Family Practice Vol. 17, No. 1, 76-82
© Oxford University Press 2000
Methodologies |
Acquiring qualitative skills for primary care research. Review and reflections on a three-stage workshop. Part 1: using interviews to generate data
Department of General Practice, University of Glasgow, Glasgow, UK.
a Research Facilitator for the Wolds Primary Care Research Network (WoReN).
Members of WoReN: Maggie Booth (Researcher); Sandra Burley (Lecturer in Practice Nursing); Brian Cook (General Practitioner); Margaret Crawley (Audit Co-ordinator); Chris Elwood (Researcher); Hayley Gibson (Audit Information Assistant, Cancer Services); Chris Jary (General Practitioner); Heather Middleton (Practice Nurse); Caroline Plews (Research Associate); Gill Rowland (Community Child Health Doctor); Bryony Simpson (Speech and Language Therapist); Shirley Stephenson (Research Health Visitor); and Jenny Wright (Assistant Project Manager, Cancer Services).
Rosaline S Barbour, Department of General Practice, University of Glasgow, 4 Lancaster Crescent, Glasgow G12 0RR, UK.
Barbour RS, Featherstone VA and Members of WoReN. Acquiring qualitative skills for primary care research. Review and reflections on a three-stage workshop. Part 1: using interviews to generate data. Family Practice 2000; 17: 76–82.
Received 27 April 1999; Revised 9 August 1999; Accepted 6 September 1999.
Abstract
This paper reflects on one Primary Care Research Network's (WoReN's) experience of running a workshop on generating interview data, provided as the first of a three-part workshop concerned with acquiring qualitative interviewing skills. It discusses the aims and limitations of the short workshop format in meeting the needs of practitioners embarking on qualitative research, drawing upon and reviewing the relevant research methods literature, and makes suggestions with regard to designing and running research methods workshops within primary care.
Keywords. Methodology, qualitative research, research training..
Introduction
This paper reflects on our experience of running a set of three half-day workshops on semi-structured interviewing for members of the Wolds Primary Care Research Network (WoReN) and represents a collaborative writing venture. It is concerned with the first of these workshops, which concentrated on developing an interview schedule and generating data. The accompanying paper discusses the subsequent workshop on data analysis.
WoReN is one of the newest networks to be set up to facilitate good quality primary care research within the NHS and covers East Yorkshire (north of the River Humber), and North & North East Lincolnshire (south of the River Humber). Reflecting the network's philosophy of a grass roots approach involving all members of the primary care team, the self-selected workshop participants (12 people) represented a range of professional groups (see participant list). Most had experience of ‘interviewing’ within the context of healthcare encounters, and five had already carried out some research interviews. Whilst all of the participants recognized the potential of qualitative methods to elicit ‘in-depth’ data, they were not fully aware of the implications for the research process either in terms of the researcher's role in generating such data or the challenges presented by the amount and range of data involved. (This is discussed in the accompanying paper.) There was, however, a general consensus that they needed to know ‘how to do it properly’. Thus, their expectations of the workshops were probably, at the same time, both too high and too low.
The collaboration, which is the subject of this paper, involved a medical sociologist (RS Barbour) in designing and running the workshops with input from V Featherstone (whose background is in psychology, sociology and education) and, of course, the workshop participants themselves.
Since the network was established in 1996, its educational programme has spanned both qualitative and quantitative paradigms, with workshops on questionnaire and study design, statistics, writing for journals and medical informatics, as well as fora on such diverse topics as research ethics and the role of anthropology in medical research. The network has sought to embrace other disciplines, accepting that clinical research needs to be open to all types of knowledge.1
Aims and scope of the workshops
Although, within the UK as a whole, there are now some excellent courses being offered in relation to developing qualitative research skills (e.g. the Royal College of General Practitioners' Master Classes and workshops run by Social and Community Planning Research, in London), there is an urgent need for local provision for busy practitioners. It may be difficult for them to find the time or to obtain cover within their practices to enable them to attend an intensive course, which is also likely to involve considerable travel.
