Hello
I am working on the following: I have 65 exsiting definitions about a concept. From these 65 definition I try to distill a representative definition via coding procedures. Thus, I code each definition to find patterns.
Now I have all definitions coded. And I can see how many times a code is used in the exsisting definitions. For example: code X is present in 20 of the 65 definitions.
As I now want to develop a representative definition (and not one which contain all codes) I would set a cut-off-level. A cut-off-level in the sense of: only codes which are more then e.g. 20 times present in the exsiting definitions will be included in my representative definition.
Now I was wondering, is there a way to determine the cut-off-level statistically?
Hi David Redaschi,
I might ask: do you believe in qualitative at all? I think that "determining cut-off level statistically" is about quantitative... You should find out what "representative" means in qualitative sense. In short "representative" can be defined in terms of conceptual variation found in the data. I think that 65 definitions should be reduced towards few information intense ones, kind of conceptual cores.
The article "Analysing qualitative data" (2000) by Catherine Pope, Sue Ziebland,Nicholas Mays could be an informative source. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1117368/
In a qualitative study your attention focuses on the properties (attributes) present in the data, and the importance and role of each attribute is not given to you by an external set of criteria. Instead, it comes from your understanding you develop during the process.
I hope this helps more than it confuses you....
One way to establish a cut-off with a frequency distribution is to look for "gap" in the data. In factor analysis, this is known as a scree test.
Use the D.Gioia approach: "You gotta get lost before you can get found". Use as many codes you need in the first step during coding...see also: Dennis Gioia, A renaissance self: Prompting personal and professional revitalization. In Stablein, Ralph E and Frost, Peter J, Renewing research practice, Stanford University Press. 2004, σελ. 97-114.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Coding