Experiments
Jan. 29th, 2013 08:35 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
colorlessgreenideasI've been feeling experimental lately. Not so much that I want to "try new things" any more than usual. I just want to run some experiments and track how they go. For most of these, I lack the means to make them scientifically or statistically rigorous enough to be conclusive. But they sound like fun to me, and I don't give a shit about publication, anyways. I thought this was as good a place as any to list these ideas. Chances are, I could find the "answers" somewhere on the internet or in a library, but that's not the point. I want to find things out myself.
Idea 1: OkCupid profiles with minor differences. How much do spelling errors influence responses? What about grammar? Type of photo (a- smiling at camera; b- brooding into distance; c- looking distracted; d- face not in photo)? Do the books, movies, or shows listed as "favorites" make a difference? Spirituality probably does matter. It'd also be fun to see the kinds of responses each profile gets. Are they well thought-out and articulate, or just a simple "Hey, you look cute"? Do women listed as bisexual get different kinds of responses from other women compared with lesbians?
There are some clear challenges to this kind of study. Other users would notice very quickly that the profiles are practically the same, and that would ruin the whole thing. The only solutions I've come up with are to run the profiles in several cities at the same time, or at different times in the same city. In either case, it introduces confounding variables. But hey, I'm not publishing this anyways.
Idea 2: Effects of various methodologies on student acquisition of verb tenses in English
Yeah, I know that could be a dissertation and then some, and in some ways it already has been. There's a fuck-ton of research on this topic already, and most of it comes down on the side of the communicative approach. So most of what I could do is replicate (to the best of my ability) those studies. But, again, what I really want to do is determine the results for myself. What techniques work best in my classroom, with my students? The literature suggestst that direct grammar instruction is not a good way to acquire grammar, but that's exactly what my students have asked for, multiple times. My students aren't dumb - they know what they're lacking, and to an extent, they know what does and doesn't work for them. One problem I have with a lot of SLA research is that it feels distant from actual classroom realities - it's hard to take the literature and apply it to real students, in my experience. And too many SLA researchers are content to say "Well, learners just don't acquire XY and Z as adults." My students want to acquire everything (at least, the motivated ones do).
Again, there's a lot of problems with this, from a scientific POV. Mainly, they're my students, and I'd be analyzing my teaching. Not exactly objective, but I could do my best. More challenging - I can't just split the class and teach one group one way and another group another way. It's all or nothing, and there's just enough overlap between quarters that I couldn't switch it up then, either. Really, the best I can do here is keep a good journal of what I've done, and try to track individual student progress based on writing samples (I find writing and speaking much better indicators than grammar tests).
Idea 1: OkCupid profiles with minor differences. How much do spelling errors influence responses? What about grammar? Type of photo (a- smiling at camera; b- brooding into distance; c- looking distracted; d- face not in photo)? Do the books, movies, or shows listed as "favorites" make a difference? Spirituality probably does matter. It'd also be fun to see the kinds of responses each profile gets. Are they well thought-out and articulate, or just a simple "Hey, you look cute"? Do women listed as bisexual get different kinds of responses from other women compared with lesbians?
There are some clear challenges to this kind of study. Other users would notice very quickly that the profiles are practically the same, and that would ruin the whole thing. The only solutions I've come up with are to run the profiles in several cities at the same time, or at different times in the same city. In either case, it introduces confounding variables. But hey, I'm not publishing this anyways.
Idea 2: Effects of various methodologies on student acquisition of verb tenses in English
Yeah, I know that could be a dissertation and then some, and in some ways it already has been. There's a fuck-ton of research on this topic already, and most of it comes down on the side of the communicative approach. So most of what I could do is replicate (to the best of my ability) those studies. But, again, what I really want to do is determine the results for myself. What techniques work best in my classroom, with my students? The literature suggestst that direct grammar instruction is not a good way to acquire grammar, but that's exactly what my students have asked for, multiple times. My students aren't dumb - they know what they're lacking, and to an extent, they know what does and doesn't work for them. One problem I have with a lot of SLA research is that it feels distant from actual classroom realities - it's hard to take the literature and apply it to real students, in my experience. And too many SLA researchers are content to say "Well, learners just don't acquire XY and Z as adults." My students want to acquire everything (at least, the motivated ones do).
Again, there's a lot of problems with this, from a scientific POV. Mainly, they're my students, and I'd be analyzing my teaching. Not exactly objective, but I could do my best. More challenging - I can't just split the class and teach one group one way and another group another way. It's all or nothing, and there's just enough overlap between quarters that I couldn't switch it up then, either. Really, the best I can do here is keep a good journal of what I've done, and try to track individual student progress based on writing samples (I find writing and speaking much better indicators than grammar tests).