There are, of course, many other smaller scale workshops and training courses which have sprung up in response to the demand. Whilst many of these may equip people with data collection skills, this is often done at the expense of addressing other aspects of the research process—most notably data analysis. This joint venture between WoReN and the Department of Public Health and Primary Care at the University of Hull offered the opportunity to develop an alternative format: to provide ‘hands-on’ experience of formulating interview schedules, eliciting data, analysing transcripts, aggregating data and considering issues around the presentation of qualitative findings. By allowing time between the workshops for the facilitators to carry out further analysis of the data generated and by linking the workshops to an earlier presentation on the epistemological underpinnings of qualitative methods, we hoped to move beyond the ‘cookbook’2 approach which has been criticized for merely providing recipes for data collection and analysis, devoid of theoretical considerations.
The case for the relevance of qualitative methods for primary care research has already been well made.3–7 Nevertheless, there can be a reluctance to use this paradigm in practice, stemming from a lack of understanding of its theoretical and philosophical background, concerns about it not being seen as ‘orthodox’ and lack of experience and skills.5 However, there is now an increasing acceptance that quantitative and qualitative methodologies can actually be compatible.8
There is clearly a demand within primary care, and elsewhere, for training in qualitative research, but the necessity of fitting this around the many other competing demands on practice staff (i.e. GPs, community nurses, health visitors, practice nurses, etc.) can result in episodic, de-contextualized (and potentially confusing) training provision. In order to be financially viable, of course, workshops need to be marketed, and this may contribute to the unrealistically high expectations which many practitioners may bring to short training courses. There is a tension between the requirement for a timely response to the specific needs of practitioners about to embark on research (whether or not training is forthcoming) and the need to ground research training in a broader theoretical framework, which may well involve a significant conceptual shift for many practitioners.
We cannot claim to have resolved this problem, but the workshops described here are an attempt at a compromise. They aimed to de-mystify the research process, and to give participants an insight into the complexities of qualitative research whilst recognizing the breadth of participants' interests, their varying degrees of commitment to qualitative methods, the different uses to which they were likely to put newly acquired interviewing skills and the constraints on their time.
Two articles9,10 were distributed to participants in advance and they were asked to read these in preparation for the first workshop. Further reading lists were provided at the workshops and we have continued to elicit feedback from participants as they have read around the topic, reflected and sought to put their skills into practice on several research projects. The unfolding nature of the workshops has allowed us to remain in close contact with those who took part, and the discussion incorporates their comments and evaluation as well as reflections from the two facilitators.
Structure and content of the workshops
At the outset, participants were advised that they should make a commitment to attend both workshops originally offered as a package. The topic for the notional research project to be explored in the workshops had already been selected by the workshop facilitators. In effect, this plunged participants into the research process some way from the starting point, as framing the research question could, in itself, have merited an entire workshop. Indeed, this task has been described as being perhaps "the most challenging part of developing a solid proposal."11 (p.16).
The notional topic selected was views about providing services for drug users within primary care (chosen due to its amenability to research using semi-structured interviews, its topicality and the previously declared interest of two group members). Inevitably, some individuals had more knowledge of service provision for drug users than had others, so we decided to begin the workshop by showing a short video on the experience of one general practice as shared/common ‘stimulus material’ (to help focus discussion).
The first two workshops comprised a mix of oral presentations and practical exercises, carried out in pairs and with the group as a whole. At the first workshop, initial ideas about possible interview questions were formulated and a consensus interview schedule was developed.
A video of an interview had been produced for this session and involved R Barbour in interviewing one of the participants about his experience of providing services to drug users and of prescribing methadone. It was considered that viewing interview video extracts would give participants a flavour of the sort of interaction involved in semi-structured interviewing and emphasize the differences between a research interview and a consultation.6 (Part of this interview was also transcribed and used to illustrate a short presentation on coding delivered at the second of the three workshops.)
Everyone also had an opportunity to experience both interviewing and being interviewed, with participants assigned to a succession of three different partners so that they were working with someone new for each exercise. The three pairs included:
- One to work on formulating an interview schedule
- One where an individual was an interviewer
- One where that same individual was an interviewee.
Logistically, however, this proved something of a challenge, especially when one of the participants did not turn up on the day, thus throwing our carefully designed programme into temporary chaos, as we attempted to introduce the workshop whilst simultaneously attending to this ‘housekeeping’ problem. In planning the workshop, we had anticipated the need for an additional researcher to act as a ‘floating’ assistant, able to respond to any problems or extra requirements as these arose. In the event, this contingency plan paid dividends: we were fortunate to have acquired the services of a colleague with many years ‘extra curricular’ experience as a Brownie leader. She was able to transfer these organizational skills and solved the problem of the missing participant, whilst maintaining the integrity of our mixed pairs.
