Discrepancy Found in US Sex Survey

WASHINGTON (AP) â€” Surveys measuring the total number of sex partners among men and among women have for years suggested that men had more partners than women. <br><br>This result has puzzled social scientists

Thursday, October 12th 2000, 12:00 am

By: News On 6

WASHINGTON (AP) â€” Surveys measuring the total number of sex partners among men and among women have for years suggested that men had more partners than women.

This result has puzzled social scientists since the surveys should show equal numbers of partners because each time a man has a new partner, so does a woman. A new sexual partnership should add one to each side of the equation.

A new study may explain the flaw: The surveys failed to measure the sexual activity of prostitutes, thus reducing the number of sex partners reported on the women's side of the equation.

``The number of partners that (heterosexual) men have had must be equal to the number of partners that (heterosexual) women have had,'' said Devon Brewer of the University of Washington. ``Each new partner for a man is also a new partner for a woman. So, in reality, it must be equal. By definition.''

But the General Social Surveys, conducted by the University of Chicago, and the National Health and Social Life Survey, funded by private foundations, found that men were claiming up to 74 percent more partners than women.

The government uses these data to design public health programs to combat sexual diseases.

Brewer said social scientists, scrambling to explain the embarrassing inconsistency, suggested two possible solutions to the discrepancy â€” survey subjects were lying, or a fundamental flaw existed in the way the data were being collected.

``One explanation was that men are boasting or bragging about their number of partners and that women were being modest,'' said Brewer.

That may be true, he said, but a study he co-authored in the Proceedings of the National Academy of Sciences says the biggest cause of the discrepancy is that the surveys ignored the professionals: women who sell sex for profit.

``We found these high-activity women, prostitutes, were inadvertently excluded by the design of the surveys,'' said Brewer, whose study appears Tuesday in PNAS.

Brewer said the national surveys covered ``households'' and not dwellings like jails, motels, shelters and rooming houses, where prostitutes are more apt to work. Surveyors usually rang doorbells at evening, on weekends and holidays, when most prostitutes are working, unavailable for interviews, Brewer said.

To test his theory, Brewer's group used other studies to estimate about 23 prostitutes for every 100,000 people in the United States and an average per-prostitute client list of 694 male sex partners a year.

``Some have far more and some have far fewer, but that is a representative sample,'' Brewer said.

Applying these estimates to the national surveys brought final numbers for sexual partners into about equal balance, said Brewer.

Male prostitutes were not included, he said, because few women buy sex from men.

Tom Smith, director of the General Social Survey at the University of Chicago, said Brewer's study ``offers a likely explanation for at least part of the discrepancy,'' but there are other factors.

Smith said that about 10 percent of the discrepancy may be because the survey did not seek responses from sexually active juveniles. This would create a bias, for example, because males over 18 might report sex with young females, but the survey would not include balancing reports from underage girls.

Also, Smith believes that another 10 percent of the bias may come from gender-based attitudes toward sexual surveys, with men overreporting, women underreporting.

The GSS, first published in 1988, is paid for, in part, by the National Science Foundation. Smith said the Centers for Disease Control and Prevention use the data to help design and target public health campaigns to control sexually spread diseases.

â€”â€”â€”

On the Net:

http://www.eurekalert.org/releases/uwash-auo100600.html