Tuesday, May 29, 2007

The Spatial Problem

JJ from NOLA sends a link that is going to get me in trouble. Thanks so much, JJ.

But, here goes:

Sexual Orientation Affects How We Navigate And Recall Lost Objects, According To Study Science Daily — Researchers at the University of Warwick have found that sexual orientation has a real effect on how we perform mental tasks such as navigating with a map in a car but that old age does not discriminate on grounds of sexual orientation and withers all men’s minds alike just ahead of women’s.

The University of Warwick researchers worked with the BBC to collect data from over 198,000 people aged 20–65 years (109,612 men and 88,509 women). As expected they found men outperformed women on tests such as mentally rotating objects (NB the researchers’ tests used abstract objects but the skills used are also those one would use in real life to navigate with a map). They found that women outperformed men in verbal dexterity tests, and remembering the locations of objects. However for a number of tasks the University of Warwick researchers found key differences across the range of sexual orientations studied.

For instance in mental rotation (a task where men usually perform better) they found that the table of best performance to worst was:

Heterosexual men
Bisexual men
Homosexual men
Homosexual women
Bisexual women
Heterosexual women

In general, over the range of tasks measured, where a gender performed better in a task heterosexuals of that gender tended to perform better than non-heterosexuals. When a particular gender was poorer at a task homosexual and bisexual people tended to perform better than heterosexual members of that gender.


Now, class, look at that last paragraph in the story. "In general...." Wow! What an absurd inference. There are only two genders. There is no, "in general" from two observations.

It is equally consistent with the data to say:

1. Men's brains and women's brains are different. Men, on average, are better at spatial problems and recognizing their location. Perhaps not a lot better, and the best women are better than the worst men, but still statistically different in terms of comparison of means.

2. And there is a clear genetic basis for sexual orientation/preference, since within gender there are differences that group by sexual preference. If there were no genetic basis for homosexuality, then we would expect uniform distribution within gender.

(Nod to JJ, who never gets lost, ever)


Dirty Davey said...

Why don't we just conclude that "in general" refers to generalizing over the range of tasks, rather than over the (nonexistent) range of genders?

Simon Spero said...

The best women are better than the worst men
Is this correct?
For example, in figure 3(a), heterosexual men outperform all women; in fig 3(e), the women dominate (nudge nudge, wink wink, say no more, say no more).