Benjamin Crabtree & William Miller (Eds.) (1999). Doing Qualitative Research (2nd edition). London: Sage, 406 pages, ISBN 0-7619-1497-8 (hardcover) $ 72.95, ISBN 0-7619-1498-6 (paperback) $ 35.95

Abstract: This text is a must for the qualitative researcher whose work primarily lies in the field of healthcare. The reader is taken through an overview of qualitative methods and then shown various data collection techniques and data analysis strategies along with some exceptional research designs. The detailed exemplars of real research problems and suggestions for solutions are invaluable. The book should not be used for introduction to or survey of qualitative methods instruction because of the complexity of strategies and research designs. For the experienced or somewhat experienced researcher, however, this text is priceless.

Key words: qualitative research, interpretivism, case study, grounded theory, participant observer, ethnography, phenomenology, hermeneutic editing, immersion, crystallization, code manual

Increasingly, over the span of the past decade, qualitative research texts and handbooks have appeared in print. It goes without saying that the interpretive paradigm has gained importance in the fields of sociology, education, nursing, and now, the broader area of healthcare. The task of designing a qualitative study principally for novices can be quite intimidating, considering there is no fundamentally "correct" way to begin. Secondly, with traditions ranging from biography, phenomenology, grounded theory, ethnography, and case study (see Qualitative Inquiry and Research Design, CRESWELL, 1998), it is imperative that researchers connect the appropriate tradition with the research questions that should drive the design, focus, and analysis. Benjamin CRABTREE and William MILLER's Doing Qualitative Research brings multiple perspectives/traditions together with an arrangement that maps out how to get started and carries the reader through to the writing up of the finished study. [1]

Doing Qualitative Research is a rich, comprehensive text that provides both new and veteran research investigators with various approaches to data collection along with analytic tools for organization styles. Moreover, the book provides examples of actual research studies complete with design, research questions, research dilemmas, and validity and reliability checks. CRABTREE and MILLER and the host of other authors offer not only multiple approaches to qualitative research design and analysis, but also furnish their own experiences and accounts pulled from recent or ongoing projects without laborious prescription. The refreshing insights in Doing Qualitative Research make this book well deserving of the reader's investment. [2]

The editors divide the book into six main areas of foci including two chapters in a summary section. The introduction importantly lays the foundation for the premises of the naturalistic/constructivist perspective. Following the brief introduction, the first section contains just one chapter by editors CRABTREE and MILLER. Rather than re-writing the contrasting "Two Paradigm" approach (positivism versus naturalistic inquiry) that seems to pervade prevailing thought (e.g., that is, single reality versus multiple realities), the first chapter explains three paradigms: Materialistic inquiry, constructivist inquiry, and critical/ecological inquiry. Materialistic and constructivist paradigms are compared and contrasted, and an overview of the circular interpretivist process is presented. [3]

Data collection is the focus of Section Two. Anton KUZEL elaborates on various sampling strategies for qualitative research, particularly targeting smaller populations that are information-rich. For example, case studies of chronic pain patients can target personal experiences of pain and suffering, enabling the researcher to better understand the phenomenon of pain in non-biomedical terms. KUZEL cites GUBA and LINCOLN (1989) in reporting that maximum variation sampling, which surveys variations in order to target specific patterns, is the preferred strategy for constructivist inquiry. Drawing from the work of MORSE (1995), CRESWELL (1998), and MAXWELL (1996), KUZEL reminds us to choose a cohesive homogeneous sample, since the researcher's goal is not to generalize. Because the goal is to reach saturation, practicing theoretical sampling from negative cases can strengthen the study's outcomes. Finally, in using a grounded theory approach, where the end goal is development of a theoretical model and where saturation is more complete, the easier it becomes to advance theory. KUZEL also emphasizes the importance of using the appropriate paradigm. He finishes the chapter with guidelines for determining the best sampling strategy. [4]

The reader is next taken through the far ranging process of participant observation. Stephen BOGDEWIC provides disparate takes on the continuum of participant/observer, explains their roles, and furnishes thorough description of the process of acquiring field notes. To illustrate, first he establishes the importance of participant observation, establishing the validity of the "natural" versus laboratory setting in which data is collected. The strength of collecting data in the natural setting is the potential for better understanding more varied aspects of human communication. BOGDEWIC urges the reader to consider the method of entry well in advance of beginning to collect data. He states several questions for contemplation such as "Who is involved?" and "What is the phenomenon happening?" and "When does the phenomenon occur?" Perhaps the most important decision is the reader's specific role in the continuum of participant/observer, from "complete observer" to "complete participant." BOGDEWIC stresses that the distinction between these two ends has become quite fuzzy. He ends the chapter with guidelines for taking field notes. [5]

