Log in

No account? Create an account
entries friends calendar profile King Rat Previous Previous Next Next
Number of partners - King Rat
Private Life
Number of partners
The New York Times asks a question I've been wondering about for a while, and talks to mathematicians who point out that men can't have, on average, more partners than women do, as many many surveys seem to say.

Are men making partners up?
Are women not reporting some partners?
Are men having sex outside the survey pools?

The answers to those questions are not known however.
8 comments or Leave a comment
differenceblog From: differenceblog Date: August 13th, 2007 04:00 pm (UTC) (Link)
It’s hard to believe that no one is looking at the activity distributions. As I discuss in my own post today, Gale is talking about means whereas the CDC is discussing medians. These are only the same under a symmetrical distribution
gkr From: gkr Date: August 13th, 2007 04:07 pm (UTC) (Link)
This is true. But pretty much every survey I've ever seen reported on, both those that report the median and mean, shows similar results. While the medians can be very different (as would happen if the high-partner women were not surveyed, e.g. prostitutes), the mean should be very similar. So while the apples-oranges comparison here is not quite appropriate, I'm glad someone is asking the question. Cause something ain't making sense generally.
differenceblog From: differenceblog Date: August 13th, 2007 04:31 pm (UTC) (Link)
That's interesting -- I haven't found any studies that report both by gender . I'd be happy to look at them, if you can point me to one.
schmooops From: schmooops Date: August 13th, 2007 04:08 pm (UTC) (Link)
I wonder if it's because surveys like this typically report the median number of sexual partners, rather than the mean. If the median number of sexual partners that women have is 4, that means that half the women surveyed have fewer than 4 and half the women have more than 4. Those who have more than four could have had five, ten, a hundred ... which would certainly help to explain how the men were getting laid. Because definitely, assuming that all the sexual contact is male-female in the number of sexual partners, on *average* (not median), the numbers of sexual partners should be the same.

In the absence of good math, I guess we just have to go with this explanation: “Some might be imaginary,” Dr. Graham said. “Maybe two are in the man’s mind and one really exists.”

malackey From: malackey Date: August 13th, 2007 05:00 pm (UTC) (Link)
Maybe the men are having sex with each other. Would explain the disparity between the number of partners reported by men and women.
laurelfan From: laurelfan Date: August 13th, 2007 10:32 pm (UTC) (Link)


One is that there are more women than men (if there is one man and two women, and he has sex with both, the man's average is 2 and the womens' average is 1). Since women have a higher life expectancy than men this is true, but I'm guessing the difference is too small to explain this.

Another one is that they throw out 0s (or they self select out). If you're advertising for subjects for a sex survey people who have never had sex might not respond. If you throw out the 0s it's possible for the averages to be different (ie. polygamy where men have multiple wives but women don't have multiple husbands)
insolent From: insolent Date: August 14th, 2007 12:51 am (UTC) (Link)
Thanks for the link, I found this very interesting.
vulture23 From: vulture23 Date: August 14th, 2007 01:51 am (UTC) (Link)
(cut and pasted from my reply to someone else's post on this very article... :) )

I am frankly very disappointed at the supposed math professor who gave an entirely invalid proof. The scenario being presented in his "proof" is entirely unlike what is actually involved in a survey, because of sampling. There are quite a few ways in which sampling error could easily change the results of a survey (the "prostitute effect" is but one of them, and many could affect the mean as well as the median), but that's not what they're saying. The fact that they are also referring to the medians, which do *not* need to be anywhere near equal if the distributions are not similar, makes their claims even weaker.

Yes, it is unlikely that, even with all of these effects added together, they would explain such a large disparity in median. However, the arguments provided are not relevant and do not actually prove anything.
8 comments or Leave a comment