======= Date Modified 25 Jul 2012 13:14:43 =======
40 interviews might be possible to reasonably analyse from a qualitative perspective. 50, however, starts to be a quite large number, depending on how deeply you want to analyse each interview from a qualitative standpoint. But from a quantitative research perspective, where you will mainly focus on answers that can be quantified, rather than free flowing open questions and answers, 50 would be easily do-able from an analytical perspective. If you are doing 50, you might want to think about having the majority of your answers in a quantifiable form (e.g. with a, b, c and d answers). Even then, it's always nice to have some free flowing answers as well with open questions.

I think your main point though is true as well, you have to be careful about how many interviews you can actually get. It may well be possible to get 50 within a longer time frame, but this could delay you writing up your PhD. If finding the interviewees is a problem, I would think very carefully about your approach to finding your sample - can you improve the quality of communications (by phone, email, snail mail) to each potential participant; can you ask each actual interviewee who takes part for any other interviewee suggestions (snowballing); are there sources to find interviewees you've not thought of (e.g. in databases at places like the British Library or an association in your topic area; any experts in your area who could suggest potential interviewees?). Also contact interviewees several times until you get a response - sometimes busy people will agree even though they've not bothered to respond to the first two approaches you've made. You don't really have to stop asking someone until they actually say 'no' - then of course it's only polite and ethical to leave them be.

I think your best argument with your supervisor that 50 is overkill would be to find published PhDs on similar topics in your discipline and show how many interviews they are using - if you can prove that 20-35/40 is the average, then that seems a good argument that your thesis should pass with that number! Also ask your supervisor what they think is not saturated in your topic with the interviews you've currently completed - can they make a good case for further interviews because some issues remain unresolved in the data you've already collected? Then you might be able to point out when you've got up to say 30 or 35 interviews, that actually the topic is well saturated. If you want to take a qualitative approach with 20-40 interviews, you might also want to say to the supervisor that you want to choose an examiner as well who will think that an-indepth approach to analysing that number of interviews will be appropriate.

I'm sure however there are some very hard working qualitative students who've done 50 plus interviews. I've come across PhDs with 60 qualitative interviews... I'm sure that's not a record by any means.