176 Qualitative Psychology Worked Example: Six Stages of Foucauldian Analysts Lei us now lake a look al how the six stages of Foucauldian analysis may be applied to our interview extract. Stage t: Discursive Constructions Since the study from which the interview extract is taken was concerned with how people describe and account for the break-up of an intimate relationship (Willig and dew Valour, 1999; 2000), it makes sense to ask questions about the ways in which 'the relationship' is constructed through ; language- in the extract above, 'the relationship' is constructed as a clearly = identifiable social arrangement with a beginning and an end, which offers - security in return for investment of lime and emotion (lines 2-26). in the ; second half of the extract, 'the relationship' is also constructed as a step on ; the way to marriage (lines 30-42). Thus, the relationship is constructed in ; two different ways. On the one hand, the relationship is constructed as a ' social arrangement between two people who agree to invest resources (such as time and emotion) in order to gain mutual support and security. Such an arrangement is hard to extricate oneself from ('It's hard .. . it's just so hard', lines 10-11) because 'ties and emotional baggage' have grown overtime. On the other hand, the relationship is constructed as a testing ground for, and a step on the way to, a superior form of involvement, namely, marriage. Here, the relationship has to be 'going somewhere' for it to be worthwhile ('it had hit the brick wall and it wasn't going any further', lines 41-2), and its quality is judged in the light of its future direction ('And even though ... 1 had no intentions of getting married for another you know four five whatever amount of years it was on that basis I was using the criteria of my wanting to continue going out with him', lines 37-9). Stage 2: Discourses Let us attempt to locate these two constructions of the relationship (as 'social arrangement' and as 'a step on the way') within wider discourses surrounding intimate relationship. The construction of interpersonal relationships as mutually beneficial social arrangements resonates with economic discourse. Notions of investment of resources in return for long-term security and the expectation that social actors exchange goods and services with one another are prominent in contemporary talk about the economy. For example, the term 'partner', now widely used to refer to one's significant other, also describes those we share business interests with, lay contrast, the construction of the relationship as 'a step on the way' to marriage draws on a romantic discourse. Here, the relationship is not conceptualized as a mutually ,. beneficial arrangement but rather as a way of moving towards the ultimate Discourse analysis 177 goal; marriage. Marriage itself is not defined or explored within the text. It is interesting that there appears to be no need to account foi why the participant uses suitability for marriage as a 'foundation' (line 31], a 'basis' (line 38) and 'the criteria' (line 39) in her account. She even points out that she has no intention of actually getting married in the near future. However, marriage as a goal forms part of a romantic discourse in which 'love', 'marriage' and 'monogamy' are inextricably linked with one another. By -invoking one, we invoke them all. As a result, suitability for marriage becomes a legitimate basis for making decisions about intimate relationships even where there is no suggestion that marriage is a realistic option in the . near or medium future. ..-: /..- /(.ý, . . .. Stage 3: Action Orientation A closer examination of the discursive context within which the two different constructions of the relationship are deployed allows us to find out more about them. When are they used and what might be their function within the account? How do they position the speaker within the moral order invoked by the construction? (See also Stage 4: Positionings.) The portion of text which constructs the relationship as a 'social arrangement' is produced in response to a question about the involvement of friends in the decisionmaking process (lines 1-2, I: 'And when you made the decision urn when you were actually working towards finishing it did you talk to friends about it?'). This question, in turn, is preceded by an account of how the participant's friends had 'taken a dislike' to her ex-partner and how they had 'talked about him with disdain'. As a result, the participant pointed out, 'everyone was glad when I'd finished it with him'. The participant's use of a discursive construction of the relationship as a 'social arrangement' could be seen, within this context, as a way of emphasizing her sense of responsibility for her ex-partner's well-being. Talk about her friends' dislike of her ex-partner and their joy at seeing the relationship break up may have created the impression that he, disliked and rejected, was the victim of a callous act of abandonment on the participant's part. In order to counteract such an impression, a construction of the relationship as a 'social arrangement' draws attention to its mutually supportive nature and to the participant's awareness of the emotional significance of the break-up ('It's hard . . . it's just so hard', lines 10-11). The portion of text which constructs the relationship as a 'step on the way' is produced following the participant's account of how her ex-partner 'didn't think there was a problem that couldn't be worked out'. The use of romantic discourse at this point allows the participant to ward off the charge that she did not give her ex-partner a chance to 'work out' the problems and to save the relationship. From within a romantic discourse, no amount of work can transform 'liking' into 'love', or an 'OK-relationship' into 'the real 178 Qualitative Psychology Discourse analysis 179 thing'. The acid test of romantic love (line 30, 'would I want to marry him?') renders redundant attempts to work out problems, because, if marriage is not a goal that can he envisaged, the relationship is not worth saving (linos 41-2, 'and as far as I was concerned it had hit the brick wall and it wasn't going any further'). From