The drumbeat of criticism that eventually drove Joycelyn Elders out of office as Surgeon General may be only a fading memory, but the controversies over sex education and contraception that dogged her tenure linger on. To conservatives, nothing symbolizes the illusions of liberalism better than the failure of permissive sexual policies. In the years since contraceptives became widely available and schools began offering sex education, haven't kids become more promiscuous? Aren't births to unmarried teenage mothers soaring? Therefore, conservatives say, the government ought to practice some abstinence of its own and stop sex education in our schools and programs that promote contraception.
But, like so many conservative arguments that appeal to a general sense of social decline, this one ignores some well-established facts. More teenagers use contraception, they use it sooner after starting sex, and they are becoming more sophisticated about its use. Pregnancy rates among sexually active teenagers have dropped, decreasing by 20 percent between 1970 and 1990. Recent evidence also suggests that sex and AIDS education programs in the public schools have encouraged youth to delay sex, limit the number of partners, and use condoms.
But, conservatives say, increased access to contraceptives and sex education has stimulated more sexual activity among teenagers. These fears were cogently voiced in 1978 by Archbishop (now Cardinal) Bernardin, who doubted, he said, whether "more and better contraceptive information and services will make major inroads in the number of teenage pregnancies—it will motivate them to precocious sexual activity but by no means to the practice of contraception. In which case the solution will merely have made the problem worse."
Was the Archbishop right? For if he was, Americans might have reason to shut down the great enterprise of sexual enlightenment that America launched thirty years ago.
An American Transformation
Turning the clock back on policies toward contraception and sex education would return us to a radically different age. Until 30 years ago, it was the policy of the U.S. government to keep contraceptives out of the hands of the poor, the unmarried, and the young. Even information about contraceptives was hard to obtain because of the legacy of the Comstock Act of 1873, which defined contraceptives as "obscene." As late as 1964 contraception was nominally illegal in some states even for married people. Public contraceptive programs, condoms in plain view in grocery stores, and magazine advertisements for contraceptive products were unimaginable. Sex education for many students before the 1960s consisted of a brief lecture about menstrual hygiene (delivered to girls by the school nurse) or nocturnal emissions (delivered to boys by the coach). Where condoms were legal, they were typically available only behind the counter in pharmacies, which often refused to sell them to customers the pharmacist knew or suspected weren't married. Doctors and clinics also often turned the young away unless they could show proof of parental consent.
Then, in a remarkably short time, the Supreme Court made contraception legal, and the president and Congress helped to make it accessible. In 1964 the Court ruled in Griswold v. Connecticut that states could not ban the use of contraceptives. In the War on Poverty that began the same year, providing contraceptives to poor women became a high priority. Prior to Griswold, only women who could afford a private physician had been able to acquire birth control because the Comstock Act had been interpreted as prohibiting, in most circumstances, public expenditures on contraception. Under antipoverty programs, the first recipients of contraceptives were poor married women, but soon teenagers were also included. In 1966 Congress mandated that birth control be offered to any woman over 15 years of age on public assistance, married or not. In 1967 the federal government reserved 6 percent of maternal and child health funds for family planning for poor women, and in 1972 it stipulated that state welfare programs offer contraceptives to "minors who can be considered sexually active." By the mid-1970s, public contraceptive clinics were the first choice of a majority of teenagers and were especially likely to be used by teenagers from poor families and by minority teens. Contraceptives became available through a wide network of public health services, hospitals, clinics, and Planned Parenthood centers—all this, despite the absence of a national health care system.
Although ready access to contraceptives is now part of the fabric of American life, conservatives hold it partly responsible for what they see as deepening moral decline. Among the sources of misperception about the consequences of liberalized contraceptive access is a series of misunderstandings about teenage pregnancy and the use of birth control. Constant references to "an epidemic of teenage pregnancy" suggest that the pregnancy rate for teens is dramatically higher than in the past and different from pregnancy rates among older women. In fact, the overall teenage pregnancy rate rose modestly between the early 1970s and the late 1980s, from 95 to 107 pregnancies annually per 1,000 women aged 15 to 19; it rose a little more rapidly from 1987 to 1991, and then fell in 1992 and 1993 (the last year for which we have data). Changes in the rate for teenagers closely track the somewhat higher pregnancy rates among women aged 20 to 29.
