So What Did Your Gay Uncle Give You for Christmas?
In the wake of the Andrew Sullivan racial-intelligence flap, Half Sigma's post on the topic devolved into a debate over the heritability of homosexuality. Regrettably, despite it being an "HBD" blog, the commenters generally proved themselves rather dense, with the host not helping much either. A common theme was the absurd "Gay Uncle" theory. This theory posits that homosexuality would be a favored trait because gay uncles help out raising their nieces and nephews. Mathematically this seems ridiculous as an uncle, sharing only 1/4 of his niece's genes, would have to be twice as beneficial to a niece/nephew than a parent (who shares 1/2 her genes with her children) to make up for the shortfall. But since I'm only mediocre in math, that doesn't bother me so much as the common-sense angle. Do these people know any gays? Have they ever seen them pay more than the most superficial attention to any children?
But perhaps gay men are more successful on average otherwise, and can bequeath benefits to their nieces/nephews. So I took a look at the General Social Survey (GSS) to see if there's any evidence of such a phenomenon. I looked at the GSS variables SEXSEX5 - which asks respondents the sex of anyone they've had sexual relations with over the last 5 years - and CONINC which is a measure of income in constant (inflation-adjusted) dollars. This question has been asked since 1991 so I tracked this over that time.While the incomes of gays are jumpy, the incomes for straight males are substantially higher for all years but three. If there were a tendency for gays to be highly successful and thus well situated to help out their relatives, it sure doesn't show up here.
But, the sample sizes for gays are small, so how reliable are this data? Gay men constitute around 2 to 3 percent of the male population (I'm leaving out those with bi-sexual relationships - their incomes are consistently lower with one outlier). So if we wanted to validate that such a small population could be represented with reasonable accuracy at such low numbers, what might we look at - particularly when it comes to income? Obviously, Jews make up about the same proportion of the population, and the fact that Jewish incomes significantly outstrip those of non-Jews is well known. So this should show up clearly in the GSS if the income figures are reliable. Well here it is:Just as we'd hope to see, incomes of Jews are significantly higher than non-Jews. Similarly, if gays were consistently more successful than non-gays, we should expect to that here as well - but we see, mostly, the opposite.
So the lesson is, if you want help from your uncle, you'll be much better off if he's Jewish than if he's gay.
Gays and Jews get a lot of attention in the media. While I know the readers of this blog are well aware of it, it might not be clear to the random person exactly what proportion of the population these groups constitute. Below, in full scale, are the percentages of those self-identifying as Jewish among all respondents and those reporting sexual relations exclusively with other men among all male respondents in the GSS:
But perhaps gay men are more successful on average otherwise, and can bequeath benefits to their nieces/nephews. So I took a look at the General Social Survey (GSS) to see if there's any evidence of such a phenomenon. I looked at the GSS variables SEXSEX5 - which asks respondents the sex of anyone they've had sexual relations with over the last 5 years - and CONINC which is a measure of income in constant (inflation-adjusted) dollars. This question has been asked since 1991 so I tracked this over that time.While the incomes of gays are jumpy, the incomes for straight males are substantially higher for all years but three. If there were a tendency for gays to be highly successful and thus well situated to help out their relatives, it sure doesn't show up here.
But, the sample sizes for gays are small, so how reliable are this data? Gay men constitute around 2 to 3 percent of the male population (I'm leaving out those with bi-sexual relationships - their incomes are consistently lower with one outlier). So if we wanted to validate that such a small population could be represented with reasonable accuracy at such low numbers, what might we look at - particularly when it comes to income? Obviously, Jews make up about the same proportion of the population, and the fact that Jewish incomes significantly outstrip those of non-Jews is well known. So this should show up clearly in the GSS if the income figures are reliable. Well here it is:Just as we'd hope to see, incomes of Jews are significantly higher than non-Jews. Similarly, if gays were consistently more successful than non-gays, we should expect to that here as well - but we see, mostly, the opposite.
So the lesson is, if you want help from your uncle, you'll be much better off if he's Jewish than if he's gay.
Gays and Jews get a lot of attention in the media. While I know the readers of this blog are well aware of it, it might not be clear to the random person exactly what proportion of the population these groups constitute. Below, in full scale, are the percentages of those self-identifying as Jewish among all respondents and those reporting sexual relations exclusively with other men among all male respondents in the GSS:
Labels: genetics, GSS, homosexuality