within a romantic discourse, the participant cannot be blamed for not trying hard enough to make the relationship work, Stage 4: Posiiioü^gs I What are the subject positions offered by the two discursive constructions of | 'the relationship'? A construction of relationships as 'social arrangement' | positions partners as highiy dependent on each other. Involvement in such a : relationship undermines the individual's freedom and mobility; partners ;' are lied to each other through investments, history and emotions (line 31, 'there's all these you know lies and emotional baggage which . . . you're carrying'). As a result, whoever decides to withdraw from the arrangement is going to cause the other person considerable disruption, inconvenience ami probably a great deal of distress. The subject positions offered by this con- : struction are, therefore, those of responsible social actors who depend on each other for support and who are faced with the difficult task of realizing their Interests within relationships of interdependence. The romantic construction of intimate relationships as 'a step on Uic way' offers provisional subject positions to lovers. While involved in unmarried relationships, lovers are not fully committed to the relationship. Their involvement contains an opt-out clause which allows them to withdraw from the relationship without penalty. Everything that occurs between lovers within such an arrangement is permanently 'under review' and there is no guarantee that the relationship has a future. Therefore, the subject : positions offered by this construction are those of free agents who reserve the right to withdraw from the relationship at any time and without moral : sanction. \ .. . , , - : Stage S: Practice '. What are the possibilities for action mapped by the two discursive con-; structions of relationships? What can be said and done by the subjects : positioned within them? Constructions of relationships as 'social arrangements' and their subject positions of responsible social actors require those positioned within them to act responsibly and with consideration for the consequences of their actions. Being part of a mutually beneficial social arrangement means that whatever we do affects the other party within the arrangement, and that we need to take responsibility for these effects. The participant's account of how she rehearsed breaking up (lines 5-10) and how hard it was for her to 'actually say to him i don't want to go out with you anymore7 (lines ()-10) demonstrates her positioning as a responsible social actor. Taking responsibility for one's partner's well-being (line 19) and breaking up in a way that demonstrates concern for that partner's future are practices which support a construction of relationships as 'social arrangements'. By contrast, being positioned within a relationship as 'a step on the way' does not require the same preoccupation with the other's well-being. Note that the section of text which constructs the relationship as 'a step on '■ tlie way"{lines 30-42) does not contain any references to the participant's ex-partner. Instead, it talks about the nature of the relationship and the criteria by which to assess its value. The subject position of a free agent who reserves the right to withdraw from the relationship at any time and without moral sanction involves a focus upon the self and its interests. This is demonstrated in lines 30-42 (note the consistent use of the first-person singular and the references to 'foundation', 'basis' and 'criteria' for decisionmaking in this section). Stage 6: Subjectivity This stage in the analysis is, of necessity, the most speculative. This is because here we are attempting to make links between the discursive constructions used by participants and their implications for subjective experience. Since there is no necessary direct relationship between language and various mental states, we can do no more than to delineate what can be felt, thought and experienced from within various subject positions; whether or not, or to what extent, individual speakers actually do feel, think or experience in these ways on particular occasions is a different question (and one we probably cannot answer on the basis of a discourse analysis alone). It '■ could be argued that feelings of guilt and regret are available to those positioning themselves within a construction of relationships as 'social arrangements' (lines 19-21, 'You start taking responsibility for them and for how they'll cope afterwards you know maybe to the detriment to your own personal sort of well-being'), while taking up a position as free agent within a construction of relationships as 'a step oil .the way' may involve a sense of time urgency in relation to decision making {lines 33-5, 'because 1 thought 'OK we've been going out for two nearly two years if we were going out for another two years would 1 want to marry him and the answer was no'). Key Differences between Discursive Psychology and Foucauldian Discourse Analysis Both versions of the discourse analytic method share a concern with the role of language in the construction of social reality. However, as I hope has 180 Qualitative Psychology become clear, iliure are also important differences between the two approaches. To conclude this „chapter, 1 w^m to _niake...a direct comparison, between the two versions of discourse analysis.and the analytic insights each one of them can generate. Key differences between the two versions are presented under three headings: 'Research Questions', 'Agency' and 'Experience' (see Box 8,4 for a summary). Research Questions Discursive psychology and Foucauidian discourse analysis are designed io answer different sorts of research questions. Discursive psychology projects typically ask, 'How do participants use language in order to manage stake in social interactions?', while Foucauldian discourse analysis answers the question 'What characterizes the discursive worids people inhabit and what are their implications for possible ways of being?' Our discursive analysis of ílu-interview extract was designed to answer questions about what the participant was doing with her talk, it allowed us to observe that