Doing Qualitative Research in Education Settings

4 Analyzing Qualitative Data

How do I make sense of the mountain of data I have collected? How do I start to analyze my data? How will I know when I am finished? How exactly do I do data analysis? This last question is one I hear all the time. Even students who have taken the courses and read the books still ask, “How do I analyze my data?” As someone who teaches some of the courses and recommends many of the books, I am sometimes frustrated that they seem genuinely to have no clue. While acknowledging the complex and idiosyncratic nature of data analysis processes, one of my objectives in writing this book was to provide first-time qualitative researchers with frameworks that would provide enough guidance to actually allow them to do qualitative data analysis. Looking at the books I had been recommending left me with the sense that, with few exceptions, data analysis processes have not been well described in the literature . In volumes of 300 pages, as few as 9 or 10 pages have been devoted to data analysis. When full chapters are devoted to data analysis, the information provided is often general and abstract, leaving new researchers without much concrete guidance. Data analysis is portrayed as messy, cumbersome, inductive, creative, challenging, subjective, nonlinear, labor-intensive, exhilarating , and time-consuming; but analysis processes are seldom spelled out with sufficient clarity that novice researchers are confident at getting started. Postpositivist researchers get the most guidance from the literature (e.g., Glaser & Strauss, 1967; Miles & Huberman, 1994; Spradley, 1980). This makes sense since research methods in this paradigm are characterized by much more structure than procedures used in other approaches. Constructivist 147 CHAPTER FOUR Analyzing Qualitative Data researchers often adapt data analysis procedures developed by postpositivists (e.g., Goetz & LeCompte, 1984; Lincoln & Guba, 1985; Van Manen, 1990), and this makes sense as well, given that both are interested in uncovering “reality”—one, the reality presumed to exist in nature, and the other, the realities constructed by social participants. Critical/feminist researchers often use data collection methods adapted from postpositivist approaches, but their emphasis on the political nature of knowledge leads them to analyses that are undertaken within particular political frames of reference (e.g., Carr, 1995; DeVault, 1990, Reinharz, 1992). Data analysis is more difficult to characterize for poststructuralist researchers. Those who gather data using methods adapted from other paradigms will adapt analysis procedures as well; those doing deconstructive or genealogical work have their own analytic approaches that have closer connections to continental philosophy and postmodern literary criticism than traditional qualitative research (see Flax, 1990; Graham, Doherty, & Malek, 1992; Sarap, 1993). I will try to make paradigmatic similarities and differences in relation to data analysis clearer as the chapter unfolds. Data analysis is a systematic search for meaning. It is a way to process qualitative data so that what has been learned can be communicated to others . Analysis means organizing and interrogating data in ways that allow researchers to see patterns, identify themes, discover relationships, develop explanations, make interpretations, mount critiques, or generate theories. It often involves synthesis, evaluation, interpretation, categorization, hypothesizing , comparison, and pattern finding. It always involves what Wolcott calls “mindwork” (1995, p. 233). Researchers always engage their own intellectual capacities to make sense of qualitative data. Even when computer programs are used to assist in the mechanics of sorting data, only the intelligence, creativity , and reflexivity of the human mind can bring meaning to those data. I conceptualize the general data analysis process as asking questions of data. What kinds of questions are asked is related to what kind of research is being done within what set of paradigmatic assumptions. Postpositivist researchers doing interview studies will likely start analysis with different questions than critical/feminist researchers doing case studies. For example, the former may read their data with a question such as this in mind: What are the criteria my informants use to make judgments about promoting or retaining students? The latter may ask: How does race influence decision making about promotion and retention in this school? Different approaches and paradigms lead to different analysis strategies, but the general approach I am proposing is built on the assumption that important information is in the data, and by systematically asking the right questions of the data, that information can be revealed. Much of the rest of this chapter describes alternative models based on this assumption. Obvi148 Doing Qualitative Research in Education Settings [3.139.107.241] Project MUSE (2024-04-26 15...