Q: How many samples (numbers) should I choose for my thesis?

Detailed Question -

I have chosen 10 extracts from each novel because I think they are representative enough, and according to my department policy, there is a limited number of pages that I should not exceed. That is, I have limited space. Nevertheless, I have just read in two articles that the smallest acceptable sample for a qualitative research design is 15 (Bertaux, 1981; Guest, et al., 2006). I have noticed, however, [that] they are always referring to interviews. Does this apply to analyzing novels/stories as well? I mean I [am] doing a comparative study. I am confused. Is 10 for each novel enough?

2 Answers to this question

The same for this one. :-) We'll get back to you shortly from our Humanities experts.


From your query, we see you are striving to seek a balance between including as many samples/numbers (extracts) as necessary for the objective of your thesis and the space limitations necessitated/imposed by your institution. This is a good thing, and also necessary, especially for real-world research. In fact, in both the studies you have referenced (Bertaux, 1981 and Guest, et al., 2006), the key theme or guiding principle is ‘saturation.’ That is, when or where does one stop researching? The answer to this (apart from that of space constraints) is when one keeps getting the same answers, that is, the saturation limit is reached. After a certain number, it may not be worthwhile to have more numbers/extracts/references (as the case may be): they will only keep proving the same point.

In this context, as we had shared in a response to one of your earlier queries, it’s also important to keep the scope of the study in mind. If you don’t establish (or restrict, as some may call it) the scope, the study may go on interminably (if not forever), which may not be feasible to do. For the same reason, a researcher sets delimitations in a study, to regulate the scope of the study. However, when doing so, it’s necessary to provide the reasoning for this.

For further insights into scope and delimitations, you may find it worthwhile to go through the following resources:

So, in your case, as you have mentioned, if you think 10 is an appropriate number of extracts because you believe these are “representative enough,” you should go with this number. While we and your department can only guide and not direct you (though your department could possible do the latter too), the final call may have to be yours, based of course, on solid rationale and research acumen (which you will no doubt keep cultivating as you go along).

To answer your other question, the referenced studies, while meant more for interviews, should apply to novels/stories as well, or rather, especially well. Your study seems to around narrative forms, which is what fiction is about, though non-fiction can have great narratives too. :-)

Hope that helps. And all the best for further progress in your thesis!

For additional discussions over the sample size, you may refer to the following queries by researchers around peer review comments of their papers: