreme views being offered, if for example there is a wide range of political views. When undertaking employment-related focus groups Lee Harvey et al. (1997) avoided having managers and workers in the same group and, as far as possible, had groups of people at the same level within the organisation. Mary Ferguson et al. (2011) split their 78 prisoners into fairly homogenous focus groups: two for adult female prisoners; one for adult males; two for female juvenile detainees; two for male juveniles; and two groups of ex-offenders, one men only and one mixed. Similarly, Douglas et al. (2009, p. 750) had noted from their fieldwork that there was a range of important prisoner constituencies within the two prisons they were studying. The inclusion of these groupings when constructing focus groups 'ensured that women of different ages, ethnicities, nationalities and conviction status were involved, gathering a broad range of perspectives on research questions'.
Jenny Hislop and Sara Arber (2003) divided their 48 participants into six groups based on age and predicted menopausal status.
Focus group participants need to feel comfortable with each other and meeting with others whom they think of as possessing similar characteristics or levels of understanding tends to be less intimidating than meeting with those who are perceived to be different (Morgan, 1988). However, Clavering and McLaughlin (2007, p. 400) argued that, despite common advice focus groups within health research is to limit the level of variation among respondents to generate comprehensive discussion and shared knowledge, consciously building in heterogeneity has the benefit of generating interaction exploring differing professional positions and interpretations of issues under discussion. They had to plan the sessions carefully and ensure that no one professional category dominated the group. So a tricky balance is required to avoid deadening homogeneity without leading to intimidatory heterogeneity.
The logistical and practical issues of organising a focus group have to be taken into account when embarking on this method. The biggest problem initially is recruiting members to a group and getting them to be present at the specified time and location. This can be a problem even when the potential group members are located in a single organisation or location. Organising focus groups usually requires more planning than other types of interviewing and setting up appropriate venues with adequate recording facilities can be time consuming.
Recruiting participants can be time consuming, especially if the topic under consideration has no immediate benefits or attractions to participants. Researchers often offer some kind of reward to potential focus group members as inducement to participate. Suter (2000, n.p.), for example did not pay participants but relied on volunteers.
I did supply refreshments, which were sometimes augmented by the host's contributions (or in one case, the host insisted that she serve the group a light lunch before we began our discussion).
Gibbs (1997) suggested that:
It is likely that people with specific interests will have to be recruited by word of mouth (Burgess 1996), through the use of key informants, by advertising or poster campaigns (Holbrook & Jackson, 1996), or through existing social networks. Incentives, whether expenses, gift vouchers or presents, will usually need to be offered.
Mary Ferguson et al. (2011) recruited the prisoners in their focus group-based study either by placing notices in clinics and common rooms or by word of mouth. In some custodial centres, participants were asked to come at short notice. Participants in prison or juvenile detention were each paid $10 to attend (although many participants were not aware of the payment, so it is uncertain whether the payment was a motivating factor). Ex-prisoners were paid $30 to cover their travel costs and associated expenses. Refreshments (tea) were provided for all the focus groups.
Focus group sessions usually last from one to two hours. Neutral locations can be helpful for avoiding either negative or positive associations with a particular site or building (Powell & Single, 1996) or in market research, a setting that disguises the product (Suter, 2000, np). Focus group meetings can thus be held in a variety of places, for example, people's homes, in rented facilities, or where the participants hold their regular meetings if they are a pre-existing group.
Gamson's (1992) study of how people refer to mass media in discussions of local and national politics used group discussions of four to six people, who knew each other, and had minimum input from the facilitator. Gamson called these 'peer group conversations' as they differed from the standard notion of focus groups. In similar vein, Press and Cole's (1999) study of abortion used groups of two to five people, which met in people's homes. These they called 'Ethnographic focus groups' to reflect what they considered the naturalistic approach of their method. Suter (2000) similarly deviated from tradition in her study of 11 small groups of friends ranging from two to five people that met at members' homes. Because of this and the way she began groups, which involved the host sharing her story that she'd already told Suter in an interview and others then telling theirs that, Suter called her groups 'narrative focus groups', the design being to make her own role as unobtrusive as possible. (This, in effect, is another version of the debate about the impact of the presence of the participant observer on the subject of the observation (see Section 22.214.171.124)).
