Can anyone suggest, apart from Triangulation, how can I merge/support primary data through secondary? Further any suggestions to use any specific strategy while analyzing it, in-order to establish rigorousness. I am doing multiple case-studies: using primary data from my interview, secondary data through content analysis. Kindly suggest some.
In my dissertation, I did not primarily focus on case studies (unless you view a whole market as a case), but used a combination of quantitative data, media coverage, some interviews and visual analysis of websites. you might get some ideas by looking at it (chapter 2 and appendix). It's downloadable from my 'old' profile 'Fredrik Augustsson' (I've changed name); called 'They Did IT'. apart from that, I'd suggest looking at work connected to critical realism which tends to favour case studies (e.g. ANdrew Sayer and a recent methods book by paul Edwards, Steve Vincent and some others.
Hi! You can apply content analysis technique also for primary data. How? First you have to identify codes in the main content (e.g. information coming from in-depth interviews); and create categories from the identified codes. Then, you proceed to analysis content from primary and secondary data using pattern matching technique and then triangulation. Other way to analysis primary data coming from cases studies using content analysis is to apply frequency count technique for qualitative methodology.
Use your conceptual framework to write the cases, ie, each case has the same conceptual framework. Then imagine the cases are lined up alongside one another and "look through" them focussing each time on the same conceptual category from each case. Then you analyse the multiple/across the cases looking at similarities and differences.
The scientific rigour in mixed-method and qualitative research is all about clear research design and transparent procedures you follow. I would suggest following:
1) Think carefully about research design. Maybe a mixed method approach could be useful help. A good book about this is: Creswell, J.W. & Clark, V.L. (2010). Designing and Conducting Mixed Methods Research.
2) If you go with case study design use same conceptual framework for each case study (as suggested by Tom Maxwell above) and be clear about the procedure of selecting and developing case studies. Very helpful and down to earth book about it is: Dul & Hak 2008 - Case Study Methodology in Business Research. Although coming from business background lots of ideas are universal and well applied to other social research.
3) Think about elements and patterns that you will be looking for across different sources. These are "codes". Depending of the type and objective of your work you can apply different coding strategies (Karen Orengo described above one of them). Each of them have quite simple but very transparent procedure. If you follow the procedure you establish scientific rigour. Definatelly the best source on coding is second edition of: Saldana, J. (2012). The Coding Manual for Qualitative Researchers.
And one more technical thing - if you have lots of qual data from different sources you can use special software for your data management and analysis (and especially coding). I work on Maxqda (good for qual-quan combination). Other are: Nvivo (good for antropological work) and Atlas (good for visualizing logical reasoning across documents).
Good luck with your research!
@ Tom Maxwell,
Thank you very much for your suggestions, your idea is quite easy to identify because of the pictorial characteristics, but one thing that is confusing me if i generate 3 conceptual frame-work, basically each term will be the similar or else there is a possibility that i can include ......issues and challenges on each one, because it's always different and then i can distinguish.
Is that what you mean sir...!
If its possible kindly provide me the "References", it would be more clearer to understand your logic
Regards,
Karun............!
Hi Karun,
You need to select a model based on which you convert qualitative data into quantitative format. There are many studies to translate online transcripts. Bale (1950) was first to start qualitative analysis and you can also search for Chou(2002) for QDA model.
Karol, your references are very helpful.
In my last email I forgot to mention a good paper about when it is useful to apply counting in qualitative research. See: Hannah and Lautsch (2011). Counting in Qualitative Research: Why to Conduct it, When to Avoid it, and When to Closet it, Journal of Management Inquiry, 20(1) 14–22.
In every study your objectives and research questions should guide you. And of course your conceptual framework. I will suggest you refer back to your obj and RQ plus conceptual framework and let these guide you in looking for similarities and differences across the case studies.
Everyone here has offered excellent suggestions. I want to reiterate a few major points: Think hard about the rationale for your cases. Are they most similar with different outcomes, or most different with the same outcome? I have not yet used NVivo, but some qualitative researchers like it for analyzing interviews. When I wrote my first book, I used mixed methods--trends from a new dataset and interviews. I transcribed 70 hours of interviews. In those days, there was no software to help me with this. After I transcribed them, I used large accounting sheets with the variables from my hypotheses across the top, respondents on the side, and in the spaces, wrote responses that were good indicators of the variables. I gave the respondents different names for confidentiality and IRB regs at the university, in this case the University of Washington. However, on another sheet, I recorded the real names with the created names. I then filled in the backstory that would organize the date according to characteristics of respondents and in what context I interviewed them. I hope this helps.
We use Atlas.ti software to organize data generated from different cases using varities of methods.
