Dr. Aaron Sell, evolutionary psychologist and Lecturer at Griffith University (Australia), fills in as guest blogger with this humorous rant about a recent Journal of Personality and Social Psychology article that got under his skin.
So here's what ruined my day.
I'm reading Eagly and Wood's new piece in Perspectives on Psychological Science. I do this because I think it's important to read people who disagree with you. They mention Clark and Hatfield's classic 1989 article documenting that approximately 0% of women (and about 80% of men) are willing to have casual sex with strangers who approach them and ask for it. Furthermore, when asked instead for dates, women became much more likely to say “yes” (about 50%) while men became less likely to say “yes” (also about 50%). So in short, about 30% of the men will have sex with a woman but not date her. Women showed nothing like that. Not surprisingly this data is a thorn (a dagger? a sword? a battering ram infested with ebola?) in the side of those who wish to argue that men and women are essentially the same when it comes to sexual relationships. This isn’t the only data point showing this difference: the field of anthropology is stuffed full of ethnographies that reveal the same pattern, evolutionary biology shows the same pattern in hundreds of species (the sex that invests less in offspring will set lower thresholds when selecting mates), market analyses of pornography, romance novels, prostitution, and so forth show the same patterns, and on and on. But the Clark and Hatfield study is the most direct, profound, scientific, succinct demonstration of the silliness of the view that men and women have the same sexual mechanisms.
"No sex differences in interest in casual sex emerged when the potential partner was a famous, attractive person or someone reputed to be sexually skilled (Conley, 2011)."
I became verbally abusive at this point and took a few moments to meditate on the oneness of the universe to restore my calm. I knew I would have to look up this study and find out how the hell she managed to find this result. Keep in mind what it claims: if a man has a reputation for being sexually skilled he should be able to walk up to female college students and elicit sex from about 80% of them with one line of dialog. Further, if the attractive man in the Clark and Hatfield study had been famous then that 0% of women would be multiplied by whatever imaginary number X is required such that X*0% = 80%. I was suspicious.
So I found the Conley paper in the prestigious JPSP (I smashed my phone). It describes tests between Sexual Strategies Theory which she defines without its cognitive element - so that men "want" to spread their seed and women are "motivated" to find the best genetic material for their ova. It contrasts this silly position with her more reasonable explanation - pleasure theory. "According to this theory, sexual reproduction is a by-product of sexual pleasure, rather than the reverse." (Aside: has anyone informed biologists of this possibility? What about the life-sustaining byproduct of food consumption?) This is simply confusing selection pressures with the cognitive motivations that evolved to solve them. It is an elementary mistake, but a common one for people not trained in evolutionary psychology. If that were the only problem it wouldn't deserve much comment. But it goes on...
In short, the paper describes three studies that claim to “evaporate” the Clark and Hatfield sex difference. First, in an unappreciated bit of irony, she drops the real-world approach and replaces it with hypothetical scenarios. "I encouraged participants to suspend their disbelief for the purposes of the current research..." I have similar advice for anyone reading the study.
Note the demand characteristics inherent in the “You can hardly believe your eyes!” and “Still more amazing...” Finally, rather than ask whether the women would have sex with Johnny Depp...which would mimic the original study, the women are asked to rate from 1-7 their likelihood of agreeing “to a sexual encounter”. “7” now means “very likely” rather than “yes.” Likert scales are fine, but you need to remember that they have implicit baselines. When I ask women how tough they are on a 1-7 scale, the mean is 4. When I ask men the mean is also 4. But this does not mean women think they are as tough as men. Furthermore, “sexual encounter” may mean something different to men and women – as has been demonstrated in the “emotional cheating” literature.
