Interviews: Sex and the American Teen

Dr. Peter Baerman

Dr. Peter Baerman: Do Virginity Pledges Work?

POV: You’ve done two studies of virginity pledges, based on the data in your survey, the National Longitudinal Study of Adolescent Health (Add Health). In your first study, you found that taking a virginity pledge had some delaying effect for many adolescents, but that certain conditions applied. Could you talk about that?

Dr. Peter BearmanDr. Peter Bearman: The first project was published in 2001. When we controlled for all the usual determinants of what we call “the transition to first sex,” we were able to show that taking virginity pledges delayed sex by about 18 months. We also found that the delay effect worked for some kids but not all kids. It worked for kids in mid-adolescence, not young adolescents or older adolescents. If there were no pledgers in a students’ community, taking a virginity pledge had no effect. And if there were too many pledgers in a student’s community — that is, more than 30 percent — pledgers didn’t benefit. Pledging works when it embeds kids in a minority community, when it gives them a sense of unique identity. And it doesn’t work when it’s a national policy that everybody follows. If everybody pledged, pledging would have no effect.

Danny, Paula, and Shelby Knox at Shelby's pledge ceremony Danny, Paula, and Shelby Knox at Shelby’s pledge ceremony

Then, of course, many kids have sex whether they pledge or not, and pledgers [who broke their pledge and had sex] were much less likely than non-pledgers to use contraceptives. So the benefits of delaying sex wash out, because of enhanced risk. Kids likely do benefit from delaying sex. But from a public health point of view, the pledge doesn’t reduce pregnancy or STD acquisition rates for adolescents.

POV: Your more recent project, just published in the Journal of Adolescent Health, involved following up on the teens from the original Add Health study. What did you learn about the longer-term effects of taking a virginity pledge?

Bearman: We looked at the consequence of a virginity pledge on the rates of STDs. Although pledgers have slightly fewer partners than non-pledgers [on average], and are more likely to be married at a young age than non-pledgers, pledgers have STD rates that are statistically the same as non-pledgers. There are three reasons for that.

The first reason is, they are less likely to use condoms [when they first have sex]. Condom use at first sex is a huge predictor of condom use subsequently. So the fact that pledgers don’t use condoms the first time they have intercourse has this long-term consequence. Secondly, pledgers are less likely than non-pledgers to think they have an STD when they have one; they are less likely to see a doctor to get diagnosed for an STD; and they are less likely than non-pledgers to get treated for an STD that they do have. And then the third reason is that kids who took virginity pledges and remained virgins were more likely to engage in what we call “substitutional sex” — including acts that can put them at higher risk for STDs, such as anal and oral sex.

POV: Are there some other characteristics or causes that might explain the differences between pledgers and non-pledgers?

Bearman: Pledgers are more likely to be religious than non-pledgers, and religious kids are more likely than non-religious kids to delay sex, anyway. Pledgers are more likely to come from two-parent intact middle-class households, and kids from two-parent middle-class households are also more likely to start having sex at a slower rate than other kids. But you can control statistically for these characteristics and still discover that pledging has an effect.

POV: Are there other characteristics that distinguish pledgers from non-pledgers? Are there differences between pledgers who are totally abstinent and pledgers who engage in other kinds of sexual activity?

Bearman: Just to take a pledge means that in some fundamental way you’re thinking about sex. Twelve-year-olds who take virginity pledges are thinking about sex in a different way than twelve-year-olds who are playing in the backyard, and therefore not thinking about sex at all. The interesting thing about pledgers is that they are more romantic than non-pledgers — pledging is built on an ideology of romantic love. Pledgers are also more likely to be in romantic relationships than non-pledgers. So they are kids who are actively thinking about the world of intimacy, and the pledge is a rhetorical device that helps them negotiate the grey zones of that world of intimacy in a very easy manner. It allows them to say, ‘Well, I like you, but I don’t intend to have sex.’ So kids who find it difficult to talk about intimacy, for example, benefit from the pledge because it draws a firm line for them.

As far as pledgers having substitutional sex, one idea is that they took a public pledge to remain a virgin and the thing that they’re fearful of is getting pregnant — which is the clearest sign of violating the pledge. So if you’re trying to avoid getting pregnant, which is a mark of having sex, you might engage in other kinds of sex activities. But of course, the thing about STDs is that you can’t see them. So [these substitutes] seem safe, but obviously they’re not.

POV: What are the implications of these findings for parents and policymakers? What can they take away from your findings?

