http://www.talyarkoni.org/blog/2010/04/01/some-thoughtful-comments-on-automatic-measure-abbreviation/ ...or you get no soup for one year! Fri, 10 Feb 2012 12:37:32 +0000 hourly 1 http://wordpress.org/?v=3.3.1 http://www.talyarkoni.org/blog/2010/04/01/some-thoughtful-comments-on-automatic-measure-abbreviation/comment-page-1/#comment-660 Sanjay Srivastava Fri, 02 Apr 2010 19:59:42 +0000 http://www.talyarkoni.org/blog/?p=476#comment-660 Thanks for taking the time to reply. My comment about being replaced by machines was a joke (I think). I don't want to go back to the days where our predecessors had to rotate their factor matrices by hand, for example. On the other hand, if SPSS 45.0 comes out with a Genetic procedure in the pulldown menu, I'm going to reserve the right to play the part of the crotchety old emeritus. I hadn't realized that the correlations on the IPIP website were from training data. That closes the gap significantly. My proof/pudding comments weren't meant as a knock against this paper (and I don't think you took them that way). That's beyond the scope of what you'd need to do in a first paper, and it's more of a future directions issue. WRT that, in my experience abbreviated measures are often most needed in large, expensive surveys and longitudinal studies. But a catch-22 is that those are the situations where the stakes are so high that researchers will be hesitant to use unproven measures based on shiny new techniques. So there'll need to be an interim period where enough researchers use GA-derived measures in lower-cost studies where maybe brevity isn't such a premium. I'm glad to hear you're already starting to do that. Thanks for taking the time to reply. My comment about being replaced by machines was a joke (I think). I don’t want to go back to the days where our predecessors had to rotate their factor matrices by hand, for example. On the other hand, if SPSS 45.0 comes out with a Genetic procedure in the pulldown menu, I’m going to reserve the right to play the part of the crotchety old emeritus.
I hadn’t realized that the correlations on the IPIP website were from training data. That closes the gap significantly.
My proof/pudding comments weren’t meant as a knock against this paper (and I don’t think you took them that way). That’s beyond the scope of what you’d need to do in a first paper, and it’s more of a future directions issue. WRT that, in my experience abbreviated measures are often most needed in large, expensive surveys and longitudinal studies. But a catch-22 is that those are the situations where the stakes are so high that researchers will be hesitant to use unproven measures based on shiny new techniques. So there’ll need to be an interim period where enough researchers use GA-derived measures in lower-cost studies where maybe brevity isn’t such a premium. I’m glad to hear you’re already starting to do that.

]]>