Pediatric Annals

The Effects of Sexuality Education on Adolescent Sexual Activity

James W Stout, MD, MPH; Douglas Kirby, PhD

Abstract

Sex education has had a presence in our country's schools since the turn of the century, when health educators hoped to reduce or curtail sexually transmitted diseases and premarital sexual intercourse.1 Both the curriculum and the agenda of sex education have experienced several generations of profound change since then. This article will acquaint the reader with recent efforts in the field of sex education toward altering sexual behavior, the theoretical underpinnings of these programs, and the effects of these efforts. This information should be directly applicable to those practitioners who provide consultation about or are directly involved in the development of sexuality education programs.

Since the turn of the century, the prevalence of adolescent sexual activity has markedly increased as has our understanding of the antecedents and consequences of this behavior. By age 15, 26% of females and 33% of males have experienced sexual intercourse; by age 20, these percentages increase to 75% and 86%.2,3 One quarter of all girls become pregnant by the time they are 18 years old; 85% of these pregnancies are unintended.4 More than 40% of all girls become pregnant before they turn 20.5

Early childbearing has a well-documented impact on both the lives of teen parents and their progeny, exacting enormous social and monetary costs. Pregnancy is the number one reason cited by girls for dropping out of school.6 Teen mothers are more likely to become single heads of households, and more than 50% are more likely than nonparents to receive welfare.5

As grim as these consequences are, the results of unprotected sexual activity are no longer dominated by the prospects of lost life opportunities, but by lost life. Although the actual numbers involved are low, heterosexual acquired immunodeficiency syndrome (AIDS) cases among 13 to 24 year olds represent the most rapidly growing subpopulation of this incurable and lethal sexually transmitted disease.7 As a result, reducing unprotected sex is no longer a matter of promoting an optimistic future, but rather a moral imperative to save young lives.

Although the impact of sex education is the focus of this article, other innovative approaches may reduce unprotected sex and therefore deserve mention. These interventions include teen theater, peer counseling, family communication programs, school counseling services, school nurse services, and condom availability programs. Of special note are schoolbased and school-linked clinic services, which couple comprehensive primary health care to the only institution with daily exposure to the majority of our youth-the public schools.

THE GENERATIONS OF SEXUALITY EDUCATION PROGRAMS

Our understanding of the evolution of sexuality education can be enhanced by dividing this movement into several generations.8 First-generation sex education programs were designed to impart knowledge about sexuality, the risks and consequences of pregnancy, and birth control. They were based on the belief that if students knew about reproductive physiology, contraception, and the consequences of unprotected intercourse, they would rationally choose either to avoid sexual intercourse or to improve their contraceptive behavior. Evaluations of these firstgeneration programs consistently demonstrated that these programs increased knowledge, but they did not consistently demonstrate changes in sexual behavior.8,9

Second-generation programs included a considerable focus on knowledge content, but paid more attention to values clarification, decision making, and communication skills. The values imparted in these curricula tended to be broad and inclusive, such as "All people should be treated with respect and dignity," and "The current and future consequences of behavior should be carefully considered." Curriculum planners reasoned that if students' values were clarified and decision-making and communication skills improved, actual decision making would improve, better decisions would be communicated to their partners, and unprotected intercourse would decrease. An evaluation of two such…

Sex education has had a presence in our country's schools since the turn of the century, when health educators hoped to reduce or curtail sexually transmitted diseases and premarital sexual intercourse.1 Both the curriculum and the agenda of sex education have experienced several generations of profound change since then. This article will acquaint the reader with recent efforts in the field of sex education toward altering sexual behavior, the theoretical underpinnings of these programs, and the effects of these efforts. This information should be directly applicable to those practitioners who provide consultation about or are directly involved in the development of sexuality education programs.

Since the turn of the century, the prevalence of adolescent sexual activity has markedly increased as has our understanding of the antecedents and consequences of this behavior. By age 15, 26% of females and 33% of males have experienced sexual intercourse; by age 20, these percentages increase to 75% and 86%.2,3 One quarter of all girls become pregnant by the time they are 18 years old; 85% of these pregnancies are unintended.4 More than 40% of all girls become pregnant before they turn 20.5

Early childbearing has a well-documented impact on both the lives of teen parents and their progeny, exacting enormous social and monetary costs. Pregnancy is the number one reason cited by girls for dropping out of school.6 Teen mothers are more likely to become single heads of households, and more than 50% are more likely than nonparents to receive welfare.5

As grim as these consequences are, the results of unprotected sexual activity are no longer dominated by the prospects of lost life opportunities, but by lost life. Although the actual numbers involved are low, heterosexual acquired immunodeficiency syndrome (AIDS) cases among 13 to 24 year olds represent the most rapidly growing subpopulation of this incurable and lethal sexually transmitted disease.7 As a result, reducing unprotected sex is no longer a matter of promoting an optimistic future, but rather a moral imperative to save young lives.

Although the impact of sex education is the focus of this article, other innovative approaches may reduce unprotected sex and therefore deserve mention. These interventions include teen theater, peer counseling, family communication programs, school counseling services, school nurse services, and condom availability programs. Of special note are schoolbased and school-linked clinic services, which couple comprehensive primary health care to the only institution with daily exposure to the majority of our youth-the public schools.

THE GENERATIONS OF SEXUALITY EDUCATION PROGRAMS

Our understanding of the evolution of sexuality education can be enhanced by dividing this movement into several generations.8 First-generation sex education programs were designed to impart knowledge about sexuality, the risks and consequences of pregnancy, and birth control. They were based on the belief that if students knew about reproductive physiology, contraception, and the consequences of unprotected intercourse, they would rationally choose either to avoid sexual intercourse or to improve their contraceptive behavior. Evaluations of these firstgeneration programs consistently demonstrated that these programs increased knowledge, but they did not consistently demonstrate changes in sexual behavior.8,9

Second-generation programs included a considerable focus on knowledge content, but paid more attention to values clarification, decision making, and communication skills. The values imparted in these curricula tended to be broad and inclusive, such as "All people should be treated with respect and dignity," and "The current and future consequences of behavior should be carefully considered." Curriculum planners reasoned that if students' values were clarified and decision-making and communication skills improved, actual decision making would improve, better decisions would be communicated to their partners, and unprotected intercourse would decrease. An evaluation of two such programs found no significant effect on sexual or contraceptive behavior.1

Rather than evolving from the first two generations of programs, third-generation programs arose in partial opposition to them. These programs, such as Sex Respect and Teen Aid, presented a moral and ideological emphasis through a strong message that young people should not engage in intercourse until marriage. In order to avoid delivering a double message, contraception was not discussed. Evaluation of these "abstinence only" programs showed that in the short term, they did reduce reported acceptance of premarital intercourse. However, what little evidence exists on long-term impact shows a marked reduction in these effects. Furthermore, limited evaluations of two curricula indicated that they did not delay the onset or reduce the frequency of intercourse.10,11

In 1989, a review of evaluations addressing the impact on sexual behavior resulting from the first two generations of programs reported that "the available evidence indicates that traditional sex education programs in junior and senior high schools have little or no effect either positively or negatively on altering the age of onset or frequency of adolescent sexual activity, on increasing contraceptive use, or on preventing unplanned teenage pregnancy."12 There are several reasons for the null findings from these first three generations of sex education programs. First, the evaluations were limited methodologically. Some used retrospective designs, especially cross-sectional surveys based on national survey data. These surveys measured respondents' reports of their own sexual behaviors and their own exposure to any sex education during their schooling. While some sexuality education programs are comprehensive in nature, most are not, and in these national surveys, the quality of the sex education was not well measured.13 Other evaluations used prospective quasi-experimental designs, but none of them had all the elements of good design; that is, none used random assignment to program and control groups, had large sample sizes, followed youth for at least 18 months, and measured behavior.

A second reason that these evaluations demonstrated little success in changing behavior is, of course, that it is difficult to change other people's behavior in general, and it is especially difficult to change adolescent sexual behavior. After all, that behavior is strongly affected by hormones and physical desire; needs for love and acceptance; family, peer, and personal values; the media; and a myriad of other factors. Thus, we should not expect brief units of instruction to have a dramatic impact on adolescent sexual behavior.