All ‘interviews' were taped (and were fully transcribed), which necessitated hiring several small rooms to ensure good quality recordings, and which added considerably to the costs of running the workshop. This meant that we had to bring six tape recorders, a dozen tapes, transformers and batteries—and spares—to the workshop. Some time had to be allowed for rearranging furniture, placing tape recorders where sound would be picked up, ensuring that ‘interviewers' were confident to operate tape recorders, collecting and labelling tapes and, on occasion, for retrieving engrossed pairs from rooms at the end of the exercise. Two 20-minute slots were allocated for interviewing, this being determined by the amount of secretarial time available for transcribing and also by the design of the workshop. In the event, this proved a realistic time in terms of allowing the ‘interviewers' to pose all of the questions on their schedules, and perhaps gave them a taste of operating under constraints such as one may encounter when attempting to interview busy professionals in the field.
Condensing a notional project into three workshop sessions meant that we did not have time to discuss broader issues of research design and sample selection. The 12 interviews which were carried out in the course of the workshop, however, represent the work which might be involved in piloting an interview schedule or a small exploratory study, bearing in mind the important qualification that our ‘respondents' were all self-selected. If we were to be carrying out a ‘real' study of professionals' views on primary care service provision for drug users, we would most probably decide to select a purposive sample12 to obtain a spread in terms of professional groups, length of experience and nature of experience (generic and specialist). This would allow us to explore diversity,13 rather than selecting a ‘representative’ sample, which is more commonly the province of quantitative studies.8
The second workshop, which allowed participants to begin analysing the data they had generated, was organized to take place 1 week later, allowing time between the two workshops for interviews to be transcribed. The other decision which was taken out of participants' hands related to transcribing the interviews. We had opted for full transcription, although researchers might sometimes opt instead for partial transcription, augmented by detailed field notes, or might carry out analysis by listening to tapes rather than reading transcripts. We also recommended to our ‘interviewers' that they collect some supplementary information from ‘interviewees' by means of a structured pro-forma. Workshop participants decided to use this approach to collect details of current posts, length of experience, previous posts and training completed. It is important to recognize that all of these decisions in the field tend to be more complex and protracted.
Formulating an interview schedule
Although practitioners may be accomplished at listening to patients, the focus of their consultations is different from that of a research interview. Qualitative research encounters involve a departure from the model of researcher as expert. Firstly, in this case, it is the respondent who possesses the information which the researcher wishes to collect and, secondly, at least with semi-structured interviewing, questions put to respondents depart from the precision required in questionnaires or surveys, resembling, instead, everyday conversational gambits. Indeed, in-depth interviews have been described as "guided conversations".14 Qualitative researchers can often feel de-skilled in the field, by virtue of the need to ask apparently ‘damn-fool’ questions, as they seek to put themselves in the position of ‘conceptual stranger’ of anthropological ventures, with the added difficulty of attempting this within the confines of their own society.
Our workshop participants tended to start off by suggesting much more complicated and precise questions than the six rather general questions which we finally agreed to include in the ‘consensus schedule’. We concluded that this could be attributed partly to their prior research training and a residual attachment to the quantitative paradigm.15 This also highlights the skills which practitioners are likely to have already developed in ‘moving discussion on’ within time-limited clinical encounters. This exercise took considerably longer than the workshop facilitators had anticipated, and perhaps we had underestimated the need to revisit the epistemological underpinnings of qualitative research. Ideally, we would have liked to have spent even longer on this task, which, hopefully might have led to participants having a greater sense of ‘ownership’ of the resultant interview schedule. In the event, the limited time available necessitated a fair amount of input from the facilitators.
After some discussion, it was agreed that we would use a small number of general questions, backed up with prompts to ensure that more specific topics were addressed, and follow-up questions to explore the implications of replies to the main questions.16 The ‘consensus schedule’ consisted of the following:
- What do you think about services being provided to drug users in primary care?
- What priority should be given to providing services to drug users?
Prompt—Implications for services to other patients?
- What are the practical difficulties?
Prompts—Issues for other patients/safety?