the extract served as a warrant for the participant's decision to terminate her relationship with her partner. By contrast, our Foucauidian analysis was concerned with the nature of the discursive constructions used by the participant and their implications for her experience of the relationship break-up. We were able to identify both economic and romantic discourses in her account, each of which offered different subject positions and different opportunities for practice and subjectivity. Discursive psychology and Foucauidian discourse analysis emphasize different aspects of human agency. Even though discursive psychology is concerned with language and its performative aspects, rather than with speaking subjects and their intentions, its focus on action orientation presupposes a conceptualization of the speaker as an active agent who uses discursive strategies in order to manage a stake in social interactions. In line with this, our discursive analysis focused upon the participant's use of discourse in the pursuit of an interpersonal objective which was to justify her decision to leave her partner within the context of a research interview. By contrast, Foucauidian discourse analysis draws attention to the power ol ^discourse to construct its objects, including the human subject itself. The availability of subject positions constrains what can be said, done and feit by individuals. Reflecting this concern, our Foucauidian analysis was interested in the discursive resources which were available to the participant and how their availability may have shaped her experience of the break-up. Discourse analysis Experience Discursive psychology questions the value of the category 'experience' itself. Instead, it conceptualizes it (along with others such as 'subjectivity' and 'identity') as a discursive move whereby speakers may refer :o their 'experiences' in order to validate their claims (as in T know this is hard because I've been there!'). Here, 'experience' is a discursive construction, to be deployed as and when required. Anything more than this is seen to constitute a return to cognitivism and this would, therefore, not be compatible with discursive psychology. By contrast, Foucauidian discourse analysis does attempt to theorize 'experience' (and 'subjectivity'). According to this approach, discursive constructions and practices are implicated in the ways in which we experience ourseives (such as 'sick' or 'healthy', 'normal' or 'abnormai', 'disabled' or 'able-bodied', and so on). As 3 result, an exploration of the availability of subject positions in discourse has implications for the possibilities of selfhood and subjective experience. This difference was reflected in our worked example. Our discursive analysis was concerned with what the respondent was doing with her talk, whereas our Foucauidian analysis was more interested in the implications of her use of discourse for her experience of the break-up. Conclusion Discourse analysis is a relatively recent arrival in psychology. However, despite its short history, it has already generated a large body of literature. As researchers use discourse analytic approaches within different contexts, they encounter new challenges which iead them to develop new ways of applying a discursive perspective. For example, early work in discourse analysis tended to concern itself with social psychological topics such as prejudice. More recently, health psychologists have started to use the method, leading to the formulation of a material-discursive approach (e.g., Yardley, 1997), while others have attempted to find ways in which discourse analysis could inform social and psychological interventions (e.g., Willig, 1999). This demonstrates that discourse analysis is not a method of data analysis in any simple sense. Rather, it provides us with a way of thinking about the role of discourse in the construction of social and psychological realities, and this, in turn, can help us approach research questions in new and productive ways. The two versions of the discourse analytic method introduced in this chapter are ways of approaching texts rather than recipes for producing 'correct analyses'. The choice of approach should be determined by the research question we wish to address; in some cases, this means that a combination of the two approaches is called for. The most ambitious discourse analytic studies may wish to pay ; attention to both the situated and shifting deployment of discursive con- 182 Qualitative Psychology Box 8.4 Key differences between discursive psychology (DP) and Foucauldian discourse analysis (FDA) Research Questions • DP asks, 'How do participants use language in order to manage a state in sociaf interactions?' • FDA asks, 'What characterizes the discursive worlds parlicipants inhabit and what are their implications for possible ways of being?' Agency Discursive Psychology • The speaker is an active agent. • The speaker uses discourse. • Discourse is a tool. Foucauldian Discourse Analysis • The speaker is positioned by/in discourse. • Discourse makes available meanings. • Discourse constructs its subjects. Experience Discursive Psychology • DP questions the value of the category 'experience'. • DP conceptualizes invocations of 'experience' as a discursive move. Foucauldian Discourse Analysis • FDA attempts to theorize experience. • Discourse is implicated in experience. • Discourse makes available ways of being. Discourse analysis 183 structions, as well as to the wider social and institutional frameworks within which they arc produced and which shape their production, in this case, both discursive resources and discourse practices need to be explored in detail so that we can understand how speakers construct and negotiate meaning (discourse practices), as weil as why they may draw on certain repertoires rather I han others (discursive resources) (Wethereil, 1998). In any event, our choice of analytic method(s) should always emerge from careful consideration of our research qucslton{s). Note Material přeseti led in this chapter is based on Chapters 6 and 7 in Willi); (2001). • further Reading Wilíig, C. (2001) Introducing Qualitative Research in Psychology: Adventures in Theory and