Saturday, January 31, 2015

Sex partners over time among men and women

Game's golden tongue on self-reported sex surveys:
The problem with sex surveys of this sort — i.e., the kind that ask in no uncertain terms just how slutty (women) or charming (men) you are — is that people lie. They lie to assuage their egos, and they lie to meet social expectations. And lo and behold, there are attenuating studies which discredit to some extent the reported results of sex surveys.
It's a point Heartiste Chrysostom has put pressure on several times before, and the finding that men average 2-3x as many lifetime sexual partners as women do has, as its prerequisite for being true, a world in which women outnumber men by a similar factor in the general population. That isn't the world we inhabit, of course, and consequently male self-reports, female self-reports, or both, are unreliable.

But just because women low-ball their counts and men inflate theirs--despite his sworn testimony, Bill Clinton probably would've counted Monica Lewinsky as a partner for the purposes of a survey like this while she likely would not have--doesn't mean data like these are worthless. There's a signal mixed in with the noise. It's difficult to definitively detect the former and separate it from the latter, but it doesn't mean that there is no value in giving a listen. Like so many other areas of human experience, the truth is stretched and excused but rarely ignored outright in perpetuity.

To the question of changes in sexual behavior over time, for example, social expectation bias suggests that we should see some increase in the number of reported partners among women as we progress forward in time due to the decrease in societal slut-shaming and the corresponding steady drop in marriage rates even if there is no change in the actual average number of partners women have today compared to what women tended to have in the past.

So without further ado, the following graphs track the percentages of white men and women aged 25-40 in lifetime sexual partner ranges over the nearly 25 years that the GSS has been querying respondents on as much. Subsequent tables show the mean, median, and modes for each year's total responses. First, men:

Female partner averages for white men aged 25-40 by year:

201227.67 (!)51

Parenthetically, mean = total instances/number of samples (what we conventionally think of as "average"); median = the value at the 50th percentile of the sample distribution; mode = the most frequently occurring value in the sample.

(!) The 2012 mean is skewed by one respondent claiming to have had over 989 female partners, the only year in which Adonis was available for survey participation. Removing him from the year drops the mean to a more terrestrial 17.15.

There is very little in the way of any detectable secular trend here, although there are faint empirical traces of a bifurcating male sexual market in which the percentages of men earning 20+ notches and the percentages of 40 year-old virgins are both increasing modestly at the expense of men in the moderating middle.

And male partner averages for white women aged 25-40 by year:


The results for women are less sporadic than they are for men, and there is an apparent slight but perceptible increase in the number of male partners women are bedding over the last couple of decades. This could conceivably be due to a reduction in shame among women in admitting to having high partner counts, it could be a result of real increases in novel sexual encounters in the US, some combination of the two, or something else.

Heartiste writes:
I’d add that, despite the above GrateFacts, it’s a good bet that lower-N count monogamy is still hanging on as the norm among Eurasian peoples. Well, serial monogamy, at any rate.
As the medians and modes show, that is indeed still the case.

Finally, 2012 was an especially lascivious year for GSS participants. The 2014 data will be released in the next couple of months. It will be interesting to see if 2012 turns out to be an aberration or the seminal year of an enduring trend towards more faux inseminations.

GSS variables used: YEAR, SEX, RACE(1), AGE(25-40), NUMMEN(0-989), NUMWOMEN(0-989)


JayMan said...

The problem with this particular set of noisy signal is that we have little ability to clean out the noise from the signal.

That of course doesn't mean there is no signal, and hence self-report sex data cannot be completely dismissed off hand.

That said, the usefulness of such data depends greatly on the intended application.

Of course, this is better than what the game guys often have, which is basically built on glorified anecdotes with no sense of quantification whatsoever.

Jokah Macpherson said...

Given what I've read about Monica Lewinsky (probably from a Heartiste link back in the day), I think she would probably count Bill Clinton in a survey like this. It definitely looks like there was the perception of "something more" on her side even if Bill was just there for the blow jobs. Tellingly, she never moved on and found a guy to marry. Sure, the national disgrace is a lot of baggage to bring into a relationship but people have overcome worse. As Heartiste might say, "One minute of alpha..."

The Last Psychiatrist, on the other hand, pointed out that Bill Clinton had to become president just to get a girl the caliber of Monica Lewinsky, which speaks more for his standards than his charisma.

Jokah Macpherson said...

For what it's worth, I trust women's parter counts in surveys more than I trust their preferred traits for men in surveys. One place where the "game guys" have it right is focusing on observable behavior since opinions didn't evolve to be mutually consistent and are often misleading or contradictory. (Not that they won't lie about partner counts but at least this is a parameter based on facts not opinions.)

Audacious Epigone said...


Nothing to argue with there. Also, I stole the noise and signal descriptors from you, so there's a little late attribution, thanks!


Yes, given that he was arguably the most powerful person in the world at the time, any regret is probably going to be more than negated by the ego boost being pursued by the highest status individual on the planet is going to give a woman. Generally speaking, though, I think the well known example illustrates the point I'm trying to make.

Dan said...

You would expect a substantial upward trend because of the increasing fraction of unmarried men.

A single guy, even a low status guy, might get one or two hookups a year and rack up a high partner count over time even though their love life is shabby.

The fact that there is not a strong upward trend suggests we aren't in an era of extreme licentiousness.