We should also recognize that when we evaluate the impact of sexuality education programs by measuring their impact on behavior outside the classroom, we are applying standards that are not applied to other courses of instruction. For example, the effectiveness of a civics class is not assessed by measuring the students' law-abiding behavior outside of the classroom.

As we shall see in the fourth generation of programs, finding a measurable impact on behavior requires both innovative and more effective curricula, as well as more rigorous evaluation methods.

FOURTH-GENERATION PROGRAMS

In contrast to past efforts, the fourth-generation programs are built on a theoretical foundation of adolescent behavioral change that emphasizes the developmental and motivational perspective of adolescents at risk. The hindsight of null findings from the first- and second-generation programs led evaluators in a well-established tradition of scientific experimentation-building on past failures. It is also noteworthy that the theoretical underpinnings of sexuality education curricula have typically followed the lead of youth-oriented substance-abuse prevention programs over the last two decades, a tradition that is especially apparent in the fourth-generation programs discussed below. This borrowing of materials and instructional methods from a similar discipline to design an intervention targeted at sexual behavior has been a pragmatic and responsible cross-pollination of limited resources.6

The fourth generation of sexuality education programs as a group are neither value-free nor moralistic. They emphasize that delaying the onset of sex is a wise choice, and they also stress the importance of effective contraception if intercourse is initiated. There are two principal theoretical models. The first is the Health Belief Model, which was first popularized in the dietary literature and is based on acquiring a number of personal beliefe.14 In the context of AIDS and pregnancy prevention programs, these beliefs include the student's perceived susceptibility to pregnancy and AIDS, the seriousness of acquiring these problems, and the benefits and costs of preventive action, such as abstinence or condom use.

The second theoretical model is Social Learning Theory, which stresses that people learn by observing their own behavior and the behavior of others and by then observing the consequences of that behavior. That is, it places a great deal of emphasis on learning from experience, both one's own and others' experience. It also emphasizes that youth need more than just a knowledge base; they must also have the motivation, skills, and confidence in those skills to avoid unprotected sex. To increase the motivation and skills, educational sessions include guided practice with knowledgeable teachers and learning through discussion, rather than didactic presentation, thereby giving the student more ownership of key concepts and factual material. Some programs based on social learning theory also use role playing to model and provide repeated practice in skills and to reinforce norms against unprotected sex. Some programs also use peer educators who serve as role models and who express clear norms against unprotected sex-norms that might not be as accepted if they were presented by older adults.

A second critical feature of these fourth-generation programs is the inclusion of a more rigorous evaluation component in the planning phase of the project. In addition to allowing a prospective design, this approach also forces the program planner and evaluator to explicitly address the desired behavioral outcomes, bringing a much-needed focus to the content of an intervention.

Rigorous evaluation designs should incorporate three fundamental concepts.9 These concepts represent a difficult standard for program evaluation. While the fourth-generation evaluations meet these standards to varying degrees, in general they come much closer to the mark than do earlier evaluation efforts. First, there should be random assignment of the study unit of analysis (eg, students, classrooms, or schools) in order to avoid selection bias. Second, there needs to be a large sample size so that programmatically significant results are likely to be statistically significant. Large samples also are needed because the statistical analyses typically require analyses of subpopulations (whether sexually inexperienced youth initiate sex and the extent to which sexually experienced youth use protection). Finally, there needs to be long-term tracking of study participants both to determine whether short-term effects endure and to determine whether new results appear in the longer term. For example, at least 18 months are needed to determine the impact of programs on initiating sex; in less than 18 months, too few members of the control group initiate sex for there to be a statistically significant difference between the program and control groups.

Four fourth-generation programs will be briefly discussed. Each of these "real world" examples employed principles of Social Learning Theory and the Health Belief Model. Although the principles of rigorous study design are not universally employed, most elements are present in most cases.