Method. Buckingham: Open University Press. Chapters 6 and 7 of this book provide a more detailed discussion of the two versions of discourse analysis. Wethereil, M. (1998) 'Positioning and interpretative repertoires: conversation analysis and post-structuralism in dialogue', Discourse and Society, 9; 387-413. In this paper, Wethereli argues in support of an integration of the two versions of discourse analysis. Willig, C. (1998) 'Constructions of sexual activity and their implications for sexual practice: iessons for sex education'. Journal of Health Psychology, 3: 383-92. This paper provides an illustration of the application of the řoucauidian version of discourse analysis. Wiggins, S., Potter, j. and Wildsmith, A. (2001) 'Eating your words: discursive psychology and the reconstruction of eating practices', Journal of Health Psychology, 6: 5-15. This paper provides an illustration of the application of the discursive psychology version of discourse analysis. Chapter 9 Focus groups Sue Wilkinson Focus groups are now a popular and widely used method in qualitative research across the social sciences. Although the method dates back some 76 years, it was relatively uncommon until a major resurgence of interest in the late 1970s. It has become popular in psychology only within the last decade, as qualitative research has burgeoned and become more generally accepted within the predominantly quantitative discipline. The early use of focus groups can be traced back to the 1920s, when the psychologists Emory Bogardus and Walter Thurstone used them to develop survey instruments - although their 'invention' is more often credited to sociologist Robert Merton and his colleagues Patricia Kendall and Marjork: Hske in the 1940s. Merton's research team developed 'focused group-interviews' to elicit information from audiences about their responses to radio programmes. Since then, they have been also known (variously) as 'group interviews' or 'focus group interviews', but the term 'focus groups' is the most commonly used - and serves (if nothing else) to distinguish tin-approach from more psychodynamically oriented forms of group work. Prior to the late 1970s, the main use of focus groups was as a market research too!, and most published studies were in the held of business and marketing - this is still an active area of focus group research today (Greenbaum, 1998). In the 1980s, health researchers pioneered the use ol focus groups in social action research, particularly in the fields of family planning and preventive health education; the method was then widely used to study sexual attitudes and behaviours, particularly in relation to HIV/ AIDS; and it continues to be used extensively today in the areas of health education and health promotion (liasch, 3 987), as well as in health research more generally (Carey, 1995; Wilkinson, 1998a). Jn the 1990s, the growing popularity of focus group research created a substantial literature on the method across a much wider range of disciplines, including education, communication and media studies, feminist research, sociology and psychology (see Morgan, 1996; Wilkinson, 1998b, for reviews). Focus group methodology is, at first sight, deceptively simple. It is a way of collecting qualitative data, and this - essentially - involves engaging 184 Focus gfoups a small number ol people in an informal group discussion (or discussions), 'focused' on a particular topic or set of issues. Focus group projects in which 1 have been involved, for example, include young women exploring how to negotiate sexual refusals; young men talking about body modification practices, such as hair removal, piercing and tattooing; nurses evaluating different types of ward management; lesbian parents discussing their children being bullied at school; women comparing their experiences of vaginal examinations and cervical smears; and partners of women with breast cancer sharing information about 'coping' with life on a day-to-day basis. There is a common misconception that people will be inhibited in revealing intimate details in the context of a group discussion - in fact, focus groups are well suited to exploring 'sensitive' topics, and the group context may actually facilitate personal disclosures (Farquhar, 1999; Frith, 2000). The informal group discussion is usually based around a series of questions (the focus group 'schedule'), and the researcher generally acts as a 'moderator' for the group:. posing the questions, keeping the discussion llowing, and encouraging people to participate fully. Although focus groups are sometimes referred to as 'group interviews', the moderator does not ask questions of each focus group participant in turn - but, rather, facilitates group discussion, actively encouraging.group members to interact with each other. This interaction between research participants is a key feature of focus group research ~ and the one which most clearly distinguishes it from one-to-one interviews (Morgan, 1997). Compared with interviews, focus groups are much more 'naturalistic' (that is, closer to everyday conversation), in that they typically include a range of communicative processes - such as storytelling, joking, arguing, boasting, teasing, persuasion, challenge and disagreement. The dynamic quality of group interaction, as participants discuss, debate and (sometimes) disagree about key issues, is generally a striking feature of focus groups. Typically, the focus group discussion is audiotaped and the data are transcribed and then analysed by conventional techniques for qualitative data - most commonly, content or thematic analysis. Focus groups are distinctive, then, primarily for the method and type of data collection (that is, informal group discussion), rather than for any particular method of data analysis. One possible reason for the contemporary popularity of focus group research is the flexibility of the method. Focus groups can be used as a standalone qualitative method, or combined with quantitative techniques as part of a multimethod project. They can be used within the psychology laboratory or out in the field; to study the social world or to attempt to change it -that is, in action research projects (see