Bearman: Pledging works for some kids in some contexts. There’s absolutely nothing wrong with being abstinent; in fact, it’s a great thing for public health. So, if pledging is useful for kids, they should do it. The problem is that eventually, pledgers and non-pledgers alike are going to have sex, and some pledgers who have sex and don’t protect themselves put themselves and other people at risk. The sex that pledgers eventually have is riskier, because they are less likely to use condoms. It’s really important that everybody have the information that’s necessary to protect themselves from the negative consequences of sex, which are STDs and unwanted pregnancy. And [on the whole] pledgers don’t get any benefit with respect to those risks. So, as a national policy, it doesn’t really impact public health.

POV: Research on adolescent sexuality, and particularly on virginity pledges, has provoked a great deal of political argument. Does such controversy make it more difficult to do good research?

Bearman: It doesn’t make it more difficult, but I find the comments by so-called abstinence-only supporters offensive. People who have no scientific credentials should in general refrain from assessing whether science is done properly or not. Leslee Unruh from the National Abstinence Clearinghouse, for example, has called the work that we do ‘bogus’ and ‘lacking scientific credibility.’ When they agree with the results, they celebrate the science. When we came out with the result that the pledge delayed sex, these same groups that are criticizing us today put that result all over their web pages, and established that study as the most scientific study ever. These are the same data, the same researchers, the same standards, so I find the politicization of this issue offensive. It also just makes it unattractive as a research area.

POV: The National Abstinence Clearinghouse claims that you are “twisting the study’s results to fit” an “ideological agenda,” and argues that your results actually demonstrate the opposite of what you’ve described. How do they reach that conclusion from your study?

Bearman: They’re just misrepresenting data in a really fundamental way. For example, it’s well known that STD rates vary significantly by race. Blacks, for example, have six times higher STD rates than whites. So any analyses that you do need to be separate for blacks and whites. [The NAC] looks at the overall STD acquisition rate for pledgers and non-pledgers, and they see that it looks like pledgers have lower STD rates than non-pledgers. When we say that these rates are statistically similar, the lay language is that the estimates are within a margin of error that overlaps. So when the two ranges overlap, for pledgers and non-pledgers, there’s no difference. So [groups like the NAC] find little pieces of data and misrepresent them. And they should know that that’s irresponsible. If we had results that agreed with them, they wouldn’t do that. Just as we had results that they liked four years ago.

UPDATE: The final findings of a recent study about the impact of abstinence-only education and virginity pledges were reported in April 2007 and found that the programs had “no effect on the sexual abstinence of youth. But it also finds that youth in these programs were no more likely to have unprotected sex, a concern that has been raised by some critics of these programs.” Download a PDF of the report at the Mathematica Policy Research website.
Here are links to some articles that appeared about the report.
No More Virginal: Spend $1 Billion Dollars on Abstinence Education. Get Nothing, Slate magazine, April 20, 2007
– Updated May 10, 2007

POV: What you’re saying is that the STD rates were statistically identical, right? Could you elaborate on that point?

Bearman: Well, let’s say that we do a political opinion poll for a presidential race, and we discover that 48 percent of the population would like candidate A, and 52 percent of the population would like candidate B, with a margin of error of 6 percent. So then we would say that candidate A is preferred by 45 to 51 percent of the people; any value in between there is equally likely. And candidate B is preferred by 49 to 55 percent of the people, with any value in between there equally likely. And you can see that there’s an overlap of values. So from a statistical point of view, those confidence intervals overlap, and so there’s no difference between them. The estimates of STD infection for pledgers and non-pledgers overlap completely, so there is no significant difference between them.

POV: What kind of research remains to be done on this subject? Does your work suggest any particular avenues for further inquiry?

Bearman: One of the things that we know is that pledgers get married younger than non-pledgers. It’s too early to see whether that leads to higher fertility, or greater divorce, to happier marriages or sadder marriages. It’s too early to see what the long-term consequences of pledging are. We know that people that marry very young are more likely to get divorced, because they marry on the basis of romantic love, or they grow differently, or for whatever reason they’re not ready. So actually, there’s a whole set of interesting studies that someone could do in four or five years.

Dr. Peter Bearman is director of the Institute for Social and Economic Research and Policy and the Paul F. Lazarsfeld Center for the Social Sciences at Columbia University, where he is also chair of the Department of Sociology. With J. Richard Udry, he designed and directed the National Longitudinal Study of Adolescent Health (Add Health), the largest, most comprehensive survey of adolescents ever undertaken, including 20,000 adolescents aged 12 to 18. His most recent article on adolescent virginity pledges appears in the April 2005 issue of the Journal of Adolescent Health.