Eisen et al15 reported on a sex education program in Houston. It was based primarily on a health belief model, but included elements of social learning theory. It was designed to "increase teenagers' awareness of the probability of personally becoming pregnant or causing a pregnancy, the serious negative personal consequences of teenage maternity and paternity, and the personal and interpersonal benefits of delayed sexual activity and consistent, effective contraceptive use."15 It used discussions and role playing. The evaluation employed a true experimental design, but the control groups received other sex education programs. The study also had a large sample size (1444 13 to 19 year olds) and tracked students for 1 year. Its results were mixed and varied by gender and sexual experience at pretest. In comparison with their counterparts in a control group, male students in the treatment group were less likely to initiate sex, but female students were not less likely to do so. Among those students who initiated intercourse after the baseline data were collected, females (but not males) in the treatment groups were significantly less likely to use contraception than their control-group counterparts.

A second program in this fourth generation is Postponing Sexual Involvement, which is described in greater detail elsewhere16 and in this issue of Pediatric Anruds. It was based on a variation of Social Learning Theory, employing a social-influences approach to changing behavior. During this program, high school students were trained as peer educators and then presented a five-session program to junior high school students.

They emphasized that the junior high school students should delay having sex and it employed a variety of exercises, including role playing to establish and emphasize that norm. Although this program focused on delaying sexual intercourse, Postponing Sexual Involvement was not moralistic and differed in other important ways from the third generation of programs. The evaluation did not use an experimental design, but the weight of the evaluation evidence did indicate that the program delayed the onset of intercourse among students, but did not affect either the frequency of intercourse among those who had already had sex nor change the use of contraception.

Kirby8 reported on an evaluation of a California curriculum called Reducing the Risk, which involved 758 high school students in intervention and comparison groups. The intervention was primarily based on concepts derived from Social Learning Theory. The curriculum emphasized avoiding unprotected intercourse, either by not having sex or by using contraceptives. Students participated in role-playing exercises that presented increasingly challenging situations about abstinence and birth control. The curriculum was presented over 15 class periods in 10th grade classes, as a component of a pre-existing health education class. Similar students in the comparison classrooms received a pre-existing sex education curriculum, similar in length to the experimental intervention. Among students who had not had intercourse prior to the intervention, this study showed a significant reduction in the likelihood that they would have intercourse in the subsequent 18 months or that they would have unprotected sex. In addition, there was a significant increase in the use of contraception among females who did engage in intercourse.8

An evaluation by Jemmott et al17 was not a study of a school-based program, but because it was an educational program that could be replicated in or through schools, it warrants discussion. Jemmott et al reported on a well-designed study of 157 inner-city Philadelphia black male adolescents who were randomly divided into experimental and control groups.17 Both of these groups received one 5-hour small-group intervention on a Saturday, the control group participating in a similarly structured career counseling session. Black male or female group leaders received special training in the intervention. Again using concepts from both Social Learning Theory and the Health Belief Model, activities were designed to increase the knowledge of AIDS and sexually transmitted diseases, and to weaken problematic attitudes toward risky sexual behaviors. Videotapes, games, exercises, and other culturally and developmentally appropriate pilot-tested materials were designed to provide accurate information in interesting ways. The 3-month follow-up survey found that the intervention group reported fewer occasions of coitus, fewer coital partners, greater use of condoms, and lower incidence of heterosexual anal intercourse than controls.17

DISCUSSION

Generalizations From the Fourth-Generation Programs

These fourth-generation programs illustrate several important points. First, to the extent that we can generalize from this limited number of studies, it appears that curricula based on social learning theory are more effective than those based on the health belief model and are clearly more effective than those curricula found in previous generations of programs.

Second, all these curricula attempted to change more than just knowledge; they placed considerable emphasis on norms and skills.

Third, these curricula use sound teaching techniques. For example, most of these programs place a premium on active or experiential learning and group participation in order for the students to personalize and take ownership of the concepts presented.

As a group, these programs appear to be most effective for those students who had not yet engaged in sexual intercourse. A general behavioral precept is that past behavior is the best determinant of future behavior, given similar circumstances.18 Once youth have developed a pattern of engaging in unprotected sex, it may be more difficult to change that pattern than to prevent a pattern that hasn't started.

