Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site ubc-vision.CDN
Path: utzoo!utcsrgv!ubc-vision!manis
From: manis@ubc-vision.CDN (Vincent Manis)
Newsgroups: net.motss,net.med
Subject: Re: Hormone markers for homosexuality?
Message-ID: <626@ubc-vision.CDN>
Date: Wed, 10-Oct-84 21:30:49 EDT
Article-I.D.: ubc-visi.626
Posted: Wed Oct 10 21:30:49 1984
Date-Received: Thu, 11-Oct-84 03:58:49 EDT
References: <982@bbncca.ARPA>
Lines: 21

This is another example of silly science: a correlative study using
Organization: UBC Vision, Vancouver, B.C., Canada
Lines: 19

ill-defined metrics. First, I'm not clear why anybody would care about
differential hormone response for people of different sexual orientations;
second, the sample group was small and there was apparently some sort
of systemic sampling bias. Result? A ``significant'' result which
although unconfirmed suggests the need for further research and certainly
further funding.

Reminds me all too much of Cyril Burt's IQ measurements, which (in the
studies which weren't faked) purported to prove that identical twins
had the same IQ, regardless of environmental differences. Turns out that
Burt's IQ metric was so vague that one has no way of validating it.

As I believe that sexual preference has both environmental and congenital
components (as do most behaviours), I have little or no patience with
studies that attempt to find causes. Why not attempt to correlate 
hormone response with socioeconomic status, political views, or favourite
programming language?  I have my own theories about Cobol programming.