It is natural to assume that a higher teen pregnancy rate means that sexually active young women are more likely to conceive than they used to be, but this assumption is false. For most of the last two decades the pregnancy rate rose because more teenagers were sexually active, not because more sexually active teens were becoming pregnant. As more teens started to have sex while unmarried, they also became much more likely to use condoms, the pill, and other forms of birth control. "Timeliness of Teenage Contraception" tells the story. It divides women who became sexually active as teenagers into three groups: those who used an effective form of contraception at first intercourse; those who used any form of contraception (including withdrawal and rhythm) in the first year after having sex; and those who did not use any kind of birth control methods within a year of starting sex (or, in many cases, ever). In 1964 only one- third of sexually active 15- to 19-year-olds used protection during their first sexual experiences, while 40 percent did nothing to prevent pregnancy for at least a year after their first sexual intercourse. But by 1988, 56 percent of sexually active teens used contraception from the start, and fewer than 16 percent were delaying contraception by more than a year. The unsurprising result is that a smaller fraction of the sexually active teens became pregnant with every year that passed between 1972 and 1990.
But what about Archbishop Bernardin's thesis that offering contraception to teenagers increases the odds that they will become sexually active and, more precisely, that they will be sexually active without using contraception? Based on the historical record in the United States and other developed nations, no one has yet been able to show that liberalized contraceptive policies increase teenage sexual activity in general or unprotected sex in particular.
Looking overseas first, we find that almost all European nations report increases similar to ours in sexual activity among teens, although they have followed widely divergent policies on access to birth control. Some have long offered publicly funded birth control to women of all ages as part of their national health care systems. Others make it difficult for even adult women to acquire contraception. These varied national strategies make up a kind of natural experiment. The evidence shows that sexual activity among teenagers is independent of any changes in the public provision of contraceptives.
In the United States the policy changes of the 1960s and 1970s responded to social changes already under way. Young people were delaying marriage but not forgoing sex. In the early 1950s American women had a one-in-two chance of being married by the age of twenty. After 1960 the median age at marriage rose four years, lengthening by about 50 percent the time that sexually mature young women (and men) are single. Norms about sex and marriage changed, and the rate of sex outside of marriage increased accordingly. As "Sex Before Subsidies" shows, the proportion of American adolescents who were sexually active and unmarried was growing steadily before any public subsidy for birth control. Not only was teen sex already on the increase, but sexual activity leveled off as funding became relatively generous in the 1970s.
Thus the first part of Archbishop Bernardin's hypothesis—providing contraception increases sexual activity—is unsupported by the available data. His second claim, that as more teens become sexually active more of them engage in unprotected sex, was somewhat true during the 1970s but not during the 1980s (when our data end).
"Unprotected Teen Sex Rises, Then Declines," shows the trends between 1964 and 1988 in the sexual behavior of all young women aged 15 to 19. The proportion of all teens who were sexually active and who waited some months to use contraception did indeed grow between 1964 and 1980, but their numbers fell during the 1980s. In contrast, the proportion using contraceptives at first intercourse increased rapidly and continuously over the entire period. And the number of hard-core non-contraceptors—those who were having sex but waited at least a year to use a method—did not increase at all between 1964 and 1980, remaining at around only 5 percent of all teens.
In short, as more young unmarried women have become sexually involved, they have also become more likely to use contraception. And while unmarried virgins are less numerous among teens than they used to be, they still remain in the majority. It is married teens who have almost vanished from the landscape.
These data, however, cut two ways. While public funding of contraception has not caused more teens to have sex, neither is there any clear correlation between public funds and teenage use of contraceptives. When federal funds were cut in the 1980s, overall teenage contraceptive use did not decline too, although these broad national data may not pick up the difference public funding makes in low-income and minority communities. Clearly, other factors affect the use of contraceptives: the determination of many teens to avoid pregnancy; increased commercial access to contraceptives in large anonymous drugstores and supermarkets, and the dissemination of knowledge about birth control—including sex education programs in schools and throughout the community that conservatives have also attacked.