From a theoretical perspective, it seems likely that the factors affecting the initiation of sexual behavior may differ from those that govern the maintenance and strengthening of that behavior.19 Furthermore, these determinants and dynamics are different for males and females, for adolescents at different developmental stages, and for those with different levels of self-esteem and future expectations. Thus, ideally, different sexuality education programs need to be designed for each of these groups. Practically speaking, it is, of course, difficult, if not impossible, to implement different approaches for different groups. Nevertheless, tailoring programs to the different needs of different youth remains one of the challenges to curriculum designers.

When the complexity of human sexual behavior and the crucible of adolescence are combined, however, it should be clear that effective solutions will require multiple resources and approaches. Although abstinence is the best method of protection against pregnancy and sexually transmitted disease, and although abstinence should be emphasized and encouraged, "Just say no" approaches will not meet the varying needs of adolescents.

FAMILY

Our society properly recognizes the important role of family values, especially in the area of sexuality. There may be no greater single potential influence on a child's sexual behavior than his or her parents. We know, for instance, that family values and communications about sexuality have a significant impact on the likelihood that adolescents will initiate intercourse or use contraception.20

Unfortunately, many parents have difficulty discussing sexuality with their children and prefer help. Some opponents of sex education programs claim that sex education thwarts communication with parents. Fortunately, the research shows that well-designed curricula can actually increase parent-child communication. For example, the Reducing the Risk curriculum included an activity in which students discuss abstinence and contraception with their parents. Consequently, there was a significant increase in such communication, and some parents discussed these topics with their children for the first time.8

THE IMPORTANCE OF MULTIPLE INTERVENTIONS

Curriculum-based sexuality education is an important component of the solution to the adolescent pregnancy problem, but we should not delude ourselves into thinking that this or any intervention provides the single answer. As Dryfoos, among others, has pointed out,

No "magic bullet" has been developed that will help young people adopt effective fertility control. . .Young people must have access to an array of developmentally appropriate interventions, beginning in the earliest years and continuing through middle school and high school. No one-shot or one-component approach can have as strong an effect as staged, multi-component efforts. This implies that at the community level, schools and community agencies must plan together to ensure that all the stages get covered.21

Thus, the kinds of curricula found in these fourthgeneration programs can be effective components, but alone they cannot solve all the problems of unprotected sex; other components are desperately needed.

BROADENING OUR DEFINITION OF OUTCOMES

In 1975, the World Health Organization defined sexual health as ". . .the integration of physical, emotional, intellectual, and social aspects of sexual being in ways that are positively enriching, and that enhance personality, communication, and love. . .every person has a right to receive sexual information and to consider accepting sexual relationships for pleasure as well as for procreation."22

This poses a real challenge for sexuality education.

On the one hand, it clearly is desirable to do more than simply reduce unprotected sex. On the other hand, to the extent that the goals of sexuality education are broadened, sexuality education programs become less focused and less effectively reduce unprotected sex. One conclusion from the research on the second generation of programs is that some of them tried to do too much and consequently had less impact on any particular outcome. One possible solution to this dilemma is to prioritize all the important goals of sexuality education. As programs effectively achieve the most important goals, then they can be expanded so that they may achieve additional goals.

CHALLENGES TO PROGRAM IMPLEMENTATION

There are reasons for optimism in our field. For example, more than 90% of parents want their children to receive comprehensive sexuality education,23 and research indicates that some of the fourth generation of programs do reduce unprotected intercourse. Furthermore, other studies are underway and their preliminary results are encouraging.

On the other hand, the challenge of implementing effective programs in a large-scale fashion cannot be overemphasized. Currently, fewer than 10% of children receive comprehensive sexuality education programs23 and the percentage that receives effective programs is unknown and small. Thus, an enormous effort is needed nationally to have adequately trained sexuality educators and to implement comprehensive and effective programs. The Division of Adolescent and School Health in the Centers for Disease Control and Prevention is providing some leadership and some funding, but much more needs to be done.

