Questions:
- Understanding exactly how many interviews I really need to conduct to achieve my stated objectives.
- Understand better the issue of generalizability within the qualitative research context
- How open ended or close ended should the interview guide be?
Interesting study! I can imagine something like a detailed concept map outlining the various factors affecting uptake of ART among HIV/TB co-infected patients.
I like that you plan to interview a wide range of stakeholders, and think that will make your data very rich. What would you think about adding a few patients/clients? They may highlight some barriers that providers may be unaware of, though I think sampling patients at each facility would be prohibitive, so maybe identifying one or two facilities where you could interview 3-4 patients would be interesting.
And related to sample size, maybe we'll talk about that more in class, but I think it depends on your analytic approach, as well as how common your themes are (e.g. if you are planning on collecting data until "thematic saturation")? A few weeks ago I found an article that was looking specifically at sample size in qualitative studies, and it came up with an estimate of 30-50 interviews to achieve thematic saturation for grounded theory. That being said, because you are interviewing a wide variety of types of people, I imagine you will uncover many themes from different perspectives, so I suspect you'll have to err on the side of more interviews to capture most of the important barriers. A strength, though, is that barriers that come up from multiple perspectives are likely to be significant, and may lead to a bigger impact if those barriers are addressed.
Monika, the importance of this study is clear and I think in-depth interviews makes sense as a format. I wonder about having so many stakeholders, does it make sense to have more subjects? I wonder if you will be able to reach thematic saturation with so many different stakeholders? I also wonder if you would learn anything valuable by observing clinic visits with HIV-TB infected patients to add an observation component to your study.
Hi Monika and great project. This is A LOT of interviews. Often a study within one group will reach thematic saturation after about 20 interviews. If you are interviewing multiple groups, it may be reasonable to do 10-20 per group, but keep in mind this will be a lot of data, and you might want to limit your study initially to the perspectives of one group. There will also be transcription and translation issues to think about - you will be publishing in English? So you will have to consider will bilingual coders do the analysis as well?
re: generalizability, i'd say its similar to quantitative research
the interview guide should be very open-ended questions - but be clear about your research question to define the scope to participants