Wilkinson, 1999, for a review; also Chapter 10, this volume). At almost every stage of a focus group project, there are methodological choices to be made. A good way to get a sense of this flexibility and variety is to look through one of the recent edited 186 Qualitative Psychology collections of focus group research, such as Barbour and Kitzinger (199'') and Morgan (1993). A focus group project can involve a single group of participant meeting on a single occasion, or il can involve many groups, with single or ' repeated meetings. It can involve as few as two, or as many as a dozen or i so participants {the norm is between four and eight). These participants may he pre-existing groups of people (such as members of families, club'; or work teams), or they may be brought together specifically for the research, as representative of a particular population, or simply on the basis of shared characteristics or experiences (for example, middle-aged men, , sales assistants, and sufferers from premenstrual tension). In addition to (or ■ instead of) a set of questions, the moderator may present group members with particular stimulus materials {such as video clips and advertisements); and, in addition to {or instead of) discussing particular questions, they may be asked to engage in a specified activity (such as a card-sorting task or a ■ rating exercise). Kitzinger (1990) provides examples of a range of such activities in the context of researching AIDS media messages. The moderator may be relatively directive or relatively non-directive. Proceedings may be audiotaped or videotaped (the former is more common in social science» research). Data transcription may be more or less detailed, ranging from simple orthographic transcription, which preserves just the words spoken, to the more complex form of transcription favoured by conversation analysts (sec Chapter 7), which also preserves a range of linguistic and paralinguistie features, such as false starts, self-corrections, overlapping speech, pauses, volume and intonation. Data analysis may be by hand (as in cutting and ! pasting sections of transcript) or computer-assisted (using programs such as • NUD.IST or THE ETHNOGRAPH). A wide variety of different types of data analysis may be undertaken - including content, thematic, phenomeno- i logical, narrative, biographical, discursive or conversation analysis (some of ■■ which are discussed in more detail in other chapters of this book). The type of analysis used depends upon the theoretical framework of the researcher rather than upon any particular featurc(s) of focus group data. One particular strength of focus group research is that it is not tied to a specific theoretical framework: the method can be used either within an 'essentialist' or within a 'social constructionist' framework. Focus group research conducted within an essentialist framework, like most psychological research, rests on the assumption that individuals have their own personal ideas, opinions and understandings, and that the task of the researcher is to access or elicit these 'cognitions'. Within this framework, the particular advantage of focus groups is the more comprehensive eiicitation of individuals' ideas, opinions and understandings than is possible in one-to-one interviews (more comprehensive in the sense that co-participants are likely to trigger memories, stimulate debate, facilitate disclosure and generally Focus groups 187 encourage the production of elaborated accounts). Focus group research conducted within a social constructionist framework does not assume preexisting cognitions located inside people's heads, but, rather, presupposes that sense-making is produced collaboratively, in the course of social interactions between people. Within this framework, the particular advantage of focus groups is flic opportunity they offer for the researcher lo observe how people engage in the process of collaborative sense-making: how views are constructed, expressed, defended and (sometimes) modified within the context of discussion and debate with others. The theoretical framework of the research will influence the kind of data analysis undertaken - essentialist research is likely to utilize content or thematic analysis, while social constructionist research is more likely to use narrative/biographical or discursive/conversation analysis. Given this breadth and flexibility of use, focus groups are obviously a multipurpose method. However, they are not, as is sometimes assumed, 'a method for all'seasons' ~ like any other method, they have particular advantages and disadvantages, and are demonstrably more suited to some kinds of research questions than others. Focus groups are a good choice of method when the purpose of the research is to elicit people's own understandings, opinions or views (note that this is an essentialist research question); or when it seeks to explore how these are advanced, elaborated and negotiated in a social context (note that this is a social constructionist research question). They are less appropriate if the purpose of the research is to categorize or compare types of individuals and the views they hold, or to measure attitudes, opinions or beliefs (although they are sometimes used in this way). Focus group data are voluminous, relatively unstructured, and do not readily lend themselves to summary analysis. While such data can be subjected to some limited quantification (as in some forms of content analysis - see below), they are best reported in ways which preserve (at least1, some of) the participants' own words, for example, by using illustrative i quotations, ideally, too, there should also be some analysis of group interactions (although, sadly, this is all too rare in the published literature). Focus groups arc unlikely to be the method of choice when statistical data and gencralizable findings are required: samples are usually small and unrepresentative, and it is difficult to make a good theoretical case for aggregating data across a number of diverse groups, or for making direct comparisons between groups (although, again, this is sometimes done). There are also practical advantages and disadvantages to the use of focus groups. They