I'm doing an insider ethnography and using mixed method approach among disaster survivors resilience. As I belong to the same population pool as that of participants, it is said that ethnography should begin with qualitative but based on my prior observation and experience of the incident, is it acceptable to use quantitative survey first and elaborate the concepts and constructs with qualitative method later with the same participants and settings later again?
In mixed methods, the most common reason for starting with a qualitative methods is a "sequential exploratory design." The most frequent use for this design is to create quantitative measures, but it is also useful for generating hypotheses. This is often summarized as qual --> QUANT
Alternatively, starting with a quantitative method is often termed a "sequential explanatory design" and the most frequent use for this design is to help understand the results from a quantitative method. This is often summarized as QUANT --> qual
A less common design is to begin with a major qualitative study and then extend that with a smaller quantitative study. In that case, you might generate hypotheses with the qualitative study and then do a preliminary test with the quantitative study. This is can be summarized as QUAL --> quant, and here is a link to a paper I wrote on this third option:
Article From Themes to Hypotheses: Following Up With Quantitative Methods
I think, in your situation, I would first do my own ethnography first so that I know which indicators /variables are important for my own ethnography. Once I know why some indicators than others are important than others for me, I can use these for survey methods to examining these on others. Then, you can make your arguments that what works for you, these may or may not work for others. If you do survey first, your ethnography may be impacted by your survey. You may lose your ethnography. it's my suggestion. It may not work for you.
I agree with Ranjan. Your design may also depend on your research question(s). Sometimes different methods are used to answer different questions related to the same topic... cheers.
Thank you @ David, Ranjan and Xavier for your insights. I'm convinced which method to use first so far my research question is concerned. It does make a complete sense why ethnography needs to be done first and the survey. But for survey, I am planning to use a pre-validated tool. Should a survey questions develop from the variables generated from ethnography or can implement a separate tool?
I think it would be best if your hypotheses and research questions were generated from the ethnography, but matched to pre-validated items in the survey.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey