University of Applied Sciences, Fulda

Methods of Analysis

Qualitative data analysis enables the researcher to analyse data by using one particular method which rests on a specific theoretical foundation or by combining different modes of inquiries. In fact, the aim, nature and method of the inquiry are decisive in choosing an analysis method. The main aim, however, is to first categorise the data in a meaningful way and then try to come up with sensible interpretations on the basis of these categories. Here the theoretical background of the method is elaborated on to clarify the rationale behind the choice.
Guba (1978) suggests that in focusing on the analysis of qualitative data an evaluator must deal first with the problem of convergence. The problem of convergence is figuring out what things fit together. This leads to a classification system for the data.
Guba suggests several steps for converting the data into systematic categories of analysis. The evaluator-analyst begins by looking for “recurring regularities” in the data. These regularities represent patterns that can be sorted into categories. Categories should then be judged by two criteria: internal homogeneity and external heterogeneity. The first criterion concerns the extent to which the data that are placed in a certain category hold together or “dovetail” in a meaningful way. The second criterion concerns the extent to which differences among categories are distinct and clear: “The existence of a large number of unassignable or overlapping data items is good evidence of some basic fault in the category system” (Guba 1978: 53). The naturalistic evaluator then works back and forth between the data and the classification system to verify the meaningfulness and accuracy of the categories and the placement of data in categories. When several different classification systems have been developed, some priorities must be established to determine which category systems are more important than others. Prioritising is done according to the salience, credibility, uniqueness, heuristic value, feasibility, special interests, and tangibility of the classification schemes. Finally, the category system or set of categories is tested for completeness.

1. The set should have internal and external plausibility, a property that might be termed ‘integratability’. Viewed internally, the individual categories should appear to be consistent; viewed externally, the set of categories should seem to comprise a whole picture . . .

2. The set should be reasonably inclusive of the data and information that do exist. This feature is partly tested by the absence of unassignable cases, but can be further tested by reference to the problem which the inquirer is investigating or by the mandate given the evaluator by his client/sponsor. If the set of categories did not appear to be sufficient, on logical grounds, to cover the facets of the problem or mandate, the set is probably incomplete.
3. The set should be reproducible by another competent judge. . . . The second observer ought to be able to verify that a) the categories make sense in view of the data which are available, and b) the data have been appropriately arranged in the category system . . .

4. 
   The set should be credible to the persons who provided the information which the set is presumed to assimilate. . . . Who is in
   

a better position to judge whether the categories appropriately reflect their issues and concerns than the people themselves? [Guba 1978: 56-57]

The second problem Guba speaks about is divergence. By this he means that the analyst must deal with how to “flesh out” the patterns or categories. He suggests that this is done by a process of extension (building on items of information already known), bridging (making connections among different items), and surfacing (proposing new information that ought to fit and then verifying its existence). The analyst closes the process when sources of information have been exhausted, when sets of categories have been saturated so that new sources lead to redundancy, when clear regularities have emerged that feel integrated, and when the analysis begins to “overextend” beyond the issues and concerns guiding the analysis.

These processes are far from being rigid or mechanical. The analyst should take the technical elements into consideration and at the same time be creative in her/his analysis. Although Guba’s model is not the only qualitative method of analysing data, other models follow more or less the same procedure. Due to its limitations in the collection of data, the present analysis could not restrict itself to one procedure only, but, of course, the writer has followed the general path.
Phenomenological Analysis
The theoretical orientation of this method was discussed in the Review of Literature chapter. Phenomenological analysis framework is one of the most general forms of analysing the data corpus. The following explanations are based on the work of Douglass and Moustakas, [1984]; Moustakas, [1990] who have developed a framework for this type of analysis and are quoted in Quinn Patton[1990].
The first step in phenomenological analysis is that of Epoché. During this phase the researcher looks inward to become aware of personal bias and to eliminate personal involvement with the subject material. During the Epoché phase it is essential that the researcher eliminate or at least gain clarity about preconceptions. Rigor is reinforced by a phenomenological attitude shift accomplished through Epoché.
The researcher examines the phenomenon by attaining an attitudinal shift. This shift is known as the phenomenological attitude. This attitude consists of a different way of looking at the investigated experience. By moving beyond the natural attitude or the more prosaic way phenomena are imbued with meaning, experience gains a deeper meaning. This takes place by gaining access to the constituent elements of the phenomenon and leads to a description of the unique qualities and components that make this phenomenon what it is. In attaining this shift to the phenomenological attitude, Epoché is a primary and necessary phenomenological procedure.
Epoché is a process that the researcher engages in to remove, or at least become aware of prejudices, viewpoints or assumptions regarding the phenomenon under investigation. Epoché helps enable the researcher to investigate the phenomenon from a fresh and open view point without prejudgment or imposing meaning too soon. This suspension of judgement is critical in phenomenological investigation and requires the setting aside of the researcher’s personal viewpoint in order to see the experience for itself. [Katz, 1987: 36-37]

Following Epoché, the second step is phenomenological reduction. In this analytical process, the researcher brackets out the world and presuppositions to identify the data in a pure form, uncontaminated by extraneous intrusions:

In bracketing, the researcher holds the phenomenon up for serious inspection. It is taken out of the world where it occurs. It is taken apart and dissected. Its elements and essential structures are uncovered, defined and analysed. . . It is treated as a text or a document; that is, an instance of the phenomenon that is being studied.... Bracketing involves the following steps:
1. Locate within the personal experience, or self-story, key phrases and statements that speak directly to the phenomenon in question.

2. 
   Interpret the meanings of these phrases, as an informed reader.
   
3. 
   Obtain the subject’s interpretations of these phrases, if possible.
   
4. 
   Inspect these meanings for what they reveal about the essential, recurring features of the phenomenon being studied.
   
5. 
   Offer tentative statement, or definition, of the phenomenon in terms of the essential recurring features identified in step 4. [Denzin, 1989:55-56]
   

Once the data are bracketed, all aspects of the data are actually treated in the same way. The researcher has some set of data which he can treat equally. The next step is to put them into meaningful categories. The researcher then goes through the data over and over again to eliminate the repeated parts and refine the categories. S/he tries to observe and investigate the phenomenon from all possible angles.

The next step is to abstract the experiences that provide the content and illustrate them as they are. Each theme is dealt with separately and the inquirer describes what is actually not evident, that is, the feelings or deeper values in the background.
The final step in phenomenological analysis is the development of a “structural synthesis”. This synthesis will contain the “bones” of the experience. The true meanings of the experience for the individual will be described. In the structural synthesis, the researcher looks beneath the affect inherent in the experience to deeper meanings for the individual. This reveals the essence of the phenomenon. [Quinn Patton, 1990: 409]

The analysis of the collected data will be discussed in detail in next chapter.