A new study from the University of Michigan is sure to get trotted out for years to come as proof that “real men have safe sex.” But to my eyes, what the study actually says is pretty unclear — and, in fact, it may not really say anything.
The study was about testosterone levels and a positive correlation to condom use in college-age men, but the headline on the Live Science article was as follows: ‘Macho’ Guys More Accepting of Safe Sex. It was then picked up by, GULP!, The Daily Mail, among other publications, all of whom reported it as de facto proof that macho guys are more likely to use condoms.
But the very notion that machismo is only about testosterone levels is insulting to me, as a proud collector of guns and swear words. Live Science shouldn’t be drawing an equals sign between testosterone and “Macho,” even if there is a vague (and squiggly, and sometimes dotted) arrow.
More importantly, Live Science’s article contradicts itself on the topic of what constitutes risk. That would be more or less irrelevant…if it weren’t the entire assertion of the paper.
The paper is “Safer sex as the bolder choice: Testosterone is positively correlated with safer sex behaviors in young men.” It appeared in Journal of Sexual Medicine, November 15, 2011. The title that seems to imply that safer sex is a risk taking behavior. But then the lead author on the paper, Sari van Anders, PhD, seems to imply that the importance of the study is that it shows a correlation between high testosterone and safer behavior, when she tells Live Science that “the finding may reveal that for young college men, insisting on safe sex could feel like a riskier move than unprotected sex.”
So, in fact, she seems to be saying that high-testosterone behavior is actually correlated to taking a non-risky action, except that it might be riskier than its opposite, which is why she can say “One of the things that is interesting about these results is that they’re one of the first to demonstrate a link between higher testosterone and less risk-taking in any domain,” in the same breath as establishing that using a condom is actually riskier, on an article titled “Safer sex as the bolder choice.”
Huh?
What is van Anders trying to say here? Are we trying to reify the idea that high testosterone = risky behavior? Because portraying safer sex as the bolder choice in order to preserve the idea of high-testosterone guys as risk-takers seems, to me, to be meeting the conventional wisdom more than halfway. It’s redefining risk on the fly, in order to generate a soundbite.
The moral of the story? Scientists who try to say “We don’t know for sure yet what this study means” by speculating on-the-record are almost guaranteed to generate soundbites that don’t actually describe their findings…especially when their paper has a soundbite-worthy title.
Studies are not gospel, and the news will garble anything. But more importantly, I have to call attention to what looks like van Anders’ apparent — and I do mean “apparent” — I don’t know what she really thinks — attempt to paint safer sex as a “more manly” choice.
Whether that choice is morally proper is not the question here…hell, I want to save lives, too. I want people to use condoms when it’s prudent to use condoms, and have sex when they feel like having sex. I want risky behaviors to be evaluated in the context of acceptable and reasonable risk, and I want as few unwanted pregnancies, new HIV infections and cases of other STDs as possible.
But tortured logic is tortured logic, and self-contradictory (or speculative) definitions of risk are not a helpful ingredient in risk-aware safer sex education. I have no problem with portraying rolling a condom on as a butch-as-hell activity, right up there with punching walls, guzzling bottles of Jim Beam and strapping rocket engines to the trunks of our Chevy Novas. But I don’t believe that the data says what the news is claiming it says.
To her credit, that may not be the fault of Dr. van Anders. She has done a lot of saying that “further study is needed,” and reminding reporters and readers that every study opens up questions more than it gives answers. I don’t want to come down too hard on Dr. van Anders, especially since her other research implies she’s strongly committed to a pluralistic view of science and sexuality. Dr. van Anders is not at all the kind of sexual determinist that usually makes The Daily Mail do a little dance by telling them that watching porn will make your head explode.
But again, tortured logic is tortured logic.
Anyway, here’s what Live Science says:
‘Macho’ Guys More Accepting of Safe Sex
In a turnaround on the usual stereotype of macho, testosterone-laden guys making risky life choices, a new study finds that young men with higher levels of this sex hormone are more likely to accept safe-sex practices.
The research focused on a population of 18- and 19-year-old men in their first year of college, a time when many are just beginning their sexual lives, said study researcher Sari van Anders, a behavioral neuroendocrinologist at the University of Michigan, Ann Arbor.
The study found that men with higher testosterone levels were more likely to have an accepting attitude toward condoms and protected sex. The finding may reveal that for young college men, insisting on safe sex could feel like a riskier move than unprotected sex, van Anders told LiveScience.
“There’s this body of research showing that people often view safer sex behaviors and people who engage in them in a somewhat negative light,” van Anders told LiveScience. Thus, she said, the “social risk” of insisting on using a condom might require more boldness and confidence than having unprotected sex.
Testosterone is linked to boldness and confidence, as well as in perilous decisions, such as making high-risk financial bets. But the concept of risk is culturally defined, van Anders said. In the case of sex, the obvious risk would be of pregnancy or sexually transmitted infections (STI). But for some people, those risks seem far-off and unlikely, she said, while the risk of having a partner think you’re untrustworthy or already infected if you insist on using a condom is very immediate.”Do you care more about the possibility of acquiring an STI, which might seem unlikely despite all these educational efforts, or do you care more about your partner right now, potentially thinking negative things about you?” van Anders said, adding that cultural ideas about sexuality can actually influence how hormones such as testosterone are linked to attitudes and behaviors.
[Link.]
What van Anders does, really, is propose two different ideas on how to interpret the data:
HYPOTHESIS A: Men with higher levels of testosterone are not inclined toward risky behaviors (despite “conventional wisdom” to the contrary) when it comes to safer sex. Perhaps this means men with higher levels of testosterone are not inclined toward risky behaviors when it comes to safer sex.
HYPOTHESIS B: Men with higher levels of testosterone are inclined toward risky behaviors as “conventional wisdom” suggests, but (counter to what one might expect) insisting on using a condom during penetrative sex is more risky than not using a condom, because when you insist on using a condom, your partner might think you’ve got an STD. Perhaps this means that men with higher levels of testosterone are inclined toward risky behaviors when it comes to safer sex…but having safer sex is actually a riskier behavior.
Offering two possible interpretations of the data is far from a hanging crime — except that the second possibility sounds to me like an attempt to rescue some sort of “conventional wisdom” that men with high testosterone levels are risk-takers.
More importantly, I see it as redefining risk on the fly — proposing a new definition of risk, after-the-fact, to explain the data.
That seems like a crappy place to start building hypotheses about sexual behavior; it’s dangerously close to the place where all science does is “react” to what people “already know,” or already think they know, about what constitutes male and female.
Here’s an even more appalling aspect of this study…it mingles those who have had sex and those who have not.
As part of a larger study on hormones and behavior in college freshmen, van Anders and her colleagues asked 78 men, who were mostly heterosexual and from high-income families, to answer questionnaires about their health, sexual activity and attitudes toward condom use and other safe-sex practices. Each guy provided a saliva sample, from which the researchers extracted a measurement of testosterone levels.
Of the participants, 46 percent had already engaged in vaginal or anal sex, while the rest had not. But because the researchers were studying attitudes about safe sex and not actual safe-sex behaviors, they were able to include even sexually inexperienced men in the study.
The claim that “because…they were able” is not an “if…then.”
Both virgins and non-virgins were included in the study because that’s who they had in the study…period. That doesn’t make it okay.
Does it compromise their data? In my view, absolutely. And I would be less inclined to say so if the article didn’t try to rescue the correlation by claiming:
The results were strongest for men who were sexually active, possibly because their attitudes were based on actual behaviors, van Anders said, though more research would be needed to test this.
We are talking about a study of 78 men. I’m presuming all of them are students at U of M, one of the most exclusive learning institutions in the country, so socio-economically and in terms of academics and intelligence, they’re all within the same ballpark.
But some are virgins, but some are not. They are “mostly heterosexual,” which means that others are “not heterosexual,” which means they’re mixing up at least two communities that have enormously different contexts for safer sex, pregnancy risk, and HIV risk.
Those things are not worth nothing. Performing an end-zone save by saying “the results were strongest for men who were sexually active” is pretty absurd, since to my mind the inclusion of sexually active and non-sexually-active men makes the study’s actual findings contextually weird…its gooey, undefined center can’t be saved in my estimation by suggesting sexually active men have the “strongest” instance of a correlation that I reject in the first place.
But then van Anders says:
One of the things that is interesting about these results is that they’re one of the first to demonstrate a link between higher testosterone and less risk-taking in any domain.
Except that she established earlier that one of the interpretations of this data is that insisting on condom use is more risky, not less!
The finding may reveal that for young college men, insisting on safe sex could feel like a riskier move than unprotected sex, van Anders told Live Science.
Is it risky or not risky? When interpreting data requires one to have it both ways, the interpretation is definitely suspect. Van Anders can’t redefine risk on the fly and then present the findings as “demonstrating” anything.
I applaud much of van Anders’ work and her apparent political leanings — most of which I strongly share. Her writing appears to show a feminist, pro-pluralist and pro-sex sensibility, and from everything I can see she’s a damned fine scientist.
But I gotta say, here I think van Anders is working with dangerous guesswork.