Workload?
Demands on time?
- What types of support should be available for drug users?
- What do you think can be achieved in working with drug users?
- What wider issues are involved?
Revisiting this interview schedule with the benefit of hindsight, the facilitators' input is more apparent, and it might have been worth taking time in the workshop to draw attention to the ‘theorizing' already influencing the process of question content and wording. The schedule does not pre-suppose detailed knowledge of the topic area and could be used with individuals with varying degrees of experience. It does not seek to gauge interviewees' levels of knowledge, but is designed to allow ‘interviewees' to focus on issues which are salient to them (whilst also attempting to elicit views on certain issues identified as being of prime importance for the notional research project). It also encourages ‘interviewees' to think about both practical issues and wider policy issues involved in service provision. The schedule thus allows for the expression of a wide range of views and does not assume, for instance, that all ‘interviewees' will necessarily be in favour of the provision of such services within primary care. The open-ended nature of the questions may, even, potentially invite ‘interviewees' to ‘problematize' the questions being posed, define them as irrelevant and suggest an alternative focus.
Using an interview schedule
Commenting on a research methods workshop run for undergraduates, Collins17 says, "There is a tendency at the outset for students to see the interview as a kind of smash and grab opportunity in which they accost some innocent bystander and relieve them of whatever useful ‘data’ they may have" (para 1.1). This notion can, of course, be reinforced by the stress on question wording which suggests a ‘model’ interview tool at the expense of emphasizing the interactive nature of interviewing. Offering an alternative view of interviewing, Mason argues that it is more helpful to speak of data ‘generation’ rather than data ‘collection’ and reminds us that ‘method’ in qualitative research implies more than technique, but, instead, refers to a wide range of "intellectual, analytic and interpretive"18 activities (p. 36).
Notwithstanding the apparent casualness of question wording, the art of asking questions in qualitative research is an extremely subtle one, and it is quite hard for the new researcher to strike the required balance between precision and flexibility. When we examined transcripts of interviews carried out at the workshop, we found, for example, that half of the ‘interviewers' re-worded the first question. The agreed—but, admittedly, potentially ambiguous—first question on the schedule "What do you think about services being provided to drug users in primary care?" was intended to elicit people's views about the appropriateness of providing services for drug users in primary care. Three ‘interviewers' asked, "What do you think about the services being provided to drug users in primary care?" The extra word changes the meaning and directs the ‘interviewee' to consider the content and adequacy of service provision, which may or may not include observations on the appropriateness of primary care service provision. Some ‘interviewers' also wandered off the schedule to explore topics which clearly held interest for them personally, sometimes at the expense of posing all of the key questions.
Occasionally ‘interviewers' asked leading questions, most notably in relation to the fifth schedule item, "What do you think can be achieved in working with drug users?" Two ‘interviewers' asked the question exactly as it appeared on the schedule and then added, respectively, "Is anything achieved?" and "Do you think we can do anything about it?"
As a result of the shared understanding of the purpose of the interviews and the group work which we had carried out in formulating the interview schedule, ‘interviewees' were unusually well informed and frequently did provide responses to the questions which they should have been asked instead of replying to those articulated by ‘interviewers'. Outside of the protected workshop environment, though, such departures from questions are likely to result in data which address a different set of issues, rendering problematic the making of comparisons between individual transcripts. This could constitute a serious problem in a study which employed several researchers, although this set-up is not common in qualitative research.
Using prompts
It was suggested that ‘interviewers' make use of prompts in order to encourage potentially reluctant ‘interviewees' to voice their ideas. The use of prompts in eliciting responses reflects the notion of silences (or failure to address specific issues) as problematic, but amenable to being rectified through technique: if only the researcher can develop enough skills, he or she can coax the reticent interviewee to reveal all. Of late, several commentators have drawn attention to this underlying and largely unquestioned assumption, and have suggested that silences (or failure to address particular issues) may, in themselves, be illuminating.19,20 However, for the purposes of this workshop, we took a pragmatic approach to silences, as our priority was to collect data on the same topics from all respondents in order to enable comparison and aggregating of findings, rather than to explore in depth how salient particular issues were for interviewees. In practice, there is, of course, frequently a tension between these two aims.
