Take-away #1: Small sample sizes aren't necessarily a problem. Both Tracy and Leslie (presenters of the evening) were taking about their work (well, the work they are gearing up to do), and they both have between 6 and 12 participants for their research. I am thinking about my own dissertation process, my own "problem" (which isn't a problem, so I hate using that term, but whatever), and how many people can be my informants (at most 16, but most likely 10 or so will agree to be part of it). I've been thinking that AU might have issues with such small sample sizes. However, considering that I am not aiming at generalization (and neither are the presenters from last night's session), I am encouraged to continue on my current path for a dissertation proposal. It seems that AU is open to qualitative, small-sample, research for dissertations. I had a fear that I'd be stuck in a qualitative, "you must have something generalization" nightmare - a nightmare because that's now where I come from in my own research views :-)
Take-away #2: Just like the boy-scouts: Estote Parati (Be prepared). When you're doing interviews (live interviews) make sure you have a back-up. Good advice. I had never thought about it (perhaps because I am not at that stage of data collection yet). I was considering using Google Hangouts and perhaps using something like Camtasia to record it. This way I could (if there were video) also record any paralanguage that exists in our interviews and it could be an additional data point for analysis.
In terms of interviews and transcription, I was also thinking of outsourcing the transcription to a company. Looking briefly into this, it seems that $1/minute (or $2/minute if the audio isn't that great). If I assume that 8 people sign-on to be my study participants, at 2 interviews per person, between 40 and 60 minutes (making sure that I don't monopolize their time), the cost come out to around $1000 (wow!) Comparatively, Dragon 13 Premium (the educational version) costs $100, but I'd have to go back and review everything and cross check text produced with audio. That is 1000 minutes of audio maximum, so if we assume 3 minutes of corrections for each minute of audio, that's 3000 minutes, which works out to 50 hours of work. Hmm.... wonder if there is a grant to pay for transcription work ;-) I think there is a benefit of having to do the hard work yourself - it makes you more intimately familiar with the data you are working with, but from a student's perspective (who is on the clock to be done with their dissertation in by year 5), is that the best place to spend your time? I don't know, but we will find out ;-)
Some bits and ends:
- IPA was mentioned in Tracy's presentation (Interpretive Phenomenological Analysis). I just find it funny how acronyms bleed through to other disciplines. For me IPA means International Phonetic Alphabet and it's associated with linguistics. A good reminder to define your acronyms and your terms! (Thanks Tracy!)
- Have someone interview you with your open-ended questions! This seems like a given, but to me this was an "aha!" moment. Since I am interviewing some potential colleagues for my dissertation research (on group/collaborative processes), and I was a member of those collaborations, it makes sense to have someone ask me those questions. I was going to answer them anyway, from my perspective (researcher as the person being research, too!) but I think a dialogic approach makes a ton more sense (Thanks Leslie!)
- Finally, Leslie's point about not sending transcripts to participants without giving them some direction as to what to do with them is important. I wasn't even thinking about sending transcripts back for checking because I plan to bring in the participants' voices at many parts of the dissertation: case study approach, open document when I have more down, which will be open for commenting, suggestions, and corrections! So, I didn't want to inundate my fellow participants with too much stuff (granted it's all optional), but this gave me something to think about.