So yesterday I claimed that the actual methodology doesn’t matter that much if you want to discuss Gina Kolata’s The Myth, the Math, the Sex in last weekend’s New York Times Week in Review section.

Turns out I was mistaken. Incredibly mistaken! I couldn’t be more delighted!

Before I get into that here’s quick point about my methodology: As you’ve probably noticed that contrary to the Mars-Venus/two-sphere model of explaining gender differences I try starting with the assumption that men and women start out virtually identical. Now I don’t really believe there really aren’t differences between men and women. But I certainly believe there aren’t as many as the dominant paradigm tells us there are. And, I assert, that dominant paradigm is men’s self-defeating but frantically defended conviction that women are the “no-sex” class, incapable of mustering sexual arousal on their own such that men (believe they) must actively manage women’s sexuality for them.

Turns out that, based on men’s and women’s respective reactions the Kolata article fabulously illuminate the “no-sex” class paradigm! Particularly just how male-centric it is. Check this out.

By an overwhelming margin men are attacking the methodology and defending the idea that, ding-dammit, women just aren’t as motivated to have sex. An awful lot of them are just dead sure that Dr. David Gale, the Berkeley professor emeritus of math the article is about, the reporter, her editors and fact checkers, and everyone else the reporter interviewed can’t tell the difference between statistical medians and statistical means. A few moments ago I did a Google “BlogSearch” with the keywords “median mean”. Virtually all resulting posts that came up were written by men.

Their objections to boil down to a reading (I would say misreading) of the following lines in the original article. Here’s how I read them (with emphasis mine.)

Surveys bear this out. In study after study and in country after country, men report more, often many more, sexual partners than women.

One survey, recently reported by the federal government, concluded that men had a median of seven female sex partners. Women had a median of four male sex partners. Another study, by British researchers, stated that men had 12.7 heterosexual partners in their lifetimes and women had 6.5.

But there is just one problem, mathematicians say. It is logically impossible for heterosexual men to have more partners on average than heterosexual women. Those survey results cannot be correct.

Source: New York Times Week In Review: Gina Kolata, Aug. 12, 2007

And here’s how all the (again mostly male) mean vs median bloggers read it (implied emphasis theirs.)

Surveys bear this out. In study after study and in country after country, men report more, often many more, sexual partners than women.

One survey, recently reported by the federal government, concluded that men had a median of seven female sex partners. Women had a median of four male sex partners. Another study, by British researchers, stated that men had 12.7 heterosexual partners in their lifetimes and women had 6.5.

But there is just one problem, mathematicians say. It is logically impossible for heterosexual men to have more partners on average than heterosexual women. Those survey results cannot be correct.

Got that? I read that multiple studies with multiple methods by multiple researchers return a consistent skew and agree with the conclusion that data collection is flawed. “No-sex” class victims see mention of one mention of medians near a mention of averages and, already predisposed to dismiss the actual story, decide that based on one admittedly weak copyediting decision the premise of the entire article is a crock, the reporter’s got a vagina, Professor Gale has a girly-sounding last name, and editors wear ballet slippers, and InstaPundit gets to perve out with his dingie little rubber stamp (“Indeed!”)

Bottom line: whereas the point of the article would stand unaltered if the “offending” paragraph were struck, all “mean/median” counter-arguments would fall.

Nope. If you’re stuck in the “no-sex” class paradigm you simply have to read it as an accounting error because it’s simply impossible that heterosexual men and women generally have the same number of sex partners.

When, in fact, that’s the most sensible explanation.

Not that there aren’t plenty of other possible explanations involving thought experiments about, say, desert islands with one man and 100 women. And then there’s this young respondent at Matt Yeglesias’s:

If promiscuous women die younger than promiscuous men, and thus become unable to answer surveys about their promiscuity, then it’s perfectly reasonable for the mean sexual partners of each gender to diverge.

Biologically I understand that it makes sense that promiscuous women die younger than promiscuous men…

Who thankfully was quickly corrected by someone else... who then went a bit too far the other way…

Given that we seem to have redefined “biologically” to mean “Based on no examination of any evidence whatsoever”:

Biologically it makes sense that men with many sex partners will die younger. Young men who are prone to risky behavior, like sex with lots of partners in the age of HIV, are probably more likely to be murdered or die in accidents…

Others go searching for answers such as sex tourism (a squintillionth of the male population) or overseas military deployment (just a couple percent of the American male population under most survey’s age-59 cutoff) or speculating that maybe the surveys weren’t accounting for non-heterosexual encounters (they were) or that men define “sex” differently from women (the surveys generally do account for that) or generating elaborate tables expounding on just how few women have to be total sluts to count for the discrepancies that would be there if Gale and Kolata really meant “median.”

But the bottom line is the vast, vast, vast majority of men who choose to respond to the question are just dead set on the idea that nope, the research has to be wrong because women couldn’t possibly be influenced by stereotypes to fudge their numbers down. (Note the paradigm even shows up in the disproportionate assumption that it’s probably men fudging their numbers up rather than women fudging them down. Despite at least some counter-evidence.)

Funny thing, though, is that while men are just all over defending perception over mathematics, to the extent women have been mentioning the study they’ve tended to view logical contradiction as obvious or else they mention that they’ve felt that such skewed numbers put pressure on them.

Counterexamples are welcome in comments, but I don’t think you’re going to find a lot of them.

Oh, one last thing, Jordan Ellenberg, an assistant math professor writing at Slate.com takes a more comprehensive look, agrees that the article could have been clearer about medians and means, but ultimately concludes…

In the end, then, Kolata is right. Studies that report these numbers should emphasize that the reported difference between men and women is an anomaly that can’t be taken at face value.

Now I still happen to think that in feet-on-the-ground terms the methodology itself is sort of irrelevant. In fact, given the reception inside the “no-sex” class paradigm, I don’t thing methodology, or data, or, for that matter, any underlying reality, is making any difference at all.