Skilful prompting (using both prepared gambits and the ability to pick up on and explore comments as the interview unfolds) is the key to successful interviewing. In addition to ensuring that particular topics are addressed, opportunistic prompts can also be used in order to get the respondent to contextualize their views. One example from the workshop transcripts shows one of the participants using this device to good effect, asking the ‘interviewee', "To what extent do you think that other people share your views about supporting drug users with whatever it takes. Do you think that's a common attitude?" Other examples of opportunistic prompting might include asking a respondent to attempt to explain why he or she holds particular views, or asking whether his/her attitudes have changed significantly over time. Thus semi-structured interviews provide the means to move beyond a statement of opinion to an explanation of that opinion. These uses of prompts anticipates analysis, and interviewees can even provide new leads in terms of conceptualizing our data. Anthropologists have long remarked on the usefulness of key informants in this respect, and skilful use of prompts can allow us to harness interviewees' expertise in a similar way. Prompts can also be used to clarify what the interviewee is saying, or can be used to summarize and invite confirmation from interviewees of the interviewer's understanding.
Although thinking on one's feet as an interviewer can produce particularly useful supplementary questions which may be added to interview schedules as a project progresses, it is also very easy to slip into asking leading questions, which may result in interviewees agreeing out of politeness or simply drying up. This can happen not only to novices, but also to experienced researchers. Indeed, this is exactly what we observed occurring in the ‘training’ video we used, which showed R Barbour interviewing one of the workshop participants:
CJ: As a GP there are two sorts of services we provide. One is general medical services, which we provide to everyone, which I'm quite happy with —that's not a problem. Then there's the more specialized drug abuse service—the prescribing methadone and things like that, which we do provide in our practice—but I wouldn't say I'm happy about it. I wish we didn't have to. I wish there was another way this service could be provided ...but there isn't ...RB: So you see it as a ...sort of ...necessary evil?
CJ: Yes ...we ...ell (raising eyebrows and smiling) ...
RB: Maybe that's putting it a bit too strongly?
CJ: I mean, it's a specialized area of work and we've had to develop some specialization in that. But there's a feeling that we don't have all the specialist knowledge that we should have.
However, interviews are conversations and it can be possible to ‘effect a repair’. In this case, a hasty amendment, "Maybe that's putting it a bit too strongly?" provided the interviewee with the opportunity to qualify his reply.
Given the time constraints of this workshop and the very brief training on using prompts which was provided, participants sometimes found it difficult to use these effectively. Probably due to their eagerness to ensure that they did not miss the opportunity for exploring particular issues, some of the ‘interviewers' ran a question into a prompt: for example, one asked, "What priority should be given to providing services to drug users? So, really what we're looking at here is the implications for services to other patients." This strategy, whilst it may be effective in getting the respondent to address implications for other patients, does foreclose discussion and may prevent the individual from furnishing interesting ideas about, for example, organizational or funding implications. "When we train interviewers, perhaps too much emphasis is placed on asking questions, when the real skill may be listening"20 (pp. 296–297). After completing her stint as ‘interviewer', one of the workshop participants commented that she had found herself repeating the phrase, "That's interesting", and asked whether this was good practice. This comes into the category of "encouragement questions"16 and can be a useful gambit, which underlines the importance of conveying to interviewees that the researcher is listening.
Feedback from participants and implications for future workshops
All the participants attended the two afternoon workshops, seven attended the optional third, several met with the facilitators informally and all commented on drafts of this paper. This reflects their commitment to and interest in our approach. The exercises involving interviewing in pairs were not experienced by many of the participants as enjoyable—even within this supportive environment—and one participant commented afterwards that she had "felt exposed". In retrospect, we wondered whether it might have been preferable to have asked workshop participants to generate their data outside the workshop environment. However, participants concluded that it had been especially useful to have hands-on experience of both sides, as ‘interviewer' and ‘interviewee', and that this had alerted them to the vulnerability of those at the receiving end of interview questions. Practising new skills in front of one's own professional peer group, some of whom are known to each other and likely to have professional contact in the future, undoubtedly created a highly contrived social situation and perhaps resulted in additional stress. However, carrying out ‘interviews' in the ‘virtual reality' afforded by the workshop format was considered extremely useful in that it provided first-hand experience in a situation where the results of the interviews per se did not matter.
