Physical attractiveness and intergenerational social mobility

I would look seriously into how that attractiveness was evaluated, and the measurement might be more accurate for men:

1. Criteria for men and women are very different: tall vs. thin is key. If the profiles list height but not BMI, that would lead to discrepancies.

2. Heterosexual women are more likely to agree on which men is attractive then the reverse (which is a classic problem in dating apps: then attention of fewer women is also more focused)

3. Make-up: unattractive men don’t have that many commonly used options to be more attractive (growing a beard, painful surgery), while women are more willing to spend half an hour to change that. It levels the playing field making the original measure less relevant.

4. When the data was collected:

> assessed when individuals are around 15 years old

My memory of biology class is faint, but I believe that men change less after 15 than women do.

In general, that’s an area that was understudied 25 years ago. I’m happy to read so many references, but couldn’t find my old sociology professor. He was struggling with the impact of height on one’s professional career. He published a lot using a sample of alumni from the other university where he taught (to control for education). Strangely, the impact of height became weaker over the decades… He had so many theories (computers, mostly)—only to realize (after two publications) that it was because the cohort was increasingly feminine. So, he published in succession that:

1. height was key;

2. the gender gap was more than explained by how women were shorter (that paper was not popular);

3. height mattered a lot less for women, therefore the gender gap thing was still a problem, but athleticism was (he had an athletics grade). (That’s how I learned that `log_salary~height*gender`, `log_salary~height+gender` and `log_salary~height:gender` meant completely different things.)

3 bis. strike that: athletics mattered for both, but less than expected; BMI was key for both but mostly women. (That was the example other professors used to teach us about the scissor effect in regression: BMI and height are related.)

4. Height and BMI were proxies for attractiveness that mattered for both but had to be measured differently. He had several sets of photos, one typically earlier and without and another later with make-up… More importantly, students were 23-24 at that point.

But all that became obsolete when most students ended up studying and finding jobs abroad and that effect dominated everything. That professor moved to study accents.

Anecdotally, as a 6’0” moderately-attractive dude in SF: I’ve laughed at the statistically unlikely number of founders/CEOs/upper management who are not only taller than me, but also are not subtle when they use their imposing height in conversation to get everyone else to just go along with whatever BS they are spouting. Bonus giggles if they are part of the hoard with thick hair and a square jaw.

Everyone is stupid except me.

I suspect it’s the lamppost effect: the data was here, so they used that rather than a more relevant observation that would have cost a lot more money.

As it’s a health-related database, I suspect it’s a photo taken to control for acne, but that’s just speculation.

But definitely agree that it’s probably skewing the results a lot.

Selection bias? Most of the women who _commented_ feel that way, in the same way that 70-95% of people who read reddit/hn/whatever only lurk. Vocal minority maybe.