HI..
i'm sorry if i posted this on a wrong thread..I really need an advice on this. I have 30 interviews data which I have trnscribed and I want to analyse it.

My problem is, how do I do that? I know that I can code it into themes based on what I found from teh literature...but i also found out that I could do confirmatory factor analysis (CFA).. I've read some books about it but most of them dit not discuss on open ended data but more towards scale data (e,g: likert scale). Does anyone have an idea on how should i do CFA on open ended data?

Thanks!

Hi trapped,

I dont think that you can do CFA on interview data. But I could be wrong as this is not my expertise. I think that you need questionnaire items to do CFA. I guess the question you ask yourself, is does CFA help you answer your research question and contribute to your findings. If no, then forget about it. If yes, pursue it further.

A method of pursuing it further is to find an author who has done it, and email them. They will probably be happy that you are doing this and give references or advice. Hope that helps.

Potatoes.:-)

Hi Trapped

this website might be worth checking out - it gives good overview of lots of qualitative modes of analysis. Some very useful video links as well.
http://onlineqda.hud.ac.uk/

HI Potatoes and Ady,

ok...i think i found the answer..CFA is for quantitative not qualitative....i thought I could actually use it for qualitative.
Thanks for the reply!

:)

======= Date Modified 15 Feb 2012 22:56:42 =======
Trapped

With a sample as large as 30 transcripts of rich data you may like to consider a methodology such as Template Analysis. This appears to have advantages in being a very pragmatic approach. I take if from your first post that you are not coming at this from a particular epistemological stance.

See http://www2.hud.ac.uk/hhs/research/template_analysis/ for a solid overview of TA and some practical tips on looking at data and writing up.

This article has some useful information in about identifying themes in qualitative data interviews:
http://nersp.osg.ufl.edu/~ufruss/documents/ryan%20and%20bernard%20themes.%20FM%2015(1).pdf

HI Everyone..thanks for the reply..really appreciate it!

I'm now in the midst of identifying themes for my data..i'm just wondering on the process of the analysis...do I need to grouped them into BROAD theme first and then identify the SUB THEME? or do I need to Identify the SUB THEME first and then geouped them into teh BROAD theme?

Which one is the first step? DO you have any reference to support the procedure?

Thanks! :D

I tried it with doing the broad themes first. I noticed that when I started with the sub-themes, that the sub-themes did not fit as nice and neat packages in the broad theme as I had planned. This took place for a couple of iterations of coding until I as able to see how the sub-themes worked together and adjusted the broad themes accordingly. Point being: don't get locked into one way of doing it and don't just do the coding once. Every good reference I read talked about the importance of repetition in coding in order to be sure your codes are useful and used consistently. This repetition will help guide you towards using the data to answer your question(s). Good luck and have fun exploring the data, that is uncharted territory!