Why might this be interesting? For one thing, some people have fetishes for things that seem, well…bad. Getting hurt. Letting other people control and abuse them. As if they have a drive toward weakness and unhappiness. This is kind of reminiscent of the self-sabotage and bad decisions some people make throughout their lives (for example, marrying a spouse who treats them the same way as an abusive parent). Sometimes I conceptualize this as them having a set point of low self-esteem and degradation that they try to enforce, regardless of its cost to their well-being. If this had the same roots as sexual masochism, that would be worth studying.
But I didn’t find anything interesting like that in the data.
BDSM preferences were heavily gendered. Of people who expressed a preference, 71% of cis men preferred the dominant role, compared to only 16% of cis women (18% of trans women; insufficient sample size of trans men). This was such a big difference that gender swamped every other effect, so I limited the analysis to cis men from this point on, since they made up most of the sample.
80% of straight men preferred the dominant role, compared to only 34% of gay men. This was such a big difference that orientation swamped every other effect, so I limited the analysis to straight cis men from this point on.
In order of importance, here are some factors that made the men in this sample more likely to be dominant, rather than submissive. All of these are self-rated:
– Good social skills
– High risk taking
– High ambition
– Conservative political values
– Low anxiety
– High status
– Belief that others are trustworthy
Forget the lurid stories about high-powered executives who secretly want to be abused and degraded. Most of these are pretty straightforward. The more successful and less neurotic a man’s personality, the more likely he is to be sexually dominant. But none of these are big effects. For example, conservatives are more dominant than liberals, but the exact numbers are 83% vs. 70%. Straight men are still mostly dominant, regardless of politics.
Submissive men reported lower sex drive, fewer dates (9 vs. 18), fewer sexual partners (5 vs. 11), and fewer long-term relationships (2.7 vs. 3.5) than dominant men. This probably has to do with the worse social skills and decreased risk taking.
This was so boring that I tried switching to sadism vs. masochism. Most self-reported sadists were also self-reported masochists, so I took only the subset of people who reported one but not the other. This showed similar patterns to dominance vs. submission, so much so that it’s not worth going over them separately.
Two mildly interesting findings. First, although psychiatric issues in general only affected these roles weakly and inconsistently, men with OCD were four times less likely to be sadists (and somewhat more likely to be masochists) than anyone else. This seems like an extreme form of the effect of high anxiety, plus maybe an obsessive fear of hurting someone else, or a response to feelings of guilt.
Second, although there was no effect of self-reported childhood trauma, men who grew up poor reported about twice the sexual sadism rates of people who grew up rich or middle-class; whether or not you were currently poor mattered a little less. Given how many comparisons I did, I’m not very confident in this result even though the effect size is pretty big.
The data didn’t support any kind of connection between dominance/submission/sadism/masochism and more prosaic forms of selfishness or self-sacrifice. Men were about equally likely to give money to charity, or identify as effective altruists, or hold various opinions in moral philosophy, regardless of their sexual roles.
Although I’ve made every other part of the survey publicly available, given the sensitivity of fetishes I’m keeping these particular answers private. If you are a professional researcher (or an amateur researcher with a good track record of professionalism and data integrity), and you want to test these results, please email me at scott[at]slatestarcodex[dot]com and we can discuss how to make that happen.