Interviewing is covered in two chapters. Valerie GILCHRIST and Robert WILLIAMS contribute a chapter on key informant interviewing. They stress the importance of selecting appropriate people to be key informants (KI) by explaining who a KI should be. The authors emphasize the importance of determining the purpose of the KI and the necessity of establishing criteria for good selection. These criteria primarily fall under knowing the type of information needed for the research and the compatibility of the KI with the researcher, especially for repeated interviews. GILCHRIST and WILLIAMS assert that a successful interview is contingent upon not directing the conversation, but deciding what is meaningful enough to the KI to introduce in the interview. Additionally, drawing from the work of SPRADLEY (1970), the interviewer must conclude strategically what descriptive questions, what questions of inclusion/exclusion, and what questions of verification to use. The authors also point out that a successful interview is contingent upon obtaining correct information, which means not assuming understanding but making sure to confirm comprehension. William MILLER and Benjamin CRABTREE follow with depth interviewing, affording examples of SPRADLEY's grand tour questions along with examples of good prompts and probes. This chapter builds on the previous one but plunges into detail with case examples and a brief section on transcription of the interviews. [6]

The section ends with a chapter on focus groups. Judith BROWN traces the history of focus groups in the primary care setting before composing a detailed description of the purpose and process of focus group research. Specifically, BROWN says the purpose of a focus group is to gather information based on the participant interactions. It is extremely imperative to avoid leading questions while still providing guidance to the topics discussed. Additionally, she provides advantages and disadvantages for heterogeneous versus homogeneous groups. BROWN reminds the reader to devise the research questions in advance so that the appropriate sample can be gathered. [7]

Part III consists of interpretation strategies. It opens with an overview of the five main steps of the "dance of interpretation" by William MILLER and Benjamin CRABTREE. These steps are describing, organizing, connecting, corroborating/legitimating, and representing the account. The three main types of strategies for interpretation, including a grounded hermeneutic editing approach, a template organizing style (codes and code manuals), and immersion/crystallization, are detailed in the following substantive chapters. [8]

The editing approach contributed by Richard ADDISON sets up the grounded theory orientation with an extended research example throughout the chapter. ADDISON's perspective follows the grounded theory model set forth by GLASER and STRAUSS (1967) in terms of constant comparison. His angle, however, highlights varying aspects of everyday events and practices by drawing from "social, cultural, historical, economic, linguistic, and other background aspects" (p.149). ADDISON lays out very clearly the assumptions of this approach. To name a few, a beginning focus for researchers should be on misunderstandings in human interactions, on understanding behaviors of their participants, including not only communication but also other behaviors and practices, and on a belief that "facts" are value-laden. Investigators, therefore, must consider their biases in their interpretation of the data and framing of the research questions and problems. [9]

Benjamin CRABTREE and William MILLER offer the template organizing style via a codebook as a way to approach a study if the topic of the study already has an attached theoretical perspective. They weigh advantages and disadvantages of the other organizing styles in contrast with the template style. For example, using a template or code manual may be more focused and certainly more time efficient than other types of styles because the research can be geared toward specific pieces of the text. In the hermeneutic editing approach, the analyst must attend to more specific attention by moving line by line in the interpretation phase. Even more time-consuming is the immersion/crystallization approach, because a very prolonged time just reading and absorbing the text is needed. Another comparison among the three main strategies is the specific codes used for analysis. The template approach uses codes devised apriori, whether from previous studies or from theoretical perspectives. On the other hand, both the editing approach and the immersion/crystallization position begin with more careful reading of the text and inductively trying to discover codes that either lean toward a theoretical perspective or lead to the discovery of a brand new theory. [10]

The next chapter is helpful in unlocking the organizing style of crystallization/immersion. Stemming from the notion that the researcher is truly the analytic tool, Jeffrey BORKAN asserts that the investigator be emersed in the data and rely more heavily on intuition than in the other styles of working the circular steps of the research process. He stresses that the analysis does not necessarily come after the data has been collected, but asserts that vital insights might occur during the data collection process. As the analysis is being carried out and as new themes and codes are emerging, therefore, the researcher can better decide how to proceed in further data gathering. [11]

In the last chapter of the third section, Lynn MEADOWS and Diane DODENDORF, in a concise and yet comprehensive manner, introduce computerized data management and computers as an analytic aid. These researchers do a great job in describing the different main types of software programs such as basic word processing. They provide brief descriptions of text managers as tools to organize text for easy retrieval of information, mentioning Folio Views®, ask-Sam®, MAX®, and ZyINDEX®. Another main type of program is the code-and-retrieval type that allows the analyst to "chunk" data for coding and grouping. These include HyperQual®,NUD*IST®, QUALPRO®, and Ethnograph®. MEADOWS and DODENDORF stress that while these programs do not do the coding for the researcher, they do facilitate in code building by managing codes and helping to make connections among categories of codes. Software programs that aid in theory building are NUD*IST®, ATLASTM, HyperRESEARCH®, QCA®, and AQUAD®. Additionally, programs that help in building conceptual networks and feature graphics are Inspiration®, Decision Maker®, ATLASTM, MECA®, and NVIVO®. [12]

Part IV lends specific understanding to distinct studies such as narrative analysis, the utilization of video recordings, unique case studies, and participatory inquiry. Jessica MULLER begins with narrative analysis in primary care settings, underscoring the importance of bringing to the biomedical model the "voice of the lifeworld" (MISHLER 1984, p. 63). [13]

Virginia ELDERKIN-THOMPSON and Howard WAITZKIN describe videotaping in specific kinds of research, particularly in interventions and evaluations. To illustrate, giving feedback to healthcare providers in terms of their communication behaviors and in providing instruction to patients about a particular illness or condition, BANDURA's social learning theory (1977) has been shown in studies to be very effective. Showing the health care provider's exact behaviors can demonstrate much more effectively than observations alone what was positive and what improvements can be made. Likewise, chronically ill patients can "see" via tapes of other chronically ill patients how to modify and adapt their daily routines, and new diabetics can see other diabetics taking their own glucose readings. [14]

Virginia AITA and Helen McILVAIN trace the history of case study and explicitly lay out the theoretical framework, data collection, and sampling strategies for this method. Lastly in the section, the concept of participatory inquiry is introduced as a somewhat new concept within the social sciences. Janecke THESEN and Anton KUZEL divide participatory inquiry into the five traditions (see CRESWELL's Qualitative Inquiry and Research Design, 1998), providing understandable examples for even the inexperienced analyst. For those unfamiliar with CRESWELL, his divisions of five main traditions in qualitative research are biography, phenomenology, grounded theory, ethnography, and case study. Briefly, the biographical perspective looks at one person and fits his/her life with cultural interpretations of that historical period and focuses on markings of that life to aid in the cultural implications. Phenomenology follows the format of explicitly examining one particular phenomenon and how the carefully chosen participants make meaning out of that. (Grounded theory supports the crystallization/immersion already depicted in Paragraph 11.) Ethnography details description and interpretation of a social or cultural group and data is therefore gathered via observation and learned behavioral patterns and customs through prolonged submersion into that culture. Case study is examining a single participant through many different means of data collection such as interviews, observations, documents, and reports over an extended period of time. [15]

The fifth section illuminates the research craft as a whole with Benjamim CRABTREE and William MILLER's description of real social scientific/healthcare investigations in real social scientific/healthcare settings. They walk the reader through one in-depth study of family physician practices by utilizing the direct observational method. The study uses semi-structured and key informant interviews with the goal of understanding organizational components that have affected preventive health care. CRABTREE and MILLER make their journey through the comparative case study design. In a carefully developed plan, the research team was able to get 18 practices ranging from large and small facilities, independent clinics, health system-affiliate practices, and rural and urban clinics. The section ends with a chapter by Kristi MALTERUD that argues for a flexible investigative design: changes in key questions should be made along the way if the process evaluation shows it vital to the integrity of the research. [16]

Part VI puts forth a summary, but actually it seems to be much more like treading new waters. Richard FRANKEL elaborates on the standards of qualitative research. He stresses two primary points: 1. the research question should determine the method and, 2. the researcher is always the instrument. Finally, Lucy CANDIB, Kurt STANGE, and Wendy LEVINSON end with looking ahead to the future in qualitative studies. Some thoughts for consideration are that qualitative work can help researchers and especially health care providers understand and thus generate solutions for the complexity of obstacles they encounter on a daily basis. Additionally, with the colossal changes in the cultural practice of medicine, as well as the institutional changes themselves, research is demanding more interdisciplinary studies that can investigate these new waves of uncertainty. [17]

