David Epstein wrote:

> On Wed, 3 Mar 2004, Beth Benoit went:

> > Didn't we conclude that Hite's data had all kinds of statistical
> > problems. (Particularly her lack of random sampling?)

> I've seen comments to that effect; I can't remember whether they were on
> TIPS.  (I search the archives for "hite" and didn't find much.)  I
> think you're right--Hite's findings probably can't be generalized
> beyond the relatively elite population from which she sampled.

As I recall, it�s a lot worse than that! It can�t be generalized *for* the
population she surveyed (not �sampled�). If I remember correctly, her
survey was a textbook example of how not to conduct a study. Not only did
she contact specific groups and magazines for her subjects, they were
self-selecting! Some study! Some �statistics�!

Unfortunately some people (especially the kind of literary types who
review books like those of Sherry Hite) are impressed by lots and lots of
figures and the trappings of statistical procedures. (There�s nothing
wrong, of course, with being a �literary type�. I just wish more of them
would get to know a bit about scientific methods and logical fallacies
before pontificating about the results of dubious surveys.)

Allen Esterson
Former lecturer, Science Department
Southwark College, London
[EMAIL PROTECTED]

http://www.human-nature.com/esterson/index.html
http://www.butterfliesandwheels.com/articleprint.php?num=10

---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]

Reply via email to