have been seen as a way of collecting a large volume of data relatively quickly and cheapiy. However, it can be difficult to recruit and bring together appropriate participants; moderating a group effectively is a skilled technique, which (ideally) requires training and practice; and data transcription and analysis (of whatever kind) is an extremely painstaking and time-consuming process, which requires a range of data-handling and 188 Qualitative Psychology Focus groups 189 interpretative skills. The following section of this chapter offers a muiv detailed practical guide to what is involved in doing focus group research. Practica! Guide The focus group literature includes a substantial number of 'handbooks', which offer a wealth of general information and advice about the process of doing focus group research, as well as a consideration of issues specific to particular types of focus group. The most useful of these guides for the psychologist are by Krueger (1994); Morgan (1997); Stewart and Shamdasani (1990) and Vaughn et al. (1996); the most comprehensive is by Morgan and Krueger (1998). Here, I draw both on the advice offered by these handbooks and on my own experience of focus group research, to review the key stages of a focus group project and to suggest the key practical considerations ai each stage. i also illustrate each stage of a focus group project with examples from my current research on women's experiences of breast cancer (Wilkinson 1998a, 1998b, 2000a, 2000b). In this project, a total of 77 women took pari in 13 focus groups, each lasting 1-3 hours. Participants were recruited through a symptomatic breast ciinic at a general hospital in the north of England. Most were working-class, middle-aged or older, and within live years of diagnosis. The focus groups were held in a university setting, and each woman attended only one group on a single occasion. Discussion ranged across the women's feelings on diagnosis, their relationships, their experiences of treatment, and the changes that cancer had created in their lives. Data were audiotaped and transcribed orthographically in the first instance; they are currently being retranscribed in more detail and analysed by a variety of techniques (see Wilkinson |2000bj for a comparison of three : methods of analysis of one of the focus groups). For any focus group to provide the best possible data (and to be a rewarding experience for the participants, an outcome which wiil also lead to better data), two things - at least - are necessary: an effective moderator and a well-prepared session. Ideally, the moderator should have some basic interviewing skills, some knowledge of group dynamics, and some experience of running group discussions. Although some of the skills involved in moderating a focus group are similar to those involved in one-to-one interviews (for example, establishing rapport, effective use of prompts and probes, and sensitivity to non-verbal cues), the number of research participants involved in a focus group requires more in terms of active 'people management'. The shy participant must be encouraged to speak, the talkative one discouraged at times, and instances of discomfort and/or disagreement must be handled with care. The handbooks provide substantial detail on the principles of 'people management', but are no substitute for the experience of moderating a focus group in practice. The most common mistakes of novice (and/or nervous) moderators are; failure to listen - and so follow up appropriately: inability to tolerate silence; talking too much; and sequential questioning. You should not embark on a focus group project without some kind of practice run - or, preferably, a full-scale pilot study. Proper preparation for, and efficient planning of, the focus group session itself is just as essential as moderator skills for obtaining high-quality data. A well-run focus group session might look effortless, but it almost certainly is not: a surprising amount of preparatory work is needed before, during and after the session itself. After you have determined that focus groups are an appropriate way to address your research question, here are some of the main practical considerations in setting up an effective focus group project. ĹJíísiijii issues ■ Hrst, you will need to decide on the broad parameters of your project - that is, the overall timescale; how many focus groups you will run; what kind of focus groups they will be; the number and type of participants you will have (and how you will recruit them); and how you will record, transcribe and analyse your data. These parameters need to be set before you can address the more detailed practical issues below. In almost all cases, the design of the research is likely to be a compromise between what would be ideal, and what is actually feasible, given the practical constraints of time, resources and your own expertise and energy. Ethical Issues Focus group research, like any other psychological research, must be conducted in accordance with the ethicai guidelines of the relevant professional body (that is, in the UK, the British Psychological Society). Broadly speaking, you need to obtain the necessary permissions and ethical clearances from the institution where you are based (such as a university or human subjects ethical committee); and from the institution where you wiil collect your data (such as a Health Authority or a National Health Service Trust ethical committee), as well as any key 'gatekeepers' within it (such as consultant or service manager). You must also obtain your participants' informed consent to take part; you are responsible for protecting their confidentiality, and you should take all reasonable steps to ensure that they will not be subjected to any stress or anxiety beyond and above what they might reasonably experience in their everyday lives. Confidentiality is a particular issue within focus groups, because of the number of participants, and 'ground rules' must be set to ensure that personal details and potentially sensitive material are not discussed outside the 190 Qualitative Psychology context of tlie group (that is, participants should be requested to rcspec! ami preserve the confidentiality