The problems of limited resources are real and pervasive for public schools and public health agencies, and effective pregnancy and sexually transmitted disease prevention programs must compete within the public health and education systems for limited resources with other worthy programs such as immunization efforts or dropout prevention. On the other hand, because the costs of unintended pregnancy, sexually transmitted diseases, and human immunodeficiency virus are so enormous, reasonable resources should be allocated to their prevention, especially when we now can have greater confidence of success.

The recent election year has finally turned the gaze of our nation's leaders toward domestic needs such as these. Our children cannot speak for themselves, and we must not underestimate our collective voice as advocates for children's well-being. We should strive to find resources needed to improve the effectiveness of programs, to evaluate them, and to implement broadly those that are effective.

REFERENCES

1. Kirby D. Sexuality Education: An Evaluation of Programs and Their Effects. Santa Cruz, Calif: Network Publications; 1984.

2. Centers for Disease Control. Premarital sexual experience among adolescent women, United States, 1970-1988. MMWR. 1990;39:929-932.

3. Sonenstein FL, Pleck JH, Ku LC. Sexual activity; condom use and AIDS awareness among adolescent males. Fam Plann Perspect. 1989;21:152-158.

4. Dryfoos JG. Factbook on Teen Pregnancy. New York, NY; Alan Guttmacher Institute; 1981.

5. Hayes CD, ed. Risking the Future: Adolescent Sexuality, Pregnancy, and Childbearing. Vol 1. Washington, DC: National Academy Press; 1987.

6. Walker G, Vilella-Velez F. Anatomy of a Demonstration. Philadelphia, Pa: Public/ Private Ventures; 1992.

7. Centers for Disease Control and Prevention. Update: acquired immunodeficiency syndrome-United States, 1981-1990. MMVCR. 1991;40:358-363.

8. Kirby D. Reducing the risk: impact of a new curriculum on sexual risk-taking. Fam Plann Perspect. 1991;23:253-263.

9. Kirby D. School-based programs to reduce sexual risk-taking behaviors. J Sch Health. 1992;62:280-287.

10. Roosa M, Christopher S. Evaluation of an abstinence-only adolescent pregnancy prevention program: a replication, family Relations. 1990;39:363-367.

11. Christopher FS, Roosa MW. An evaluation of an adolescent pregnancy prevention program: is "just say no" enough? Family Relations. 1990;39:68-72.

12. Stout JW, Rivara FP. Schools and sex education: does it work? Pediatrics. 1989;83:375-379.

13. Selverstone R. Sexuality education for adolescents. Adolescent Medicine: State of the Art Reviews. 1992;3:195-205.

14. Janz N, Becker MH. The health belief model: a decade later. Health Educ Q. 1984;11:1-47.

15. Eisen M, Zellman GL, McAlister AL. Evaluating the impact of a theory-based sexuality and contraceptive education program. Fam Plann Perspect. 1990;22:261-271.

16. Howard M, McCabe J. Helping teenagers postpone sexual involvement. Fam Plann Perspect. 1990;22:21-26.

17. Jemmott JB III, Jemmott LS, Fong GT. Reductions in HIV risk-associated sexual behaviors among black male adolescents: effects of an AIDS prevention intervention. Am J Public Health. 1992;82:372-377.

18. Bandura A. Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall; 1977.

19. Bandura A. The Social Foundations of Thought and Actum. Englewood Clifts, NJ: Prentice-Hall; 1986.

20. Hofferth SL. Factors affecting the initiation of sexual intercourse. Risking the Future: Adolescent Sexuality, Pregnancy, and Childbearing. Vol 2. Washington, DC: National Academy Press; 1987.

21. Dryfoos JG. Adolescents at Risk. New York. NY: Oxford University Press; 1990.

22. World Healrh Organization. Education in human sexuality for health practitioners. WHO Chronicles. 1975;29(2):49-54.

23. Donovan P. Risk and Responsibility: Teaching Sex Education in America's Schools Today. New York, NY: Alan Guttmacher Institute; 1989.

10.3928/0090-4481-19930201-09

Sign up to receive

Journal E-contents