Analysis of Ethnographic Data Analysis of Focus Group

Ethnographic Data Analysis

Analysis of Ethnographic Data Analysis of Focus Group Data Interpretation Of Qualitative Findings

Analysis of Ethnographic Data

    Spradley's (1979) developmental research sequence is one method that is often used for data analysis in an ethnographic study. His method is based on the premise that language is the primary mechanism that relates cultural meaning in a culture. The task of an ethnographer is to describe cultural symbols and to identify their coding rules. His sequence of 12 steps, which includes both data collection and data analysis, is as follows:

1. Locating an informant

2. Interviewing an informant

3. Making an ethnographic record

4. Asking descriptive questions

5. Analyzing ethnographic interviews

6. Making a domain analysis

7. Asking structural questions

8. Making a taxonomic analysis

9. Asking contrast questions

10. Making a componential analysis

11. Discovering cultural themes

12. Writing the ethnography

    Thus, in Spradley's method there are four levels of data analysis: domain, taxonomic, componential, and theme. Domain analysis is the first level of analysis. Domains, which are units of cultural knowledge, are broad categories that encompass smaller categories. There is no preestablished number of domains to be uncovered in an ethnographic study. During this first level of data analysis, ethnographers identify relational patterns among terms in the domains that are used by members of the culture. 

    The ethnographer focuses on the cultural meaning of terms and symbols (objects and events) used in a culture, and their interrelationships. In taxonomic analysis, the second level of data analysis, ethnographers decide how many domains the data analysis will encompass. Will only one or two domains be analyzed in depth, or will a number of domains be studied less intensively? After making this decision, a taxonomy a system of classifying and organizing terms is developed to illustrate the internal organization of a domain and the relationship among the subcategories of the domain. 

    In componential analysis, multiple relationships among terms in the domains are examined. The ethnographer analyzes data for similarities and differences among cultural terms in a domain. Finally, in theme analysis, cultural themes are uncovered. Domains are connected in cultural themes, which help to provide a holistic view of the culture being studied. The discovery of cultural meaning is the outcome.

Analysis of Focus Group Data

    Focus group interviews yield rich and complex data that pose special analytical challenges. Indeed, there is little consensus about the analysis of focus group data, despite its use by researchers in several qualitative research traditions. Unlike data from individual interviews, focus group interviews are very difficult to transcribe, partly because there are often technical problems. For example, it is difficult to place microphones so that the voices of all group members are picked up with equal clarity, particularly because participants tend to speak at different volumes. 

    An additional issue is that, because of the group situation, it is inevitable that several participants will speak at once, making it impossible for transcriptionists to discern everything being said. A major controversy in the analysis of focus group data is whether the unit of analysis is the group or individual participants. Some writers (eg, Morrison Beedy , Côté -Arsenault, and Feinstein, 2001) maintain that the group is the proper unit of analysis. Analysis of group-level data involves a scrutiny of themes, interactions, and sequences within and between groups. 

    Others, however (eg, Carey and Smith, 1994; Kidd and Parshall, 2000), argue that analysis should occur at both the group level and the individual level. Those who insist that only group-level analysis is appropriate argue that what individuals say in focus groups cannot be treated as personal disclosures because they are inevitably influenced by the dynamics of the group. 

    However, even in personal interviews individual responses are shaped by social processes, and analysis of individual-level data (independent of group) is thought by some analysts to add important insights. Carey and Smith (1994) advocate a third level of analysis namely, the analysis of individual responses in relation to group context (eg, is a participant's view in accord with or in contrast to majority opinion, and how does that get expressed or suppressed?). 

    For those who wish to analyze data from India visual participants, it is essential to maintain information about what each person said a task that is impossible to do if researchers are relying solely on audiotapes. Videotapes, as supplements to audiotapes, are sometimes used to identify who said what in focus group sessions. 

    More frequently, however, researchers have several members of the research team in attendance at the sessions, whose job it is to take detailed field notes about the order of speakers and about significant nonverbal behavior, such as pounding or clenching of fists, crying, aggressive body language, and so on. Many focus group researchers agree, regardless of their position on the unit of analysis, on the benefit of certain methods of enhancing data quality and analytic rigor. 