With the benefit of hindsight, we realize that we under-emphasized the theoretical base on which the ‘consensus interview schedule’ was founded. Perhaps in addition to providing the stimulus material in the form of a video, we could have pre-circulated participants with a couple of ‘orienting paragraphs’, which could have summarized developments with regard to providing services for drug users, the policy rationale behind providing these within primary care, and the level of knowledge regarding practitioners' experiences and views. Had the interviewing been part of a real—as opposed to ‘virtual’ —project, of course, such a text would already have been available in the form of the introductory section of the application form for funding which we would have had to complete for the funding body involved.
Professional backgrounds emerged as an important influence on interpreting data (discussed in the accompanying paper) and is also an important consideration in terms of planning a study. Interviewing one's own patients, for example, may not be advisable, depending on the topic.21 Researchers should also bear in mind the possible impact of interviewers' professional identities (where these are known) on the data collected. Other aspects of the researcher's identity (as presented or as perceived by respondents) will also affect the content of the interview.22 One of our workshop participants was debating whether it was appropriate for him (as a male GP) to interview women who had had repeat terminations of pregnancy. We concluded that there was no right or wrong approach, provided that one recognized the impact of gender and professional identity and acknowledged that individual interviewees may well collect different data, although using the same methods, or, indeed, the same interview schedule.23 In this particular case, women might be even more reluctant to discuss repeat terminations with another woman who could be perceived as a conventional and potentially disapproving mother.
One of our participants commented that she would have benefited from having attended the workshop prior to conducting some interviews which she had just carried out. Indeed, as with all training courses, timing may be an important issue: unless one has the opportunity, relatively soon after training, of putting into practice what one has learnt, such knowledge may be lost. The whole research process, however, is a long, slow one where practitioners gather skills and experiences as a project progresses. Perhaps the effects of workshops attended is cumulative, and ideas, theories, etc. gradually become embedded in their ‘research consciousness' as they undertake more research-oriented activities. As they travel the research road, they can also highlight areas where expertise is lacking and identify needs for specific training.
As one participant pointed out, we did not attempt to ascertain prior to the workshop what level of experience participants previously had of interviewing. It is possible that collecting more information about participants might have helped us to tailor the workshops more closely to their needs. Bearing in mind the previous point regarding ‘using or losing’ skills, one potential solution might be to offer additional research methods workshops in specific areas such as how to use prompts or developing coding skills, for project teams or for individuals at similar stages in their research. The areas not covered or glossed over in the course of the three workshops discussed here (e.g. developing a research question; designing a project; selecting a qualitative sample) could provide a starting point for identifying such needs.
Organizing the workshops took up a lot of time, and this is unlikely to be reduced if, as we suggest here, future workshops are tailored more closely to the needs of individual researchers, project teams or researcher ‘cohorts’. The workshops were also relatively expensive to run, due partly to the need to hire a number of rooms. Costs might be kept down by holding workshops in a surgery or University Department over a weekend when there is considerably less demand on space. The one area where we would argue that cuts could not be made, however, is in terms of staffing, as it is essential to be able to respond to practical problems or emergencies as these arise.
The workshops discussed here were a valuable learning experience for the organizers themselves and were viewed as the first step in a long and unfolding process for the participants. The research consciousness and activity in the area in which WoReN is working was quite low at the time, and this workshop was an early attempt to raise awareness and activity. Undertaking any research, whatever paradigm it is rooted in, is a long and time-consuming task, and it is hoped that these workshops illuminated this to a degree. Workshops such as this could perhaps take place at more appropriate times in the careers of these professionals as they need to rise to challenges which present themselves as they undertake research activities.
The constituency for our workshops comprises busy professionals, who can take on only a certain amount of information and training with regard to something not immediately related to their working life. In retrospect, 18 months on, it was surprising that all participants did manage to attend both workshops originally scheduled and that so many made time to attend later optional workshops and further informal discussions. Three of the original participants have attended subsequent related workshops run by R Barbour and WoReN, illustrating the need for continual reinforcement of methodological issues and, perhaps even more importantly, for general peer support and the opportunity to discuss progress and pitfalls fallen into or narrowly avoided. Novice researchers need this nurturing, and this is one of the needs which WoReN does attempt to meet, both as a stand-alone organization and with considerable support from academic consultants.