Regardless, the data show that women and men are both “somewhat likely” to agree to have sex with a famous attractive celebrity while vacationing in Malibu by themselves. I should point out that, unless there’s a sex tape I’m unaware of, we have no idea how good Johnny Depp is in bed. He could be all Edward Scissorhands in the sack for all we know. The point of pleasure theory is that women say “no” to casual sex because they assume men aren’t any good at sex. What evidence do we have that Johnny Depp can provide good skilled union work? I have no idea. But regardless, the sex difference appears to be gone when you crank attractiveness, wealth, respect, cool, and famous up to the max, make the scenario hypothetical, replace a “yes or no” question with an ambiguously baselined 1-7 question about a vague sexual encounter, encourage your participants to suspend their disbelief and cue them in that you think it’s amazing that Johnny Depp sees them and wants to have sex with them (I went outside my office and beat a passing graduate student senseless with one of Tooby & Cosmides’ shorter papers). Finally, the subjects were not asked about whether they would go on a date with Mr. Depp, so the crucial pattern elicited in the Clark and Hatfield study was not tested at all.
The data show women being likely to agree to an encounter with Depp at 4.09 on a 1-7 scale, and a bunch of lying men at 4.16 for Angelina Jolie. Incidentally, there is no explanation for why real world tests show average female college students able to wrangle 80% of men back to their place but Angelina Jolie operating at a Malibu beach can only pull a 4 out of 7. Are we meant to assume that subjects think Angelina Jolie is sexually unskilled?
Now for the even less believable part. Is it really true that men with a reputation for being sexually skilled can have sex with 80% of college-aged women upon approach? Once again we have a hypothetical example. The subject pool: “participants were recruited by research assistants who posted the URL for the study on their social networking web pages.” (I ran downstairs and strangled the hobo that begs for change outside the student center.)
To her credit, the author recognized that asking women about Johnny Depp isn’t really a clean control for sexual skill, but actually manipulates (massively) a host of other variables that the alternative position (Sexual Strategies Theory) claims are relevant to women’s mating psychology. So to control for this, she decided to replicate the results by asking women about another familiar group: their friends. Participants were asked to imagine an opposite sex friend with whom they were never romantically involved, to whom they felt the closest. I suppose I could quibble that having sex with your “closest” friend is not casual by definition, but there are better shots to take.
So read it: “During a conversation with your male [female] friend, he [she] says to you, “I have been noticing how attracted I am to you. Would you go to bed with me tonight?” Again, rather than answer you are instructed to give your 1-7 probability of agreeing to a “sexual encounter.” Nonetheless, an odd thing happened. Women said no: M = 1.97 out of 7; and men said sort of: M = 2.84; Cohen’s d = 0.48.
So were Eagly and Wood just wrong? Well, not quite. It turns out that this massive sex difference in (not at all casual) sex offers completely “evaporates” (i.e., lowers to d = 0.17) when you control for “sexual capabilities.” The sexual capabilities covariate contained two items: “the proposer would be a great lover” and “would provide you with a positive sexual experience.” Keep in mind, these items were rated by the subjects just after deciding how likely they would be to have a sexual encounter with the person! If you control for two tightly correlated variables any effect disappears. The gap in male and female pay completely disappears when you control for testicle number... So how tightly correlated is “likelihood of agreeing to sex” with “would provide a positive sexual experience”? We don’t know; it’s not reported. JPSP. Seriously. (I pulled out my Mastercard and made a $100 donation to the Discovery Institute...the hell with it.)
Think for a moment about what it would mean for the measure of "likelihood to have a sexual encounter" to be divergent from "belief that the person would provide a positive sexual experience." What would be required would be a woman willing to say something like this: “I would definitely have sex with him, and I’m sure he would provide a thoroughly negative sexual experience.” Or, “I would definitely not have sex with him, but if it did happen anyway I’m sure it would be a thoroughly positive experience.” Only these two types of women remain to generate variance in the regression analysis that “evaporates” Clark and Hatfield’s data. We’re basing our lack of sex difference on the existence of women who either desperately want to have terrible sex or kind of enjoy being raped. If these women don’t exist in high numbers, then the regression is meaningless. Now I suspect few if any of those women exist. But then again I thought men and women had different attitudes about casual sex, so what do I know?