Trends in teenage sex have shifted significantly, according to the Centers for Disease Control and Prevention. Between 1991 and 2001, the percentage of U.S. high school students who had had intercourse dropped from 54.1 to 45.6, and condom use increased from 46.2 percent to 57.9 percent. These findings, heartening to most parents, may be due in part to work done by John Jemmott III, Ph.D., a communications professor at the University of Pennsylvania at Annenberg. Here, Jemmott discusses his research on sex education and how parents and educators can make it more effective.

Susan Fiske: Based on your research, what advice can you give parents who want to protect their kids from sexually transmitted diseases (STDs)?

John Jemmott III: Parents may not feel comfortable conveying their values about sexual behavior. But when do you want to talk to your child-before, or after he's had sex?

SF: What kinds of interventions are effective?

JJIII: Interactive, upbeat programs tend to be more engaging. Lectures can't hold an adolescent's attention, and dated educational films get a negative reaction because teenagers are so trendy.

SF: What topics of conversation would you encourage?

JJIII: There's a lot of emphasis now on abstinence until marriage. Given today's average age of first marriage, that's asking young people to delay sexual involvement for a considerable length of time. [According to the U.S. Census Bureau, the median age for first marriage is 26.8 for men and 25.1 for women.] Most of our work has focused on encouraging youth to use condoms. Recently, we did a study with 650 sixth- and seventh-graders. Even at that young age, about 25 percent had had sex. We randomized them to receive an intervention that focused on either abstinence or condom use. We found that the abstinence curriculum did have a significant impact on reducing sexual activity at the three-month follow-up, but after one year the effect disappeared. In contrast, the safe-sex intervention was consistently effective in increasing condom use over the entire one-year period.

SF: Have you discovered any cultural differences?

JJIII: You do have to use different kinds of activities with different populations. By using social cognitive theory we try to tap in to things that are consistent with cultural orientations. The basic premise is that people are motivated by the perceived consequences of their behavior: "If I don't have sex, my boyfriend's going to be angry with me"; "If I use condoms, it's going to ruin the sexual experience." Part of the intervention is to change their view of those consequences. Another part is addressing whether people important to the adolescents would approve of their behavior. And a third component is a person's confidence that they can engage in the new behavior.

SF: You also mentioned that interventions should be "upbeat."

JJIII: I don't believe that scare tactics work. On some level, young people have to believe that they're vulnerable, or they won't protect themselves. But focusing only on fear-arousing messages is not effective. Also, children want the conversation and they're not getting it. In surveys, parents ranked very low when adolescents were asked, 'From whom do you get most of your information about sex?' When asked from whom they would prefer to get information, parents were at the top.

SF: No kidding?kids really want to hear from their parents?

JJIII: It's really surprising, because the difficulty in talking exists on both sides. One problem is that many parents have little or no experience using condoms. So part of the process is getting parents comfortable with condoms.

SF: Do you see problematic trends in parent-child communications?

JJIII: When we ask "Have you talked to your child about sexual behavior?" oftentimes parents will say yes. But children typically say that the conversations have not occurred.

SF: So what parents consider a conversation may not be adequate from the kid's point of view?

JJIII: Exactly. We conducted another study and recruited mothers from public-housing developments who had sons ages 11 to 15. We randomized the mothers to receive either a sexual risk-reduction intervention or a control intervention. After three months, we found that mothers who were taught risk-reduction skills reported increased communication with their sons about condoms and sexual behavior, and we've found the same increases reported by the sons. Boys whose mothers received the intervention on STDs also reported reduced sexual activity compared with boys whose mothers received a control intervention.

SF: Should parent organizations and schools teach us how to have these difficult conversations with children?

JJIII: What's controversial about adolescent sexual behavior is how you deal with it in terms of prevention. But regardless of one's politics, the notion that you could effectively intervene with parents is very feasible. Then parents can frame sex education in terms of their own particular values.