How gay are American football players?

Every week, the Anti-Scientist picks a study or news report that was stillborn from the start. Then he resurrects it and makes it do a jolly dance!

How gay are American football players?

Eric Anderson of the University of Bath, Center for the Tedious Verification of the Self-Evident, has determined that college-level male cheerleaders who played football in high school, are likely to have engaged in “acts intended to sexually arouse other men.”

This is news how? Still, we have it, the neon factoid, nineteen of forty-seven footballers sampled “had sexual relations with men.”

To those of us it is intended for, this study is almost as gratifying as a post-game towel-whipping. But the sex appeal of the hypothesis is not the only reason the piece will be difficult to take down. It is all about spin, and unfortunately, not only is the study as methodologically spurious as a quadruple reverse, it is as slippery as a sweaty halfback. The study employs several shady maneuvers against debunking. Chief among these is that Anderson dodges his own paraded thesis. The data is positioned to inform us of a high incidence of homosexual activity among high school football players.

But explicitly, Anderson draws only the softer, untestable conclusion, that these cheerleaders’ confessions indicate a decreasing prevalence of homophobia among young male athletes. Various media will then be trusted to take the ulterior thesis to the moon, as they have already begun to.

But how are we to say that this is intentional on his part? The U of B press release: “He said the study was not biased by talking to sportsmen who were now cheerleaders . . . Those he interviewed were selected to represent men that considered themselves traditionally masculine, typical American Football players.” The study is forthcoming, but we will assume that it is only as willfully distortive as the press release from the institution that sponsored it. Headline of that press release: “Over a third of former American Football players interviewed had sexual relations with men, says study”

On the methodological front, the sample-size of this study is only 47 people. Only in rude circumstances, archaeological for instance, are such piddling sample-pools legitimate. There are roughly 1.2 million high school football players in the country. So stop being coy. Take more than a four-hour sabbatical. Talk to several thousands.

But perhaps most importantly, the sample is confined to a self-selecting group, college male cheerleaders. This move is clever because it settles the burden of debunking on the inverse premise, the contentious position that male cheerleaders more often engage in same-sex partnering than your average football player. That position immediately becomes a political one fraught with impolitic implications about the nature of masculinity itself. But by flatly refusing to interrogate the overlapping category, cheerleaders, and by smirkingly assuming it as a control group, the bombastic ghost-thesis is rendered null.

But the very bottom line is this: if he’d really wanted to determine whether “homophobia is on the rapid decline among male teamsport [sic] athletes in North America at all levels of play,” he might have asked even a single one of them, and asked twice. Any high schooler, whether he plays football or not, can tell you that the the calculation of change requires the comparison of a single variable at two positions along the axis of time.

The study will appear in the (avowedly programmatic) journal Sex Roles in January.
~Alan Bajandas