    First, it is usually recommended that member checking occur in situ. That is, moderators develop a summary of major themes or viewpoints in real time, and present that summary to focus group participants at the end of the session for their feedback. Especially rich data often emerge from participants' reactions to those summaries. Second, post- session debriefings are critical. Team members who were present during the session meet immediately afterwards to discuss issues and themes that arose. 

aDuring these debriefings, which should be tape recorded, team members also share their views about group dynamics, such as coercive group members , censoring of controversial opinions, individual conformity to group viewpoints, and discrepancies between verbal and nonverbal behavior. 

    Transcription quality is especially important in focus group interviews: Emotional content as well as words must be faithfully recorded because participants are responding not only to the questions being posed but also to the experience of being in a group.Because of group dynamics, focus group analysts must be sensitive to both the thematic content of these interviews, and also to how, when, and why themes are developed. Some of the issues that could be central to focus group analysis are the following:

• Does an issue raised in a focus group constitute a theme or merely a strongly held viewpoint of one or two members?

• Do the same issues or themes arise in more than one group?

• If there are group differences, why might this be the case were participants different in background characteristics and experiences, or did group processes affect the discussions?

• Are some issues sufficiently salient that they are discussed not only in direct response to specific questions posed by the moderator, but also spontaneously emerge at multiple points in the session?

• Do group members find certain issues both interesting and important? Some focus group analysts, such as Kidd and Parshall (2000), use quantitative methods as adjuncts to their qualitative analysis.

    They conduct such analyzes as assessing similarities and differences between groups, determining coding frequencies to aid pattern detection, examining codes in relation to participant characteristics, and examining how much individual members contributed. They use such methods not so that interpretation can be based on frequencies, but so that they can better understand context and identify issues that require further critical scrutiny and interpretation. 

    Focus group data are sometimes analyzed according to the procedures of a formal research tradition, such as grounded theory. Typically, data from such studies are integrated after the fact. That is, quantitative data are analyzed statistically, qualitative data are analyzed through qualitative methods, and then researchers interpret overall patterns in light of findings from the two study components. There is, however, emerging interest in exploring alternative methods of analysis.

    One such approach is the construction of a meta-matrix that permits researchers to recognize important patterns and themes across data sources. This method, which can also be used to integrate multiple types of qualitative data (eg, interviews and observational field notes), involves the development of a two- or three-dimensional matrix that aligns variety of types of data (Miles & Huberman, 1994). Typically, one dimension involves a particular study participant. 

    Then, for each participant, a chart is constructed in which data from multiple data sources are entered, so that the analyst can see at a glance such information as scores on psychosocial scales (eg, scores on the Center for Epidemiological Studies Depression scale), comments from open-ended dialogue with participants (eg, verbatim narratives relating to depression), hospital record data (eg, physiologic information ), and the researchers' own reflective comments. 

    A third dimension can be added if there are multiple sources of data relating to multiple constructs (eg, depression, pain). The construction of such a meta-matrix allows researchers to explore such issues as whether statistical conclusions are supported by the qualitative data for individual study participants, and vice versa. Patterns of regularities, as well as anomalies, may be easier to see through detailed inspection of such matrices, and can allow for fuller exploration of all sources of data simultaneously.

Interpretation Of Qualitative Findings

    In qualitative studies, interpretation and analysis of the data occur virtually simultaneously. That is, researchers interpret the data as they categorize it, develop a thematic analysis, and integrate the themes into a unified whole. Efforts to validate the qualitative analysis are necessarily efforts to validate date interpretations as well. Thus, unlike quantitative analyses, the meaning of the data flows from qualitative analysis. 

    Nevertheless, prudent qualitative researchers hold their interpretations up for closer scrutin self-scrutiny as well as review by peers and outside reviewers. Even when researchers have undertaken member checks and peer debriefings, these procedures do not constitute proof that results and interpretations are credible. 

    For example, in member checks, many participants might be too political to disagree with researchers' interpretations, or they may become intrigued with a conceptualization that they themselves would never have developed on their own a conceptualization that is not necessarily accurate. Thus, for qualitative researchers as well as quantitative researchers, it is important to consider possible alternative explanations for the findings and to take into account methodologic or other limitations that could have affected study results. 

    In drawing conclusions, qualitative researchers should also consider the transferability of the findings . Although qualitative researchers rarely seek to make generalizations, they often strive to develop an understanding of how the study results can be usefully applied. The central question is: In what other types of settings and contexts could one expect the phenomena under study to be manifested in a similar fashion? 

