Medicine

Why sex surveys fail.

By Chris Rangel MD ⋅ August 13, 2007 ⋅  Email this post ⋅  Print this post ⋅ Post a comment

The “observer effect” is what happens when the very act of measuring something changes the properties of what is being measured. This is common in physics as well as sociology where people tend to change their behavior if they are aware that they are being observed (much like the waving idiots in the background of any given local TV broadcast).

Yet it’s surprising how often we forget about this observer effect when it comes to such measurements like surveys. For example, surveys about sexual behavior in western society routinely find that men report having several more lifetime sexual partners than women. Well, duh! Everybody knows this. These results fit perfectly with how we think of gender differences i.e. men as being more sexually aggressive and adventurous and women tending to be more pure and sexually innocent. The problem is that mathematically these results are impossible.

Simply put, if women in a given population had an average of X number of sexual partners and men in the same population had an average of X + Y number of sexual partners then the mathematical question is . . where did this Y number come from? The average number of sexual partners in a given population should be statistically the same! There are many possible explanations;

1. The results are skewed by a few massively promiscuous women (like prostitutes or spring breakers) who were not part of the survey. However, this would mean that the vast majority of those extra “Y” numbers are from visits to prostitutes and Daytona Beach and this is highly unlikely.

2. The results are skewed by men going outside the study population (to other countries) to have extra sexual partners. have you been on a sex tour of southeast Asia lately? Again this is highly unlikely.

3. A significant number of men in these surveys are closet homosexuals who lied about their sexual orientation but included their gay partners in the survey tally. Yea, right. Be gay. Be proud! Unlikely.

4. A majority of men surveyed actually believe that they have had more sexual partners then they really did. This one is plausible for a certain number of men who are legends in their own mind but not for the vast majority. Unlikely.

5. Filling out the survey in the presence of an interviewer changed the behavior of those surveyed causing men to tend to inflate the number of their sexual partners while women tended to deflate the number. Ah, the “observer effect”. The 1999 US government study was conducted by “detailed, in-person home interviews” and obviously this changed the results.

This study says more about how people answer personally sensitive questions in the presence of an interviewer than it does about the reality of their sexual behavior. It also says a lot about what is perceived as socially expected and acceptable as far as differences in sexual behavior between men and women. In the presence of an official government researcher, study participants tended to answer with what they believed was acceptable and/or expected, i.e. that men should have more sexual partners and that women should have fewer.

The reality is likely that men have fewer sexual partners than we expect and women have more and Americans probably still have a very inaccurate and incomplete picture of our sexual behavior because we still rely on this survey method in order to gather data.

From Mr. and Mrs. Smith;

John Smith (Brad Pitt): “How many? Ok… I’ll go first, then. I don’t keep exact count, but I’d say, uh, high 50s, low 60s. I mean, I know I’ve been around the block an all, but… ”
Jane Smith (Angelina Jolie): “312.”
John Smith: “What? How?”
Jane Smith: “Some were two at a time.”