I'm afraid the video is way too jerky and the sound on the low side, but if you wanted to get a sense of Lisa Handley, here's about three minutes of her analysis of Alaska voting patterns. This is from Tuesday, May 24, 2011 at the Redistricting Board meeting.
As mentioned in the previous post, you can see Dr. Handley's powerpoint notes on her Alaska Analysis at the Redistricting Board's website. The first part of the power point is here. It covers the Voting Rights Act in general.
I am a bit tired as I type this, so hopefully I won't regret any of it in the morning. :)
ReplyDeleteEcological inference is taken very seriously by political scientists, but from her presentation, one might reach the conclusion that there is only one technique available for ecological inference; in fact, there are dozens. There is a lot of debate over which one is the best one.
Second stage ecological regressions that simply truncate the final distribution are well-liked by some people I highly respect. Beta binomial ecological inference models avoid the truncation problem but also generally bring non-symmetrical distributions into play, which can create some headaches. There doesn't appear to be a lot of research on second stage ecological regression on manifolds, but that would be an ideal approach with the right manifold.
Well, don't you learn the "science of presentation making" in the US? We are told all the time that we shouldn't put too much info to one slide because it takes time to read them and it drives the attention away from the presenter.
ReplyDelete