Doing Qualitative Research assumes one of the top spots for qualitative research texts. Like CRESWELL, DENZIN, LINCOLN, GUBA, among others, CRABTREE and WILLIAMS tackle the insurmountable job of taking the more common qualitative traditions, compiling data collection strategies, amassing data analysis methods, organizing everything according to paradigm, and writing it in such a fashion that is both articulate and reasonably stated. Finally, investigators in the field of health studies have at their disposal a reference containing examples from various health areas including BALINT groups, patient case studies, ethnographies of single clinics, and provider-patient communication, to name a few. The incorporation of perhaps unfamiliar concepts such as participatory investigation and clarification between hermeneutic editing versus immersion/crystallization earns this book a very favorable evaluation. I especially found the chapter on computerized data management software very informative and helpful because my own data collections tend to be of large size (13-100 conversational interactions or interviews) for qualitative studies. Computerized data management is becoming a real necessity for me and especially for others in similar research situations in order to adequately manage data. [18]

Although health researchers are likely to find Doing Qualitative Research a worthy investment, some points should be noted. Firstly, while most of the book is written in a rather sophisticated albeit educational fashion, several of the chapters do not follow this format and are written in what seems to be a "less-than-formal" manner. Secondly, I would have benefited from more discussion of triangulation, especially triangulation among co-collaborators. Team research appears to be on the upswing in qualitative studies, notably in investigations where the sample size and/or the amount of data collected would overburden a sole investigator. Examples and suggestions, therefore, for research partners would prove very useful. Thirdly, while Anton KUZEL alludes to discourse analysis (and conversation analysis) by way of the tradition of sociolinguistics, additional references and examples of discourse analysis throughout the rest of the book would have been contributive to the extensive coverage of the numerous qualitative approaches. Finally, I applaud the efforts of the editors and other contributors in augmenting our mindfulness of human subject issues. I might go so far as to say that with regard to recent events in research institutions (especially biomedical research facilities), an entire chapter on Institutional Review Boards (IRBs), their purpose, and especially the function of the investigators in maintaining protection of human subjects be explored in more detail. [19]

Doing Qualitative Research should be at the top of the faculty member's reading list who teaches health research courses and/or directs theses at the MA or Ph.D. level. That said, the complexity involved in piecing together research methods with strategies and understanding how these fit in with the various overarching paradigms would surely cause the typical undergraduate student enrolled in a senior-level methods class to steer clear of graduate school! [20]

Bandura, Albert A. (1977). Social learning theory. Englewood Cliffs, N: Prentice Hall.

Creswell, John W. (1998). Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks: Sage.

Glaser, Barney G., & Strauss, Anselm L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

Guba, Egon G. & Lincoln, Yvonna S. (1989). Fourth generation evaluation. Newbury Park: Sage.

Maxwell, Joseph A. (1996). Qualitative research design: An interactive approach. Thousand Oaks: Sage.

Mishler, Elliot G. (1984). The discourse of medicine: Dialectics of medical interviews. Norwood: Ablex.

Morse, Janice. M. (1995). The significance of saturation. Qualitative Health Research, 5(2), 147-9.

Spradley, James P. (1979). The ethnographic interview. New York: Holt, Reinhart, and Winston.

Jacqueline M. BARNETT is a lecturer in speech communication in the School of Communication at Sam Houston State University, Huntsville, Texas. She has taught a number of speech and composition courses including interpersonal communication, public speaking, and small group communication, to name just a few. Her research lies in the field of health communication, particularly in doctor-patient communication. Her most recent project included working on a federal grant-funded tobacco cessation project at a family practice residency clinic in Southeastern Texas. There she trained staff, nurses, and doctors in a behavior change approach to the medical interview. Currently she is finishing a large study of audio-recorded doctor-patient conversations.

Sam Houston State University
School of Public Communication
322K Dan Rather Communication Building
1803 Ave I
Huntsville, TX 77341, USA

Barnett, Jacqueline M. (2002). Review: Benjamin Crabtree & William Miller (Eds.) (1999). Doing Qualitative Research (2nd ed.) [20 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 3(4), Art. 3, http://nbn-resolving.de/urn:nbn:de:0114-fqs020432.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License