of others). There are also some ethical issnci. specific to the interactional nature of focus group research, for example, very occasionally a participant may be visibly worried or distressed hv the experiences or opinions being aired, an argument may 'turn nasty', <>i several focus group members may collude to silence or intimidate a pant cular individual, ft is important to handle such a situation immediately, within the group (this may include, in the last resort, terminating tin-session); it may also be necessary to address it further with the individuate») involved once the group has finished. In practice, though, focus group research is usually an interesting, and often enjoyable, experience for all concerned, and such 'difficult situations' rarely occur. Finally, as with any research, it is a good idea to have contact details available for relevant counselling services, helplines, self-help groups and other sources of information, in case they are needed or requested following the group. Preparing Materials You will need (at least) a focus group schedule, perhaps also written 01 í pictorial materials. In devising a schedule, make sure that it is likely to ; engage the participants, that it uses appropriate vocabulary, that the ques [ tions flow logically, that it provides the opportunity for a variety of view-j points to be expressed, and that it allows participants to raise points which 1 may not have occurred to the researcher. Box 9.1 shows the schedule used in my breast cancer project. Try out all the materials you intend to use - to ensure they are intelligible, legible, visible and the right length. If you are intending to use slides or video clips, make sure that the appropriate projectors are readily available, and that you know how to operate them. Have back-ups available in case of equipment failure. Write out your introduction to the session (include a recap on the project, the procedure to be followed and the 'ground rules' for the focus group) and your closing comments (include a summary of the session, any necessary debriefing and a reiteration of thanks); see 3iso the procedural points covered in 'the session itself below. Recruit! nt; Participants This is much harder than the novice focus group researcher ever imagines. Make sure that potential participants know what is involved in the focus group procedure - this is part of giving informed consent. Consider whether you will pay them (or offer other incentives - for example, simple refreshments are almost always appropriate) and/or reimburse travel expenses. Always ovcrrecruit by about 50 per cent (that is, recruit nine participants for Focus groups 191 Rox 9.1 Focus group schedule Women's Experience of Breast Cancer Introduction (recap on purpose o! project, procedure, ground rules) Questions (used in all groups) 1. How did you feel when you lirst became aware of a breast problem? 2. How did you feet when you were first told it was breast cancer? 3. How did people around you react to knowing you had breast cancer? - Partner/family/friends/others 4. What kind of support did you need? - When you were first aware of a problem? - When you knew tor sure it was cancer? 5. What kind of support did your parlner/tamily/others close to you need? - When you were first aware of a problem? - When you knew for sure it was cancer? 6. What do you think caused your breast cancer? 7. What kind of effect has having breast cancer had on your life - including your genera! outlook on life? - On you personally? - On those around you? Supplementary questions (used in some groups, when lime) 8. What is the worst thing about having breast cancer? 9. Has anything good come out of having breast cancer? - What? 10. Have you been concerned about your appearance? - In what way? - Those around you? In what way? 11. Is there anything else you would like to say about your experience o! breast cancer? - Or about this research project? Conclusion (summary, thanks and debriefing) 192 Qualitative Psychoíogy .1 six-person group) - however much enthusiasm/commitment participants express, some of them always fail to turn up on the day, for one reason or another. Make sure they have clear directions for finding the venue, and (particularly if you recruit some time in advance of the session), issue several reminders, including - most crucially - a telephone call the day before the focus group meets. Choosing the Venue Sometimes - particularly in action research projects - there is no choice of venue: you have to conduct the focus group on the group's own 'territory' (that is, wherever the participants usually meet, or wherever they are prepared to meet you), which may not be an ideal research environment. Where there is a choice, however, the main consideration is balancing participant comfort and a good recording environment. A few universities now have purpose-built 'focus group suites' (more often in the business school than the psychology department), and most psychology departments have a laboratory with a one-way mirror - this might be worth considering, particularly if observation/video recording is part of the project. Mosi important is a relatively comfortable, quiet room where you will not he disturbed or under time pressure to finish. Participants should be seated in a circle - either in easy chairs or around a table (your choice may depend on what participants will be asked to do, but note the different 'feel' of these two options). Easy access to lavatories and to a telephone is essential. PfBparinc; for the Session There are two aspects to this: thinking through the logistics of the day itself and preparing supplementary materiais. Ii is idea! to have an assistant, especially for larger focus groups. Whether or not this is possible, think through how you will handle arrivals and departures {including late arrivals £nd eariy departures), refreshments, dealing with unforeseen queries or problems, and taking notes and/or operating the recording equipment while moderating the group. Remember that Murphy's law {'if anything can go wrong, it will') holds as much for focus groups as other types of research ■■ but seems to apply particularly to recording equipment! This should be checked