How To Achieve A Titillating Headline

Glenn Reynolds linked this titillating headline:


Of course, when you read the article (of course I clicked through, I have no pride),  you find that:

  • The sample size is approximately 40
  • The sample was from a group of people who self-selected to attend a seminar by the owner of a sex-toy business

The "3% who participated in bestiality" is actually 1 person out of 40 who have a self-selected interest in pushing sexual boundaries.  With a little larger sample size, a bit poorer math, and a bit more work goal-seeking to a desired outcome, this might almost meet the standards of climate research.

Which is all a relief to me -- after 30+ years of being a Yale hater, I was afraid I might have to admit it was a more interesting place than I thought.


  1. MingoV:

    I've been very disappointed in Glenn Reynolds' choices of citations. He doesn't seem to have a functioning bullshit filter. He doesn't read much of what he links to and, therefore, makes inappropriate comments. He loves anything about (essentially nonexistent) nanotechnology. He commonly cites medical "advances" that are based on unsubstantiated hype from pseudoscientific "researchers" trying to get venture capital funding (just like most of the "green" energy hucksters). I realize he's one of the few bloggers who makes a profit, but I believe he would do better if half his links weren't nonsense.

  2. Gattsuru:

    Yale does tend to recruit from urban, upper-class folk who are less likely to have constant exposure to animals (and especially unfixed animals). The general population achieved a greater percentage in the Kinsey reports, and while those had a number of procedural issues, they were at least statistically significant.

    But it is stupidly bad reporting, and I would hope for better from Mr. Reynolds.

  3. marque2:

    I understand there are lots of issues with the Kinsey report, overstating non-hetero human relationships in general. It is one of those things, if you do a study and show that everything is normal, it won't get much press, you show things are more wild and wacky than assumed and your report will get broadcast everywhere.

  4. Gattsuru:

    There's some issues with Kinsey's work (ethical issues about reporting, oversampling of prisoners and runaways, selection bias given the time period), but most of the results remain when you filter the data or in separate studies. Even more recent works, which I expect would reduce the importance of the volunteer bias.

    The numbers don't actually say what a lot of modern-day readers think they do, but that's a different issue.