Sexual Activity

David C. Howell

bar bar


Sexual Activity is Not Susceptible to Easy Measurement

Some very interesting data on the sexual activity of male and female subjects can be found in Chatterjee, Handcock, and Simonoff (1995) A casebook for a first course in statistics. New York: Wiley. They provide data on the reported number of sexual partners for 1682 males and 1850 females. These data can be found at SexualActivity.dat, where the first column contains the data for males, and the second column contains the data for females. The dependent variable is the number of reported partners.

These data are interesting for a number of reasons. If we simply decide to use the means as our measure of average number of partners, we find that men have an average of 11.72 partners, while women report an average of 3.32. We all know that there is a double standard, but this is ridiculous. Since sex is usually between one male and one female, how could there be such a huge difference between the two sexes? In something wrong, other than lots of lying on the part of males or underreporting on the part of females? For an interesting discussion of these issues, see a 1995 issue of Chance News, where I first came across the example.

Perhaps the mean is not really the statistic we want. If you look at the histograms for both sexes, you see that both distributions are extremely positively skewed. That skewness is probably doing a lot to distort the mean. (I have not provided the histograms here because they are so skewed that it is difficult to get a meaningful histogram from most statistical software.) Instead of the mean, we could look at other measures of central tendency. These are given in the following printout.

Here you can see that when we look at the medians, the two groups are considerably closer together, which results from the effect of eliminating the role of extreme values. Moreover, notice that the modal value is 1 for both sexes. Not only does this say that the most common number of sexual partners is the same for both males and females, but it also shows us that (modally), both sexes are monogamous.

We could go one step further and record the percentage of males and females who report having no sexual partners. Although the "word on the street" would suggest that this figure would be much higher for females than for males, the data do not support any such conclusion. In fact, 9.9 percent of females, and 9.8 percent of males (ignoring the missing data that result from entering males and females as adjacent columns) report having no partners. You can't get numbers much closer than that.

Where, then, do the peculiar data we started with come from? Part of it is undoubtedly reporting biases of males and females. Another, related, part may come from the extreme rounding (up) that occurs at higher reported levels of activity. I recommend looking at the Chance News, article for what it has to offer.

One final note concerns the exagerated accuracy of some of the data. The most obvious is the male who reports the most sexual activity, with 253 partners. Is there anyone out there that believes this number? Note that it is not 252 or 254, but exactly 253. Either we have some sort of wierd macho kind of guy who notches his bedpost with each new partner, or he just made that number up. (And if you had had 250 partners, do you really think you could keep track of which one was a new one?)

bar bar


Home Icon Return to Dave Howell's Statistical Home Page  



Cows Icon University of Vermont Home Page  

Send mail to: David.Howell@uvm.edu)

Last revised: 06/19/01