I used a semi-structured interview form to collect data from teachers, students, and parents in our broader project that aimed for understanding and comparing the values education processes in a succesful public and private middle(secondary) school as a multiple case study (public and private cases). I am trying to analyze the data with content analysis method. I have encountered that some respondents had responded in a more amotivated manner, and provided data that is somewhat less complex from others. They have lower magnitude, less codes assigned, etc. I think this may be relevant to report. Although I am OK with reporting just what I said as a discussion theme; I am curious about whether there are established methods (or perhaps statistics) useful in documenting such differences.
Dear Çınar,
I doubt whether such methods/metrics exist as you are carrying out an interpretive exercise. Clearly differences in motivation might lead to different degrees of richness in the responses and this could measured by the number of categories identified or the shorter length of time to reach data saturation.
If some of your respondents were not very motivated this might also have affected the validity of their responses.
If you wish to be more scientific in measuring such differences it might require your data being analyzed by several researchers independently. Perhaps this might be a useful type of study to look for in the literature?
I agree with Peter Samuels that it would be difficult to develop code-able ways to capture this difference. In particular, how would you decide what was a more "complex" or "motivated statement?
The other question is what you would do with such data. If the issue is whether these participants say something different, that is one thing. But if they basically have similar content but different ways of saying the same thing, then what difference does it make?
Yes I agree with the comments - you need to keep true to your research question/s and ask yourself the purpose of coding - what is the end point. What are you wanting to report on here?
From what I can understand in your inquiry, I think you are making a point more proper of quantitative research than qualitative. The point should be how much is my data telling me about a relevant topic, and that "how much" is entirely up to your interpretation. I agree with the comments above that the issues are more of validity and quality of the findings: in qualitative research, the more depth and facets you have, the better. If there are noticeable differences between your participants in this sense, then yes you should recognize it and be open about it. Trying to recognize the reasons behind the differences, potential biases, etc. is a very good starting point, some might even say it is the "established method" you are looking for.
Thanks for all the responses. I will take the points seriously. I realize that I have a strong urge to quantify/reduce perhaps partly because of my previous learning history on quantitative research.
I agree with what has previously been said but would like to add something important. The persons you have interviewed have not used "codes" when answering your questions in the interview. They have simply tried to answer your question using their own words. It is you who creates and assigns codes to the interviewiees words in your analysis of the collected data.
I also think that the thing you have discovered , differences between persons way of expressing themselves is an intereseting thing to discuss in your article, and not least the consequences of this differences.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude