Our research is about human capital in professional accounting firms. The respondents will answer a structure questionnaire of about 70 questions. However, we think that selecting a few respondents to explain deeply their answers will help put context to the quantititative statistics. The question is "What the optimal number of these in-depth interviews" given the initial sample was about 100
Its easy to answer: 6 may be enough.
Questions / issues for futher thinking:
- Do you already have the basic results of the 100? Or which is going to be first in the data collection?
- - If you are having/ going to have the questionnaire results first, its WISE to use the results for the interview selection.
-- You study questions + the (possible) questionnaire results will tell you wheather 6 is enough: ---> from where/which context do you need futher/deeper data?
-- If you are finilizing you Research plan, you can plan to make 5-10 in-dept interviews, because its always also the question the resiources which you can use for the data collection ans analyses of the qualitative data.
From the perspective of grounded theory, the optimum "number" of interviews is determined by the content of the interviews themselves. Once you are hearing nothing new in the interviews, assuming you have intentionally selected the subjects to include diverse perspectives, then there is no strong reason to continue to seek out new interviews.
It is simple. Once you reach to the point of saturation, meaning when you hear nothing new, then you will come to know that it is enough. According to my personal experience, 6 interviews are usually NOT enough. You have to think to at least double the number.
A lot depends on the research design you are using -- in particular, whether you are doing a mixed methods study, where you want to use your follow-up qualitative interviews to understand your quantitative results. This is often known as a sequential explanatory design, which is summarized as QUANT --> qual.
In that case, you would want to obtain your quantitative results before defining your qualitative sample. The approach is to do a "purposive" qualitative sample that will help explain your quantitative results. For example, you might compare 3 informants who are high on some characteristic and 3 who are low on the same characteristic. Or, you might select a set of outliers to learn why they don''t match the larger sample.
Regardless of your design, you should pay attention to Ibrahim Khalid Rayes' emphasis on saturation. Otherwise, you might do just 6 interviews and discover that you have a wide range of responses, without enough overlap to help you make sense of the data.
Here are two other current RG questions that address your topic:
https://www.researchgate.net/post/How_many_interviews_are_enough_in_qualitative_research_case_study_studies
https://www.researchgate.net/post/what_are_the_research_method_for_small_samples
One thing is clear the number of interviews cannot be fixed as it can be done in a quantitative study. Two of the key concepts which are relevant for solving your problem are saturation and purposive sampling. Saturation goes back to the fathers of grounded theory. It remains difficult to handle, though it is an essential criterion. Concerning purposive sampling I would refer to the work of Janice M. Morse.
Normally a qualitative research namely formative research precedes quantitative research which gives us detailed information about the area of inquiry and from there we can pick the aspects we want to focus on in the quantitative research. In qualitative interviews you stop when the information gets saturated. Here I presume the qualitative interviews are desired to provide context to the quantitative data. I am of the opinion that after the data is collected then depending upon the results for which we need to provide context We can choose representative persons using purposive sampling, for in-depth inquiry depending upon the variance one sees in the study population. The results from quantitative data will guide us as to how different the population was. If it was very similar small number of interviews even 6 should be good enough, else one can conduct about 8-10 in-depth interviews for this sample size.
Thanks for the many useful comments and ideas. We shall select the interview sample purposively and indeed will be greatly informed by the results of the quantitive study. We shall also borrow the saturation idea because the interviewees may at some point be saying the same thing.
Saturation it's very important for qualitatives studies
I share with you a good paper of QMR :
http://www.emeraldinsight.com/doi/full/10.1108/QMR-06-2016-0053
You can also read qualitative books
Belk R.W . (2006), Handbook of qualitative Research : Methods in marketing, 609 pages. Denzin N.K, Lincoln Y.S & al. (2000), Handbook Qualitative Research, Sage, 2nd edition.
Flick, U. (2009). An introduction to qualitative research. Sage.
Neuman, L. W. (2014). Social research methods: Qualitative and quantitative approaches, 7th Edition, Pearson Education, 599 pages, .
Tracy, S. J. (2012). Qualitative research methods: Collecting evidence, crafting analysis, communicating impact. John Wiley & Sons.
Van den Hoonaard, D. K. (2012). Qualitative research in action: A Canadian primer. Oxford University Press Canada.
Good luck
In depth interview...6 may not enough. Interview until reaching saturation point. Stop when no more new information from the individuals. Most of the time around 10 - 12 individuals.
you have already done a quantitative study i suggest you sample 1/10 of your initial population for the second part.
You don't start with a fixed number in mind. You begin your interviews and continue until no new information is forthcoming, i.e you have reached data saturation. It can be after 5 interviews or not even after 8.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India