Sex, lies and statistics...

"There are three kinds of lies: lies, damned lies, and statistics."

Mark Twain

Not all things can be a gender issue. After seeing an article claiming that women were more promiscuous than men I had to laugh. I laughed again when I saw one claiming the opposite. Where the articles were written and who took the time to carry out the "research" is immaterial, so let us leave those issues aside. Also, since most of these articles are driven by an agenda to either showcase men as womanisers or women as "manisers", let's only consider the heterosexual population.

My argument is the simple fact that, on average, no gender can be any more promiscuous than the other. Let's just consider the population of an imaginary world called Miscu. Its population is composed of two genders, the "pro" and the "iti".

In Miscu, each sexual union "is" a real union, for establishes an unbreakable bond between the participants.

In the above example, both individuals have engaged in sex with one other individual. The sexual bond is not directional so both individuals, and hence genders, are equally promiscuous. Since the number of men and women in our own world is approximately equal, both genders must be equally promiscuous by definition. If you don't believe my simple example above, look at this more complex one below, where 10 individuals are involved...

In the example above, the number of links between individuals (9, precisely) remains the same for both the "iti" and the "pro" genders. Both genders are still equally promiscuous, on average, meaning that the mean is still meaningless. This leads to a more interesting point though: despite the fact that both genders have an equal number of sexual partners, the distributions of these bonds in each gender are different in a number of different ways...

• The promiscuity of the "pros" is less variable, ranging between one and two partners compared with a range between zero and five in the case of the "itis".

• The median number of sexual partners (i.e. the middle number in the distribution) is smaller for the "pros" (2) than for the "itis" (3).

• The mode (i.e. the most common number), on the other hand, is greater for the "pros" (2) than the itis (0).

• I won't even start writing about the skewness and the kurtosis, as by now you'll probably have seen my point :-)

This point is that statistics at its simplest can be a dire tool of deception and delusion, but if used properly, can give us much information. So let's just go beyond our simplistic gender biases and past the typical "itis are more promiscuous than pros" headline and try get a little closer to the truth...