There is also the possibility to run virtual focus groups through on-line mechanisms, such as webinar technology or time-limited interactive blogs especially set up for a specific group of people (Stewart and Williams, 2005). The advantage of this kind of approach is that the discussion is easily recorded but the interaction is more 'forced' and it may be harder for the moderator to keep the discussion on track. More importantly, remote engagement makes it more difficult to ensure verisimilitude and there may be more scope for 'acting' or fantasising.
126.96.36.199.6 Focus goup moderator Typically the researcher moderates, or runs, the discussion. The moderator needs to provide clear explanations of the purpose of the group. The moderator uses a question or series of questions to get the group discussion started. In some cases a different stimulus, such as a video, film, photograph, vignette or game might be used to set the scene. The moderator then guides the discussion being more or less intrusive depending on the purpose of the focus group. Moderator need good interpersonal and leadership skills, being good listeners and non-judgmental, helping the group members feel at ease, facilitating interaction and keeping the group on topic. As Anita Gibbs (1997) noted:
During the meeting moderators will need to promote debate, perhaps by asking open questions. They may also need to challenge participants, especially to draw out people's differences, and tease out a diverse range of meanings on the topic under discussion. Sometimes moderators will need to probe for details, or move things forward when the conversation is drifting or has reached a minor conclusion. Moderators also have to keep the session focused and so sometimes they may deliberately have to steer the conversation back on course. Moderators also have to ensure everyone participates and gets a chance to speak. At the same time moderators are encouraged not to show too much approval (Krueger, 1988), so as to avoid favouring particular participants. They must avoid giving personal opinions so as not to influence participants towards any particular position or opinion.
It is recommended that there are at least two people on the research team, one to act as moderator and the other to check the audio or video recording devices and make notes on the interactions, identifying speakers as an aid to transcription and recognition of the participant in the recording.
In their study of prisoners, Mary Ferguson et al. (2011, p. 24) explained that:
All focus groups were facilitated by two members of the research team who used a semi-structured tool to facilitate discussion. The topic areas in the tool included: health issues of concern, health information, seeking health advice, health services in custody and health services in the community. A full explanation of the purpose and ground rules regarding confidentiality was given at the start of each focus group. Participants provided verbal consent to the discussion being audio-recorded, and were assured that all material would be de-identified.
Gibbs argued that there needs to be consistency across groups, so there is a need for careful planning of roles and responsibilities.
JanetBillson (2006, pp. 3–4) saw a rather more intrusive role for moderators than some other commentators:
A specially trained moderator facilitates the discussion through a process of “guided interaction” within a controlled environment. The moderator or other participants can pursue ideas generated by the group. The moderator can draw out motivations, feelings, and values behind verbalizations through skilful probing and restating responses. Participants stimulate each other in an exchange of ideas that may not emerge in individual interviews or surveys. The moderator can link ideas for further exploration. Group interaction generates insights that might not occur without the cross-fertilization of ideas that occurs in a well-moderated focus group. Focus groups afford depth and insight into the research question and help contextualise quantitative data (Krueger and Casey, 2000; Puchta and Potter 2004; Billson 2004; Billson 2006a).
188.8.131.52.7 Sensitive subjects The presumption is that focus groups, because of their interactive nature and that they are, therefore, not fully confidential or anonymous are suitable for non-sensitive, non-controversial and low-involvement topics. However, focus groups have been used to explore sensitive subjects such as sexuality (Frith, 2000), violence against women, sexual abuse (Overlien et al., 2005) HIV education (Munodawafa et al., 1995), contraception, abortion, child-rearing, health among prisoners (Douglas et al., 2009; Ferguson et al., 2011), corruption) (Jakobson, 2012) and people's fear of woodlands (Burgess, 1996).
Focus groups can actually help in the exploration of sensitive subjects because, as Hannah Frith (2000) argued such groups can provide conditions in which people feel comfortable discussing and sharing sexual experiences.Lori Peek and Alice Fothergill (2009) similarly argued that focus groups can serve a social support or empowerment function especially with marginalised, stigmatised or vulnerable participants.
184.108.40.206.8 Telephone focus groups Focus groups can be held remotly, using some form of conference call technology, be it conventional telephone or internet-based aural or video calls.
