I met with two of my committee members (chair and sponsor) on Thursday to talk about steps forward for the dissertation. My sponsor thinks I should add another (set of) analyses to the dissertation, as he thinks my main aims aren’t ambitious enough for a dissertation. I agree, but I think where we disagree is what I should do.
I have to say that my dissertation sponsor is pretty much everything an adviser should be. He’s been my primary adviser since my first year of graduate school, and he’s been really involved and invested in my professional development since the start. We generally meet every other week (although it’s been less frequently lately, but that’s to be expected). Because he’s untenured, he needs to be really productive, so that’s translated to me being moderately productive myself. I’ve got to learn first hand what the role of an assistant professor looks like. I always proclaim this on boards or when talking to prospective students about graduate school; sometimes they get worried or hear from others that they shouldn’t choose untenured faculty members as advisers because what if they leave? Not that that’s not a threat, but there are some pros to choosing an untenured faculty member as a sponsor too.
However, my sponsor’s #1 suggestion was to do some coding of a qualitative question in the longitudinal part of my dissertation data (some of the data is cross-sectional, taken at time 1, and some of it is qualitative taken at times 1-8). I think this is a very very bad idea. I’d have to come up with a codebook and code 8 weeks’ worth of qualitative statements, however short they are. I really dislike coding qualitative data (as much as I realize that it’s useful). I’m a quantitative researcher, and I want my dissertation and the subsequent papers to be examples of my work not only as a substance use researcher but also as a methodologist and quantitative researcher. But more importantly, the qualitative work will take a LOT of time and throw off my timeline substantially, I think. (I might also add that my sponsor has a little bit of a history of suggesting work that will take a really long time.)
My chair, thankfully, agrees with me.
My idea was to do a latent class analysis or structural equation model looking at one of the constructs I’m investigating as a latent variable. The idea is to see what elements really make up that latent variable, and try to put some definition to the construct. There’s been calls in the literature in my field to examine this construct in more depth in the context of my area, and some really relevant recent research, so I’m thinking it’ll be well-received in the field. My chair likes this idea. My sponsor does too, but he’s a bit skeptical about whether it’s “enough.” (I don’t know what “enough” is, unfortunately.) There’s another person on my committee who’s big into this area of the field and so I think I want to make an appointment with him to ask him about it.
But then a thought just occurred…if he didn’t think it was “enough,” why let me go forward with the proposal? After all the proposal WAS accepted! I mean, I don’t mind because I really like thinking about this stuff and the prospect of the new area actually excites me, even though I know it will be difficult.