The Efficacy of Sex Education
Critics claim that sex education has failed primarily on the basis of research that has shown no appreciable difference in behavior between students who have taken sex education courses and those who have not. But only in recent years have most schools offered education about birth control to young teens, timed to occur before most of them are sexually active.
For many young people, sex education has come from a partner, not from a class. In the 1988 National Survey of Family Growth—the most recent, large-scale survey available—almost half of all young women (44 percent) born between 1963 and 1965 had sex education about contraception after they had become sexually active. But as schools became willing to teach sex education in lower grades, this pattern began to change. Of teens born between 1966 and 1968, 38 percent had sex before sex education, but of those born between 1971 and 1972, only 19 percent had been sexually active prior to any instruction about contraception.
Most types of sex education offered after sexual initiation have little effect on behavior. Yet the popular view that sex education does not work was based on early studies that did not distinguish youngsters who received sex education from those who sat through the instruction when they were already having sex. Any beneficial effects of sex education on the students who were still virgins were likely masked by the absence of effects among the sexually active.
Our own analyses of the 1988 survey data show a strong relationship between prior sex education and contraceptive use by teens. We found a difference of about 10 percentage points in the likelihood of contraceptive use. By 1988 young women who had had sex education were only half as likely as those who had not to delay contraception for a year or more.
The impact of sex education stems from small changes among many students. It can hasten their use of birth control, encourage more effective methods, and (though this is not our theme here) help students to resist premature or unwanted sexual activities. In short, it will nudge some students—not all—in the direction of safer behavior.
Some, of course, do not need to be nudged in school. Half of sexually active teens in the 1980s used some type of contraception at first sex even without formal sex education. Others cannot be reached even through a good program. About 3 percent of students who had had sex education had never used contraceptives even though they had been sexually active for more than a year. But between these extremes lie half of sexually active youth, whose behavior can be shaped by the information, skills, peer expectations, and adult counsel that constitute an effective sex education curriculum.
Designing Sex Education Programs
While these graphs suggest that sex education can work, aggregate data tell us nothing about what goes into an effective program. Fortunately, thanks to a panel of 14 national experts convened at the request of the Centers for Disease Control (CDC) and a recent analysis for the Office of Technology Assessment (OTA), we know more than ever before on this question. Under the leadership of Douglas Kirby of ETR Associates, the panel carefully reviewed the evaluations of 16 school-based programs and 7 studies using national data with an eye to establishing what works. Kirby subsequently reviewed an additional 33 studies for the OTA.
Both reviews first address the Bernardin hypothesis that sex education increases sexual activity among teens. None of the evaluated curricula hastened sexual intercourse or increased its frequency among participating students. Kirby and colleagues are unequivocal: "These data strongly support the conclusion that sexuality and AIDS education curriculums that include discussions of contraception in combination with other topics—such as resistance [to sexual pressure] skills—do not hasten the onset of intercourse." In fact, even those sex education programs associated with school-based clinics, which provide birth control to students, did not find that rates of sexual initiation went up.
Indeed, the news is that sometimes sex education can postpone sexual initiation if the program is based on carefully evaluated strategies and is offered to groups of students who are mostly still virgins. Kirby and colleagues note that "two curriculums that specified delaying the onset of intercourse as a clear goal . . . successfully reduced the proportion of sexually inexperienced students who initiated sex during the following 12 to 18 months. Notably, both groups also received instruction on contraception." This result may not have been found in earlier research into sex education because until recently, most curricula did not explicitly seek to discourage students from initiating sex at young ages.
Other programs that successfully influenced student behavior were focused on increasing contraceptive use or, more specifically, increasing condom use, among participating students. These programs had several features in common. They had clear goals and a relatively narrow focus, whether on postponing sexual involvement or on reducing risks of pregnancy or sexually transmitted diseases. They acknowledged the importance of peer group behavior in student learning. They offered accurate information through experiential exercises designed to let students personalize the information. And they let students practice skills in sexual communication, negotiation, and refusal.
