Gait analysis and evo psych

Dave Garbutt writes,

Perhaps you will find this interesting. It is a deconstruction rather than the original, but how to analyse the data might be an interesting challenge worthy of your readers….

I don’t have anything to add about this, except that there’s a long tradition of oveinterpreting data on menstrual cycles. I remember an example from an old and highly-recommended statistics textbook (Say it with Figures, by Hans Zeisel) that I used in a class once: he had an interesting example along with a graph and story, but then when I took a look at the original article being cited, I couldn’t find anything like what was being claimed in the textbook. (This case is a little different, because here it’s the scientific article itself that’s being called into question, but still it reminded me.)

4 thoughts on “Gait analysis and evo psych

  1. The interpretation of the data doesn't make much sense to me. From an evo-psych perspective, gaits aren't attractive in and of themselves. Rather, gaits that transmit information about fertility should become more attractive to men. This study claims to find that the more attractive gait is related to nonfertile periods (unless I missed something?) which argues against any evo-psych interpretation of the data. Men should be selected for their ability to tell the difference.

    But this is evo-psych; they can explain away anything.

  2. Hi,

    I should have mentioned bad science too in my email to Andrew.

    Deconstrunction can certainly be fun. But I was wondering if these data came to you as a consulting problem, what would you do?
    They used PCA (if I recall correctly) to discriminate the menstrual phases.
    Then they constructed gaits for the testers to review.

    What other analysis techniques can handle this amount of data?
    Perhaps Functional data analysis?

    DaveG

  3. Given the amount of data and the representativeness of the sample, I would suggest that a dartboard "handle" that data accurately.

    But then you realise it would give too many possibilities. A deck of cards?

    Well, no. Pinochle?

    Possibly. But probably euchre.

Comments are closed.