Men get uglier, Women get Prettier 04.04.2007

There’s an article out on a study done that supposedly shows that beautiful people tend to have more daughters.

While I don’t have anything against the conclusion, I do have issues with the road to that conclusion.

Among many questions, one that pops to mind is: how do they measure beauty?

I read through the actual published academic article. And well…

They do, in fact, have a point system. The study used gathered data from in-home interviews with people that volunteered for the study.

At the conclusion of each in-home interview, the interviewer is asked to rate the respondent’s physical attractiveness on a five-point ordinal scale (1 = very unattractive, 2 = unattractive, 3 = about average, 4 = attractive; 5 = very attractive)


You also simply have to love the definition of an “objective” measure of physical attractiveness (emphasis mine):

the Longitudinal Study of Adolescent Health (Add Health), the only large representative survey data which contains an objective (i.e. not self-reported) measure of physical attractiveness

I wish I was able to use such definitions of objective measurement in my academic papers.

“David, why did you write in your paper that gravity is God’s sticky paper to keep human roaches rooted away from heaven?”

“My neighbor said so. It’s objective! He’s a crazy religious nut, but don’t worry, I didn’t self-report it. He told me, so it must be true.”

I mean. Seriously? The entire basis for this so-called scientific article is on the personal evaluations by anonymous interviewers on the physical attractiveness of others? What if some of the interviewers had a freak fetish for short dwarfish women with beady eyes, receding hairlines and wooden teeth? Does that mean it’s really the short balding women with scurvy that mostly have daughters?

In the conclusions, the paper mentions that some of the things it draws from that supremely sketchy premise are confusing:

It is only the respondents rated “very attractive” who have a significantly less chance of having a son. Why there is a seeming qualitative difference between “very attractive” respondents and merely “attractive” respondents is not at all clear.

Here, I’ll clear up the confusion. Earlier, the paper states that “only 11.2% are rated as [very attractive] by the reviewer”. It also states that it only used those volunteers in the survey who had “at least one biological child (n=2972)”.

That comes out to around 332 people, rounding down.


Then, in a table provided, it breaks down per score the percentage of a male first-born child:

Very unattractive: 0.50
Unattractive: 0.56
About Average: 0.50
Attractive: 0.53
Very Attractive: 0.44

I can see the author now. You must be so amazed, oh researcher.

WHAM! JACKPOT! you must think. That 0.44 for very attractive sticks out like a sore thumb! I’m going to be famous, you think. Shit, I’m so not going to have a girl, you moan.

But, numerically, what exactly does that mean?

Let’s suppose we have 100 people. 3 groups of 100 people who are, respectively, super hot, ugly, and your neighbor Sally, who isn’t terribly attractive though you wouldn’t it regret too much if you had a wild drunken romp one day. Sally Average.

The super hot .44 of 100 gives you 44 princes of Maine.

0.56 (ugly) of 100 gives 56 kings of New England who have inherited horrible ugliness.

.5 (Sally) of 100 gives 50 supremely mundane infant boys.

The entire argument hinges on the fact that 12 people had boys rather than girls. A whopping 6 people less than about average.

But no! you cry. It’s statistically significant! How come the really fucking hot people get to have girls?

Ah, but they don’t. 56% of them had girls. 56% of an elite 11.2% of a total pool of 2972 people.

And in a pool of that size, I can find any single conclusion you want me to find.

The only thing this proves is that if the 2972 people are a truly representative sample of the population — and if that “objective” measure of attractiveness was actually objective — then it’s depressingly difficult to get someone very attractive.

Where are all the hot hos at?

Shit man. It’s hard out here for a pimp.

Though there’s an article I’d gladly endorse. I can see it now. “It’s Hard Out Here for a Pimp: A Further Implication of the Generalized Trivers-Willard Hypothesis As It Relates to Creative Entrepreneurs.”

That sound you hear?

Yeah, that’s the sound of peer review getting shot, burned, maimed, quartered, and eaten by female scurvy dwarves — I mean very attractive people.