Levels of Analysis

Levels of Analysis

1. CATEGORY CONSTRUCTION

   • The categories are concepts indicated by the data, and not the data itself
     • Units of data (bits of information)
       • Sorted into groupings that have something in common
       • Any meaningful or potentially meaningful segment of data. Can be a small word of the participants, or large pages of field notes
         • First: Heuristic, it should reveal information relevant to the study
           • Second: the unit should be the smallest piece of information about something that can stand by itself

2. THE STEP-BY-STEP PROCESS

   • Category construction
     • Reading of the first interview transcript, field notes, documents and side notes.
       • Join the notes and comments that go together. Check the second set of notes and compare them
   • Naming the categories
     • Come from three sources: the researcher, the participants and the literature.
       • Categories should reflect the purpose of the research, be exhaustive, mutually exclusive, sensitizing and conceptually congruent
   • How many categories?
     • The fewer categories the greater ease.
       • Four guidelines: what people mention with frequency indicates an important dimension, the audience may determine what is important, some will stand out because of their uniqueness, they reveal areas of inquiry
   • System for placing data into categories
     • Four basic strategies for organizing all the data in preparation for further analysis
       • Index cards, file folders, information retreival cards, and computer programs

3. DEVELOPING THEORY

   • A third level of analysis involves making inferences, developing models, or generating theory
     • Properties
       • Concepts that describe a category , properties are not examples of a category but dimmensions of it
         • The development of categories, properties, and tentative hypotheses through the constant comparative method is a process whereby the data gradually evolve into a core of emerging theory
     • Hypotheses
       • Are the suggested links between categories and properties

4. DATA ANALYSIS IN CASE STUDIES

   • Intensive, holistic description and analysis of a single, bounded unit, this range of data sources may present disparate, incompatible, even apparently contradictory information
     • Within case analysis
       • Each case is first treated as a comprehensive case in and of itself. Data are gathered so the researcher can learn as much about the contextual variables as possible that might have a bearing on the case
     • Cross-case analysis
       • The researcher attempts to build a general explanation that fits each of the individual cases . Attempts to see processes and outcomes taht occur accross many cases, to understand how thay are qualified by local conditions, and thus developed more sophisticated descriptions and more powerful explanations.

Images and files in the map