In part because of their controversial character, the early sex education curricula that addressed contraception were often forced to adopt a tone of value neutrality, focusing on clinical information to the exclusion of the social, emotional, and moral aspects of sex. The research by Kirby and his colleagues suggests that this strategy was a mistake. In many respects, the most successful sex education programs are liberal in the breadth of their discussion but conservative in their directive message.
The Europeans reached this conclusion first. They have carved out a middle ground between absolute prohibition of adolescent sexuality and the total abdication of any adult responsibilities for guiding it. The new sex education programs in the United States are trying to create an analogous middle ground. Feminists and conservatives alike can find something to admire in programs that encourage young women (and men) to resist peer pressure and take responsibility when and if they feel truly ready for sexual intimacy. While the far right will still insist on a policy of "just say no" and sexual libertarians will resent any attempt to tell adolescents what to do, the emerging consensus of the middle has much to recommend it.
The CDC's team of reviewers emphasizes that we are just beginning to understand which factors contribute most to the overall success of the programs. Their main message is that some programs do work and that the next generation of programs should take advantage of the lessons that varied approaches teach.
American youngsters in the 1990s face a different world from the one that confronted their parents. More young people are sexually active, and more report that some sexual activity is coerced. Sexually transmitted diseases that threaten health and fertility (gonorrhea and chlamydia) or life itself (AIDS) afflict many young as well as older people. Helping teens handle these challenges isn't easy. Their needs change rapidly as they mature: A youngster may need encouragement to postpone sexual involvement when she or he is fifteen, easy access to contraceptives when he or she is eighteen, and, throughout, increasingly sophisticated help in sexual negotiation and refusal.
America has a long way to go before our teenagers are as effective in preventing pregnancy as are most of their European counterparts. While we understand the desire of many people to turn the clock back to a simpler age, the crucial task now is to continue studying open-mindedly what works for adolescents, and for whom it works. It is simplistic, and mistaken, to claim that the efforts of the past two decades to help teens have been either ineffective or counterproductive. Young people from across the social spectrum have taken advantage of public policies to help them take care of themselves. Legally imposed barriers that once imperiled their well-being have been lowered or removed. That these new policies and programs have made only slow and partial progress is evidence for strengthening them and designing them more intelligently. To abandon the effort now would be a kind of collective, parental irresponsibility.
Figure 1: Timeliness of Teenage Contraception
Description: Figure 1 tracks the timing of contraception among sexually active unmarried teens. It divides women who became sexually active as teenagers into three groups: those who used an effective form of contraception at first intercourse; those who used any form of contraception (including withdrawal and rhythm) in the first year after having sex; and those who did not use any kind of birth control methods within a year of starting sex (or, in many cases, ever). In 1964 only one- third of sexually active 15- to 19-year-olds used protection during their first sexual experiences, while 40 percent did nothing to prevent pregnancy for at least a year after their first sexual intercourse. But by 1988, 56 percent of sexually active teens used contraception from the start, and fewer than 16 percent were delaying contraception by more than a year. The unsurprising result is that a smaller fraction of the sexually active teens became pregnant with every year that passed between 1972 and 1990.
Figure 2: Timeliness of Teenage Contraception
Description: Figure 2 tracks teenage sexual activity and federal funds for contraception. The proportion of American adolescents who were sexually active and unmarried was growing steadily before any public subsidy for birth control.
Figure 3: Timeliness of Teenage Contraception
Description: Figure 3 tracks the timing of contraception by unmarried sexually active teens relative to all female teens (15 to 19 years old). The proportion of all teens who were sexually active and who waited some months to use contraception did indeed grow between 1964 and 1980, but their numbers fell during the 1980s. In contrast, the proportion using contraceptives at first intercourse increased rapidly and continuously over the entire period. And the number of hard-core non-contraceptors—those who were having sex but waited at least a year to use a method—did not increase at all between 1964 and 1980, remaining at around only 5 percent of all teens.