Acknowledgments
We would like to thank Carol Emslie, Eleanor Guthrie, Guro Huby, Helen Richards and Graham Watt for helpful comments on a draft of this paper. We also gratefully acknowledge the assistance of Michelle Middleton (Administrator) and secretarial support from Liz Bradley, Maria (Winnie) Wilson and Abby Chambers. The one participant who registered for the workshops, but was unable to attend, was told by her Trust, at short notice, that she could not be released after all, due to staff shortages.
References
1 Miller WL, Crabtree BF. Clinical Research. In Denziin N, Lincoln YS (eds). Strategies of Qualitative Inquiry. Thousand Oaks, CA: Sage Publications, 1998: 292–314.
2
Harding G, Gantley M. Qualitative methods: beyond the cookbook. Fam Pract 1998; 15: 76–79.
3 Helman CG. Research in primary care: the qualitative approach. In Norton PG, Stewart M, Tudiver F, Bass MJ, Dunn EV (eds). Primary Care Research: Traditional and Innovative Approaches. Research Methods for Primary Care, Volume 1. Newbury Park, CA: Sage Publications, 1991: 105–124.
4 Miller WL, Crabtree BF. Primary care research: a multimethod typology and qualitative road map. In Crabtree BF, Miller WL (eds). Doing Qualitative Research. Research Methods for Primary Care, Volume 3. Newbury Park, CA: Sage Publications, 1992: 3–28.
5
Murphy E, Mattson B. Qualitative research and family practice: a marriage made in heaven? Fam Pract 1992; 9: 85–91.
6
Britten N, Jones R, Murphy E, Stacy R. Qualitative research methods in general practice and primary care. Fam Pract 1995; 12: 104–114.
7 Dowell J, Huby G, Smith C (eds). Scottish Consensus Statement on Qualitative Research in Primary Health Care. Dundee: Tayside Centre for General Practice, 1995.
8 Barbour RS. The case for combining qualitative and quantitative approaches in health services research. J Health Serv Res Policy 1999; 4: 39–43.[Medline]
9
Britten N. Qualitative interviews in medical research. Br Med J 1995; 311: 251–253.
10 Griffiths F. Qualitative research: the research questions it can help answer, the methods it uses, the assumptions behind the research questions and what influences the direction of research. Fam Pract 1996; 13 (Supp.1): S27–S30.
11 Marshall C, Rossman GB. Designing Qualitative Research. 2nd edn. Thousand Oaks/London: Sage Publications, 1995.
12
Mays N, Pope C. Rigour and qualitative research. Br Med J 1995; 311: 109–112.
13 Kuzel AJ. Sampling in qualitative inquiry. In Crabtree BF, Miller WL (eds). Doing Qualitative Research. Newbury Park, CA: Sage Publications, 1992: 31–34.
14 Lofland J, Lofland LH. Analyzing Social Settings. 3rd edn. Belmont: Wadsworth Publishing Company, 1995.
15
Malterud K. Shared understanding of the qualitative research process: guidelines for the medical researcher. Fam Pract 1993; 10: 201–216.
16 Bowling A. Research Methods in Health: Investigating Health and Health Services. Buckingham: Open University Press, 1997.
17 Collins P. Negotiating selves: reflections on ‘unstructured’ interviewing. Sociol Res Online, 1998; Vol. 3, No. 3. <http://www.socresonline.org.uk/socresonline/3/3/2.html>
18 Mason J. Qualitative Researching. London: Sage Publications, 1996.
19 Barbour RS. Engagement, presentation and representation in research practice. In Barbour RS, Huby G (eds). Meddling with Mythology: AIDS and the Social Construction of Knowledge. London: Routledge, 1998: 183–200.
20 Poland B, Pederson A. Reading between the lines: interpreting silences in qualitative research. Qual Inquiry 1998; 4: 293–312.
21 Hoddinott P, Pill R. Qualitative research interviewing by general practitioners: a personal view of the opportunities and pitfalls. Fam Pract 1997; 4: 307–312.
22 Green G, Barbour RS, Barnard M, Kitzinger J. Who wears the trousers? Sexual harassment in research settings. Women's Stud Int Forum 1993; 16: 627–637.
23
Barbour RS. Mixing qualitative methods: quality assurance or qualitative quagmire? Qual Health Res 1998; 8: 352–361.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||