He provided a proof, called the High School Prom Theorem:
"Each girl is asked to give the number of boys she danced with. These numbers are then added up, giving a number G. The same information is then obtained from the boys, giving a number B. Theorem: G=B. Proof: Both G and B are equal to C, the number of couples who danced together at the prom."
Researchers speculate that one of two things is happening, either the men are going outside of the interviewed population for their sexual escapades or both genders are simply lying. I'm sure it's a bit of both.
A couple of things bothered me about the article. First, the article interchanged the words median, mean, and average continuously, which I find quite irritating, since they are not necessarily the same. Second, although I can clearly see why the arithmetic mean number of copulations must be the same for both genders of a heterosexual population, I'm not sure that really tells us anything about relative promiscuity.
Let's say we define promiscuity as having more than one sexual partners in relative simultaneity. Then we imagine a graph of heterosexual pairings that reflects the common alpha-male situation. There are 10 men and 10 women. One studly/slutty dude hooks up with 4 of the women. Two other men and women engage in monogamous coitus. The rest go home alone. If we use a simple arithmetic mean, then both genders engage in 0.6 sexual encounters. However, that one guy accounts for 2/3 of the men's numbers. If you randomly selected a guy from the room, there is a 1/10 chance of him being promiscuous, by my definition. The same cannot be said for the girls, as none of them had more than one partner. So in this sense, men can be said to sleep around more.
You could translate this into a weighted average as well. If you assign weights of 0 to not promiscuous and 1 to promiscuous, then the men achieve a weighted average of 0.4 and the women of 0.0. This ranking puts the men ahead in the slut department.
It's important when applying mathematical computations to real life situations that the math is not only correct, but non-trivial. And the method you use can significantly affect the outcome, as is evidenced here.