If using telephones, fixed land lines are usually more reliable than mobile phones, which risk losing power or poor reception.
Rather than get a group of people in a single physical location, telephone focus groups just require access to the appropriate technology at a given time. The contributors still need recruiting and appointments made. This is usually, usually via email (or letter) with reassurances about use of the data, confidentiality and informed consent.
Normally, a telephone focua group would be recorded and the participants need to be aware of that.
It is sensible to remind each participant the day before the session.
If the group interview is aural only, then each respondent should state their name before they speak.
In all forms, participants should take turns in speaking (not speak over each other).
The interviewer needs to commence the session with a brief outline of the purpose and then raise the first question.
There should be as much flexibility as possible to allow free discussion, albeit the interviewer needs to keep the conversation on track.
Telephone focus groups tend to run better with just four to six people. This is smaller than for a face-to-face group but seems to work well (Krueger, 2002, p. 2). Voice recognition can be achieved quite quickly with a small group and removes the need to announce names after the first couple of interventions.
Jeffery (1998) discovered that telephone focus groups are just as effective as face-to-face meetings. Telephone focus groups are advantageous in overcoming geographical distances so a widely distributed group can be arranged. As there is no need to get people together in one place telephone focus groups can potentially involve people unlikely to have the time or ability to come to a face-to-face meeting.
Telephone focus groups also offer an increased level of anonymity (if just aural), which may be appropriate when discussing sensitive topics.
Furthermore, participants do not need to dress for the occasion, they can take part in their everyday work clothes, which may help the participants be more relaxed.
It is argued that telephone focus groups seem easier to control than face-to-face ones, probably because during telephone calls people are used to talking one at a time. Participants tend to be polite and take turns.
As everyone is speaking into a telephone that is recorded, the final recorded session is usually much better quality than can be achieved in most face-to-face group sessions.
The face-to-face interviewer needs strong interpersonal and group-process skills. The 'invisible' telephone interviewer has to have extra ability in projecting friendliness, naturalness and informality without the aid of physical gestures. The telephone interviewer needs to be able to fill in gaps in the conversation. The interviewer also has to make extra efforts to make sure that everyone is heard if they want to be and that an individual's silence does not mean that their line has gone dead.
Earlier research suggested that discussion is rather more stilted in telephone focus groups.Krueger (1994)suggested that telephone groups may stifle discussion and there can be a lack of the spontaneity and creativity found in face-to-face groups. Other commentators disagree with this view and, indeed, this may no longer be the case with the advent of mobile telephones and social media.
220.127.116.11.9 Summary of focus groups Focus groups are a good way of getting feedback from a group of people about a specified area of research. They can benefit from interaction within the group and provide multiple perspectives. There is much to commend them, especially where surveys or observation studies are not feasible. They can be used as a research method in their own right or as the initial stage, helping specify the focus for a subsequent research enquiry.
Whether focus groups are appropriate depends on the research aims, the theoretical and philosophical approach and available resources.
Focus groups may discourage certain people from participating, for example those who are not very articulate or confident. On the other hand, focus groups do not discriminate against those who cannot read or write and encourages participation from those reluctant to be interviewed on their own. People with communication problems or special needs may not easily be included in focus groups.
Focus groups, because of the interactive nature of the discussion raise confidentiality and anonymity issues. Some people may, then, not trust others with sensitive or personal information. In such cases personal interviews or the use of workbooks alongside focus groups may be appropriate.
Focus groups are usually time consuming to arrange and the sessions need to be moderated well to maximise the data that derives from them. People may need some incentive to attend. However, those who participate often find the experience rewarding. The research process is collaborative and the group research can be empowering for participants, as well as both challenging and rewarding for social researchers wanting to glean different perspectives on their field of study.
The focus group method has limitations; the researcher (moderator) has less control than in a one-to-one interview situation, although this may not be a negative consequence. The moderator provides a context in which participants talk to each other, raise questions not anticipated by the researcher and express doubts and opinions.
Focus groups are often convenience samples, and there is the possibility that they serve to reinforce a particular view or conversely, people do not express their real feelings. Some people may 'perform' in a focus group setting and the moderator needs to alsoensure that all group members get heard.