and double-checked before every group. White highly specialized recording equipment is unnecessary, it is essential to use an omnidirectional, flat microphone in order to produce a recording clear enough for transcription. These can be purchased relatively inexpensively at large high-street electrical retailers. To minimize the risk of recording failure, it is also desirable to use two sets of recording equipment, if possible (this also reduces the number of tape copies needed) - this is much easier to manage if you have an assistant. Focus groups 193 In terms o! supplementary materials, you will need some or all of the following: » refreshments: water at least, preferably tea/coffee and biscuits (not alcohol); depending on time of day and length of session, possibly simple food (such as sandwiches and pizza) ~ but nothing crunchy (this obscures the recording) e writing materials (paper and pens) - for yourself and the participants o informed consent forms; expenses claim forms • a box of paper tissues o name badges or cards (and marker pens to complete them) • recording equipment, including spare tapes and batteries. Set up the room well in advance, if possible, and check the recording equipment (again) just before using it. The Session itself You need to allow 1-3 hours (depending on the topics/activities to be included and the availability/commitment of the participants). The beginning and end of the focus group session entail specific practica! considerations. The following activities are needed at the beginning of the session (not necessarily in this exact order): • offering thanks, a welcome and introductions e attending to participants' comfort {refreshments, toilets, any special needs) o signing consent forms (if not done at recruitment), including permission to record and an explanation of what will be done with the data • reiterating issues of anonymity/confidentiality • completing name badges • recapping purpose of study • outlining procedure (including confirming finishing time) »i. 194 Qualitative Psychology e spuing ground rules for running t ho group » providing an opportunity to ask questions. You then move into the discussion itself. You should aim fo create ;m atmosphere in which participants can relax, talk freely and enjoy them selves. Although it may take a while to 'warm up', once it gets going, a good focus group discussion will appear almost to run itself. The discussion wit! 'flow' weii - and it will seem to move seamlessly through the schedule ■■ sometimes even without the moderator needing to ask the questions. Such apparent 'effortlessness' rests substantially upon good preparation and effective moderating skills fas well as a measure of good luck). Although your main energies should be directed towards effective moderation of the group discussion, it is also desirabie to keep notes of the main discussion points, and of any events which may not be captured on audiotape - for example, the occasion when one of my focus group participants reached inside her bra, pulled out her prosthesis {artificial breast) and passed it around the table! An assistant will be able to take more comprehensive notes, which could include a systematic list of the sequence of speakers (this helps in transcription, especially with larger groups). It is also worth noting that a good focus group often overruns: always aiiow participants to leave ;ti the agreed time, even if you have not finished. The following activities are needed at the end of the focus group (again not necessarily in this exact order): • reiterating thanks • reiterating confidentiality • giving a further opportunity for questions • providing further information, or possible sources of information (as appropriate) • debriefing (as appropriate) - including on an individual basis as necessary • checking that participants have had a good experience (possibly formal evaluation) • completing expenses claim forms (and making payment arrangements) • offering appropriate farewells and/or information about any follow-ups. Focus groups 1 95 Box 9.2 Sample transcription key Transcription Conventions Used for Data Extracts in this Chapter » underlining - emphasis • hyphen at end of word - word cut oif abruptly • ellipsis (...)- speaker trails off • round brackets - used when transcriber is uncertain what was said, but is able to make a reasonable guess - for example, (about) • square brackets - enclose comments made by transcriber. Such comments include inability to make out what was said [indistnet], and sounds that are difficult fo franscribe - such as [ten], [stutters], as well as interactional features of noie - such as [laughs], [pause], [cuts in], [turns to Edith). Data Management and Transcription The next step is to make back-up copies of all notes and tapes (which should be clearly labelled with the date, time, length and nature of the session). Keep them in a separate place from the originals. If you are transcribing your own data (as is usualiy the case), try to do this as soon as possible after the session, while it is still fresh in your mind. Specialized transcribing equipment is not essential - but it is strongly recommended for more advanced/ extensive transcription work (the job of transcription is considerably facilitated by the use of a dedicated transcribing machine, and a much better transcript usually results). Transcription is really the first stage of data analysis, and a careful, detailed transcription will facilitate the next steps (although the level of detail preserved in the transcription will depend on your research question and the type of data analysis you plan to use - see earlier discussion). Note that whenever you present extracts from your data, you should append a transcription key listing the precise transcription conventions you have used. Box 9.2 shows a typical (simple) transcription key covering the data extracts presented in this chapter (a more elaborate transcription key, of the type used in conversation analytic studies, can be found in Atkinson and Heritage (1984); see also Chapter 7). Whatever type of transcription you undertake, the transcription process is likely to take much longer than you might expect. A skilled transcriber