Second of Two Articles
The first generation of research findings is emerging from the National Longitudinal Study of Adolescent Health, a federally funded, multimillion-dollar, school-based study designed to identify and assess the various factors that place adolescents at risk for a host of potentially health-compromising behaviors ranging from eating disorders to vehicular safety to early sexual activity. These studies and future research using "Add Health" data promise not only to help parents, communities and policymakers understand the factors that protect against or promote risky behavior among adolescents but also to point toward interventions that will ultimately improve the overall health of teenagers and, over time, the population at large ("The 'Add Health' Survey: Origins, Purposes and Design," TGR, June 2001).

At least half a dozen published reports and peer-reviewed journal articles based on Add Health data are beginning to shed light on the range of "risk" and "protective" factors associated with the politically sensitive area of teenage sexual activity, including the roles of virginity pledges and parent-child communication (see box). While some of these early findings have been seized upon by the popular press and by conservative interest groups looking to promote their ideology, they may ultimately provide more guidance to parents than to policymakers on how to help teens postpone sexual activity.

What Puts Teens at Risk?

One of the key purposes of the Add Health survey is not only to determine the prevalence of sexual intercourse—as well as a host of other potentially health-compromising behaviors—among teenagers but also to assess the various factors that place adolescents at risk for such activity. And as one would expect, results from the Add Health survey show that teens' reports of ever having had sexual intercourse increase dramatically with grade level, from 16% among seventh and eighth graders to 60% among eleventh and twelfth graders. Additionally, the Add Health survey reaffirms the findings of a large body of existing research that teenagers who are black or from low-income or single-parent families are more likely to have had sexual intercourse than their peers.

But one of the Add Health survey's greatest strengths is that it is designed to help researchers go beyond these demographic descriptions to identify the underlying social mechanisms that influence teen behavior. In other words, the survey aims to help researchers identify the forces within a community, family, school or peer group—as well as innate characteristics—that, alone or in combination, predispose a teen to high-risk activities and to analyze how these influences are expressed in different ways within various subgroups of teens. And, indeed, studies using Add Health data show that demographic factors play only a minor role in explaining teenage sexual activity and that there is tremendous variance—or heterogenity—even within individual racial, ethnic and income groups.

In fact, the research emerging from the Add Health survey strongly indicates that whether or not a teenager has ever had sexual intercourse is largely explained by that individual's own sexual history and his or her own perceptions about the costs and benefits of having sex. This stands in stark contrast to other major risk behaviors—such as cigarette smoking, drug and alcohol use, weapons-related violence, and suicidal thoughts and attempts—which the data indicate are shaped by more generic or external factors, such as frequent problems with school or work, frequency of "just hanging out with friends" and the number of friends who regularly smoke or drink. According to a major report published last year on teen risk behavior, Protecting Teens: Beyond Race, Income and Family Structure, when it comes to sexual intercourse, the most important risk factor across all racial and ethnic groups was having been in a romantic relationship within the previous 18 months, followed by (for most subgroups) "ever [having] kissed or necked." The most powerful "protective" factors for most subgroups were the perceived personal and social costs of having sex or getting pregnant or causing a pregnancy.

Robert W. Blum of the Center for Adolescent Health and Development at the University of Minnesota, and the author of Protecting Teens, explains that "sex doesn't really fit in with the other risky behaviors" that he and his colleagues examined using Add Health data. "While we tend as a society to lump sexual activity together with other risk factors, it is fundamentally different than drug use or weapon carrying—behaviors we hope to prevent altogether. In contrast, sexual activity is a normative behavior which we merely seek to delay rather than prevent. As a result, we need to remove it from this collective framework."

Virginity Pledges

One finding from the Add Health survey immediately seized upon by the popular press is that some students who took a virginity pledge promising to abstain from sex until marriage were less likely than students who did not take a pledge to become sexually active. According to a study published this year in the American Journal of Sociology, the pledge was found to delay intercourse by an average of 18 months. However, the pledge had some significant limitations—it worked best among 15-17-year-olds, with no effect among older teenagers, and had an impact only within certain ethnic groups.

In addition, virginity pledges were found to work only in those school contexts in which the pledge essentially constitutes minority group behavior. In fact, as the proportion of students who pledge rises, the effectiveness of pledging decreases. In other words, students are attracted to virginity pledges precisely because they offer them a shared group identity that sets them apart from their peers—a counterculture of sorts—that loses its allure once it becomes normative. Say authors Peter S. Bearman and Hannah Bruckner, "if most adolescents were to pledge, there would be no pledge effect." As a result, virginity pledges "cannot work as a universal strategy" and "policy makers should recognize that the pledge works because not everyone is pledging." Michael D. Resnick, also of the Center for Adolescent Health and Development at the University of Minnesota and a lead Add Health researcher, reiterates that "while the effect of virginity pledges is real, virginity pledges are not an immunization which work in every context. Policymakers would be wrong to think that virginity pledges will have a magical effect on kids' behaviors. Pledges work only for those young people who identify with this norm. If you make it mandatory, kids will fight it."

Virginity pledges also have an unintended effect that should be of concern to researchers and policymakers alike: They actually place some teenagers at higher risk of unintended pregnancy and sexually transmitted disease (STD), because teens who break their pledge are one-third less likely than non-pledgers to use contraceptives once they do become sexually active. Write authors Bearman and Bruckner, "pledgers are less likely to be prepared for an experience they have promised to forego." So before policymakers rush to embrace virginity pledges as the answer to teenage sexual activity, they should heed Bearman and Bruckner's suggestion "that pledgers, like other adolescents, may benefit from knowledge about contraception and pregnancy risk, even if it appears at the time that they do not need such knowledge."

Communicating with Parents

A key finding of the Add Health survey is that teenagers who feel highly connected to their parents—teenagers who report that their parents are warm, caring and supportive—are far more likely to delay sexual activity than their peers. This protective effect not only applies to sexual activity but is universal for all risk behaviors, according to a study published by Resnick and his colleagues in 1997 in the Journal of the American Medical Association. Add Health findings also show that teens who feel highly satisfied with their relationship with their mother are more likely to use contraception and to delay sexual activity and are less likely to become pregnant.

At least two studies using Add Health data explore how discussions about sex and contraception between mothers and teenage children—as well as teens' perceptions of these discussions—may influence adolescent sexual behavior. One of these studies, published last year in the Archives of Pediatrics and Adolescent Medicine, found that teenagers in grades 8-11 who perceive that their mother disapproves of their engaging in sexual intercourse are more likely than their peers to delay sexual activity. And perhaps not surprisingly, teenagers who are more connected to their mothers are more likely to accurately perceive maternal disapproval of sex. But when mothers recommend specific methods of birth control, teens are less likely to perceive strong maternal disapproval of sex. Researchers working in this area emphasize that discussions about birth control do not cause teens to have sex, but may instead be temporally related to the onset of sexual activity. "What we suspect," explains author Clea S. McNeely, from the Division of General Pediatrics and Adolescent Health at the University of Minnesota "is that mothers increase their discussions about birth control and sexual activity when they sense that their teenagers are sexually active or about to become sexually active."

Another study published last year in the American Journal of Public Health explored the equally complex area of perceived maternal approval of contraception. Teenagers in grades 7-11 who perceived that their mother disapproved of contraception were more likely to delay sexual activity. But these teens were also less likely to use contraception when they did become sexually active. This latter finding is particularly crucial, notes Resnick, because one factor that most closely predicts whether or not a teen has experienced a pregnancy is whether or not that teen used contraception the first and last times he or she had sexual intercourse. And, of course, unprotected sex increases a teen's risk of contracting an STD.

Messages for Parents and Policy

Researchers analyzing Add Health data are only beginning to understand the various factors that place teens at risk for and protect against teenage sexual activity and how these factors interact. And it may be that aside from the findings on virginity pledges, which are basically cautionary, the research thus far has less direct relevance for policymakers than it does for parents, because it speaks to the quality of parenting and parent-child communication. Resnick sums up the various Add Health findings this way: "You can look at all sorts of variables related to sex, but the most powerful predictor is whether teens have been in a romantic relationship for more than 18 months. This suggests several things. Parents can discourage heavy one-on-one dating and emphasize going out in groups. They should attend to issues like the age at which kids start dating and the age difference between partners." This is particularly important from the perspective of teenage pregnancy and STD prevention, given the findings of a study published this year in Family Planning Perspectives based on Add Health data that shows that younger adolescents who are sexually active, as well as adolescents involved with partners who are more than two years older or attend a different school, are less likely to use contraception than their peers.

McNeely concurs: "What Add Health does tell us—whether we are politically on the left or the right—is that parents should focus on the appropriate supervision and monitoring of their teens. They should meet boyfriends and girlfriends, get to know their teens' friends, and always know where their children are. They should establish a warm and caring relationship with their children and foster age-appropriate autonomy and independence. Ultimately, good parenting is more important than the specific content of the communication about sex. For me, that's the bottom line."