Ch 13. Qualitative Data Analysis Flashcards
What are general strategies of qualitative data analysis (2)
- Analytic induction
- Grounded theory
Basic operations in qualitative data analysis (2)
Coding
Narrative analysis
A general research question is devised, some data are gathered, and a hypothesis is proposed
Analytic Induction
Qualitative analysis is an iterative process where the
analysis starts after some data have been collected & further data are collected based on that analysis
Because all cases must be explained, hypotheses generated may be too broad to be useful
Usually no guidelines on how many cases must be reviewed before the validity of the hypothesis is accepted
Difficulties with analytic induction
“Theory that was derived from data, systematically gathered and analyzed through the research process” (Strauss & Corbin, 1998, p. 12)
Most widely used framework for analyzing qualitative data
Grounded theory includes processes
- Coding
- Constant comparison (of data and concepts)
The indicators are repeatedly compared for concepts/categories - Theoretical saturation
A point in time when nothing new is being learned
What is the first step in interpreting data and developing theory?
coding
What are three types of coding?
- open
- axial
- selective
Open Coding
identifies initial concepts that will be categorized later
Axial Coding
reviews data for linkages, re-organized according to connections
Selective Data
selecting core categories, validating the relationships, and identifying gaps that need to be filed in
building blocks of theory
concepts
encompass two or more concepts
categories
attributes of a category
properties
initial hunches
hypothesis
observed patterns are related to each other and a theory is developed to explain the connections in that setting
Substantive theory
theory formulated at a higher level; requires data collection in different settings; applicable to a variety of settings
Formal theory
Iterative Process
a cyclical approach where researchers repeat steps or phases of their research in order to refine, improve, or validate their findings.
- The researcher begins with a general research question
- Relevant people and/or incidents are theoretically sampled
- Relevant data are collected
- Data are coded, which may, at the level of open coding, generate concepts (step 4a)
There is a constant movement backward and forward among the first four steps, so that early coding suggests a need for new data, which leads to theoretical sampling, and so on
- Through constant comparison of indicators and concepts categories are generated (step 5a). It’s crucial to ensure a fit between indicators and concepts
- Categories become saturated during the coding process
- Relationships between categories are explored in such a way that hypotheses about connections between categories emerge (step 7a)
8/9. Further data are collected via theoretical sampling
10/11 The collection of data is likely to be governed by the theoretical saturation principle and the testing of the emerging hypotheses (step 11), which leads to specification of substantive theory (step 11a)
- The substantive theory may eventually be explored using grounded theory processes in a different setting from the one in which it was generated
In this way a formal theory (step 12a)—relating to more abstract categories not specifically examined in the research— can be generated
Steps in Qualitative Analysis
- Provide reminders about what is meant by the terms used
- Are important in conceptual and theoretical comparison between cases
- Aid in conceptual and theoretical reflection, creation of concepts and categories
Memos in Grounded Theory
Differences between concepts and categories may be vague
Observation and data gathering may not be as “theory neutral” as claimed
Practical difficulties
May not result in theory (especially formal theory)
Coding may result in fragmentation, loss of narrative flow
Criticisms of Grounded Theory
- code and transcribe asap
- read through the data before considering any interpretation
- read through the data again
- do not be concerned with producing too many codes, this is normal at the beginning
- review the codes to consider associations, redundancy, & relationships to existing concepts
- consider general theoretical ideas regarding codes & data
- keep coding in perspective
Steps and Considerations in Coding
Problems with Coding (5)
- Risk of losing the context
- Fragmentation of data
- Basic Coding
- Deeper awareness of the context in the text
- Exploring broader analytic themes (refer to slides for full description)