    The implications of the findings of qualitative studies, as with quantitative ones, are often multidimensional. First, there are implications for further research: Should the study be replicated? Could the study be expanded (or circumscribed) in meaningful and productive ways? Do the results suggest that an important construct has been identified that merits the development of a more formal instrument? Does the emerging theory suggest hypotheses that could be tested through more controlled, quantitative active research? Second, do the findings have implications for nursing practice? 

    For example, could the health care needs of a subculture (eg, the homeless ) be identified and addressed more effectively as a result of the study? Finally, do the findings shed light on the fundamental processes that are incorporated into nursing theory?

Key Points

• Qualitative analysis is a challenging, labor-intensive activity, guided by a few standardized rules.

• Although there are no universal strategies, three prototypical analytical styles have been identified: (1) a template analysis style that involves the development of an analysis guide (template) to sort the data; (2) an editing analysis style that involves an interpretation of the data on which a categorization scheme is based; and (3) an immersion/crystallization style that is characterized by the analyst's total immersion in and reflection of text materials.

• Qualitative analysis typically involves four intellectual processes: comprehending, synthesizing, theorizing, and recontextualizing (exploration of the developed theory vis-à-vis its applicability to other settings or groups).

• In working with audiotaped data that must be transcribed, researchers should use transcription conventions and take steps to ensure that transcription errors are minimized and corrected.

• The first major step in analyzing qualitative data is to organize and index the materials for easy retrieval, typically by coding the content of the data according to a categorization scheme.

• Traditionally, researchers have organized their data by developing conceptual files, which are physical files in which coded excerpts of data relevant to specific categories are placed. Now, however, computer programs are widely used to perform basic indexing functions, and to facilitate data analysis.

• The actual analysis of data begins with a search for themes, which involves the discovery not only of commonalities across subjects, but also of natural variation and patterns in the data.

• The next analytical step usually involves a validation of the thematic analysis. Some researchers use quasi-statistics, which involves a tabulation of the frequency with which certain themes or relations are supported by the data.

• In a final analytical step, the analyst tries to weave the thematic strands together into an integrated picture of the phenomenon under investigation.

• Grounded theory uses the constant comparative method of data analysis

• One approach to grounded theory is the Glaser and Strauss method, in which there are two broad types of codes: substantive codes (in which the empirical substance of the topic is conceptualized) and theoretical codes (in which the relationships among the substantive codes are conceptualized).

• Substantive coding involves open coding to capture what is going on in the data, and then selective coding, in which only variables relating to a core category are coded. The core category, a behavior pattern that has relevance for participants, is sometimes a basic social process (BSP) that involves an evolutionary process of coping or adaptation.

• In the Glaser and Strauss method, open codes begin with level I (in vivo) codes, which are collapsed into a higher level of abstraction in level II codes. Level II codes are then used to formulate level III codes, which are theoretical constructs.

• Through constant comparison, the researcher compares concepts emerging from the data with similar concepts from existing theory or research to determine which parts have emergent fit with the theory being generated.

• The Strauss and Corbin method is an alternative grounded theory method whose outcome is a fully preconceived conceptual description. This approach to grounded theory analysis involves three types of coding: open (in which categories are generated), axial coding (where categories are linked with subcategories), and selective (in which the findings are integrated and refined).

• There are numerous different approaches to phenomenological analysis, including the descriptive methods of Colaizzi , Giorgi, and Van Kaam , in which the goal is to find common patterns of experiences shared by particular instances.

• In Van Manen's approach, which involves efforts to grasp the essential meaning of the experience being studied, researchers search for themes, using either a holistic approach (viewing text as a whole); a selective approach (pulling out key statements and phrases); or a detailed approach (analyzing every sentence).

• One approach to analyzing ethnographic data is Spradley's method, which involves four levels of data analysis: domain analysis (identifying categories); taxonomic analysis (selecting key domains and constructing taxonomies); componential analysis (comparing and contrasting terms in a domain); and theme analysis (to uncover cultural themes).

• Some researchers identify neither a specific approach nor a specific research tradition; rather, they might say that they used qualitative content analysis as their analytic method.

• A major controversy in the analysis of focus group data is whether the unit of analysis is the group or individual participants.

• One approach to analyzing triangulated data from multiple sources (eg, qualitative and quantitative data) is the construction of a meta-matrix that arrays such data in a table so the analyst can inspect patterns and themes across data sources.