Gang bangs can alter the study. On that one given night the woman is promiscuous, but are the men?

If there were two women in the world, and one man, and the man had sex with both women, and the women don't have sex with each other, then the man has had twice as many sex partners as the women have had, right? Each woman has had 1 sex partner, and the man has had 2 sex partners.

I always assumed that the stats were explicable that way, but statistics are not my strong suit.

The Times methodology is deeply flawed, failing to distinguish that while the means must be the same, the medians (which is what was reported in the study) can be quite different. See explanation here: http://justoneminute.typepad.com/main/2007/08/they-cant-mean-.html

The numbers would work if there are seven, super promiscuous women out there who are each having sex with every single American man. Since we are talking median here, and there are millions of American women, these seven super sluts would have no discernible impact on the median number of partners for women. Though, if this is indeed what's happening, I'd think the rates of STD's would be higher.

- Andy

There are alot of assumptions that aren't being stated here. The first thing you need to know about any statistical study is how they define their population. Are they truly surveying a representative sample of the sexually active individuals out there?

Let's say you've got a crew of um, ...friendly, women out there. Are they at home to receive the researcher's call? Likely no. Too busy out fornicating, natch. Likewise the lonely wallflower, at home with her Bible and knitting needles. Probably only too happy to speak about how she's saving it all for her one tru wuv. (And yes, I'm aware that the 'we're not at home to get the phone calls because we're more fun' routine is straight out of the republican playbook, but it actually could apply here.

So in short, these researchers are a little too quick to discount men going outside their sample population to what they call the 'prostitute' population, as quite likely, this is where the logic falls apart. Stated another way, you'd have to make sure that each of their purported partners was inside the study population in order to support the claim they are making.

From a scientific point of view, the only way a researcher could truly prove a difference in male/female promiscuity one way or the other would be to corral a large group of people (1000+) on a research farm somewhere, and radiolabel the guys' sperm or something, then follow the evidence. Obviously something like that is out of the question, from an ethical standpoint, although I'm sure someone could find willing participants. And even then you'd have a hell of a time separating out effects of behavioral modifications due to the lab environment.

Or *shrug*, you know, we could just stop obsessing so much about each other's sex lives.

http://crookedtimber.org/2007/08/13/another-post-on-dating-strategies/

"The only thing is that it isn’t logically impossible, at least as the author presents it. Ask Andrew Gelman

Jeff’s response: MEDIANS??!! Indeed, there’s no reason the two distributions should have the same median. I gotta say, it’s disappointing that the reporter talked to mathematicians rather than statisticians. (Next time, I’d recommend asking David Dunson for a quote on this sort of thing.) I’m also surprised that they considered that respondents might be lying but not that they might be using different definitions of sex partner. Finally, it’s amusing that the Brits report more sex partners than Americans, contrary to stereotypes."

Foolish math. Not all women under report their partners, just the really-really active ones. So, to use the prom example, assuming the guys accurately report their numbers of dance partners, its the under-reporting by the drunk girls which creates the discrepancy.

The average is not the same as the median. The average number of sexual partners of men and women does need to be the same, by definition, as the editorial is asserting. However, if a small number of women have sex with lots of men, while most women only have a few partners, the median for women may be low, much lower than the average. Therefore, the mismatch between the median number of partners for men and women, if accurate, would reflect different distributions of sexual activity in male and female populations.

The key phrase is "if accurate." I don't know if the original study strictly ruled out partners of the same sex, since I've been told that some people consider themselves "heterosexual" even if they have sex with partners of the same sex. Also, I don't know how strictly or precisely the survey defined "sexual activity." For example, if a man and a woman have oral sex, the man may want to count her as a "sex partner" while the women may not count him, in order to keep their "numbers" in conformity with social norms.

Since there are roughly the same amount of women and men, on average the number of sex partners for men and women should be the same. This also takes in to account of gang bangs.

Let say 1 men sleeps with 10 women. So from the women's perspective, on average they have one partner. Now, if we assume that there are equal amount of men, that means the other 9 men didn't have sex. So from the men's perspective the average is also one partner.

So what does this mean for the study? Probably men lie about having more partners and women lie about having less partners. Let's take the happy medium, on avg everybody has about 5.5 partners.

The mathemetician's argument is flawed. one can reconstruct the total numbers of couplings by multiplying the number of men and the average number of partners but not when multiplying by a median.

a median is the number which divides a population into 2 equal halves. In a population of 10 men, the median can be 7 when

e.g 1) 5 men have had 6 partners and 5 had 8 or e.g. 2) 1 had 6, 1 had 8, 4 had 1 and the last 4 each had any number of partners over 7 (whether 8 or 80).