We as management researcher write case study, we need to apply both-quantitative and qualitative- research methodologies. I am now more interested in applying qualitative approach in case studies....
This is a very good question. I usually use existing literature to help me to identify emerging themes. Specifically, I usually develop a preliminary theoretical framework based on a literature review. There are two approaches: (1) identify closely relevant research studies and use their research findings; (2) identify important themes as reported in the existing literature.
What Lihong Zhou describes is often referred to as a "deductive" approach to qualitative analysis because it begins with a predetermined theoretical framework. In contrast, and "inductive" approach would build its codebook from the data itself. A third, "hybrid" alternative is to begin with a deductively produced codebook and modify it through inductive reading of the data. You see more about these alternatives in Hsieh et al. (2005) Three Forms of Content Analysis.
You might also take a look at Braun & Clarke (2006) on Thematic Analysis, which is the most widely cited source in this area.
As far as creating cases goes, I am not aware of any sources on how to convert a set of themes into case study write-up.
I do qualitative research and publications, using data as necessary. Both work well together, but you might think of the two in terms of an airplane. The quantitative analyst often relies on correlations and flies over a country noting valleys and mountains and geographic patterns A qualitative analyst parachutes out to take a closer look and talks to people about living patterns. Both are useful.
Emerging themes might fit better with the inductive approach - as Professor Morgan said, which devotes to letting your data speak for themselves, rather than predetermining a framework and forcing your data into it.
To make themes emerge, we need to engage deeply with our data. How to engage depends on the methodological approach/strategy you use. For instance, grounded theory method (Corbin and Strauss 1990), Case study (Merriam 1998; Stake 1995; Yin 2013); Thematic Analysis (Braun & Clarke 2006), and so forth. Miles and Huberman (1994) provide a very comprehensive guide on how to conduct qualitative data analysis.
In general, the following patterns are practiced to obtain emerging themes:
Read, re-read, and re-read (or listen, re-listen) to get familiar with data
Code data into categories, sub-categories, and themes
Write memos/notes to answer 'questions' emerging from codes
Develop diagrams to connect segments (themes/categories/sub-categories) of data in relation to certain topics/themes
Develop matrixes to compare, contrast between groups/cases in your data
Develop a storyline to connect all themes appearing
Compare and contrast with themes existing in literature
Develop an overall theoretical framework to answer your research questions
I will strongly suggest the use of Strauss and Corbin's axial and selective coding techniques associated with their grounded theory method. These will help you filter conceptual and thematic parameters that form patterns from your interviews making it easy for you to prioritize and analyze results.
Hopefully the following sources could further help:
Cruzes, D. S., Dybå, T., Runeson, P. and Höst, M. (2015) Case studies synthesis: a thematic, cross-case, and narrative synthesis worked example, Empirical Software Engineering, 20, 6, pp. 1634-1665.
DeSantis, L. and Ugarriza, D. N. (2000) The concept of theme as used in qualitative nursing research, Western Journal of Nursing Research, 22, 3, pp. 351-372.
Hancock, D. R. and Algozzine, B. (2006) Doing case study research: A practical guide for beginning researchers. New York, NY: Teachers College Press.
i think the approach under qualitative research matters alot. eg ...grounded theory , ethnography & case study etc. then we should develop & design the emerging areas/themes.
I prefer Grounded Theory is a good way for qualitative data with its various types of coding techniques. Also you can use NUD.IST software for coding your data.