Reflections on Two Decades of Research on Teen Sexual Behavior and Pregnancy
Kirby, Douglas, Journal of School Health
Exactly 20 years ago, I began doing research on sex education programs designed to reduce unprotected sex and teen pregnancy. At that time there was increased national concern about teen-age pregnancy; a belief that sex education programs could provide American youth with needed information about sexual behavior, contraception, and pregnancy; and a desire to find effective sex education programs and demonstrate their effectiveness through research. At that time HIV/AIDS was not known, and not part of our concerns, but there was concern about STD more generally. During these past 20 years, our field has come a long way, both in the development and use of research methods and in its understanding of adolescent sexual behavior, teen pregnancy, and programs that affect them. This paper highlights some of those advances.
ADVANCES IN RESEARCH METHODS
Our field has made tremendous advances in research methods -- both in its understanding of how to conduct rigorous research and in its actual use of more rigorous research methods. There are at least five different areas in which major improvements have taken place.
Measurement of Sexual Behavior
Twenty years ago, only a few studies measured the impact of sex education programs, and of those studies which measured impact, still fewer measured impact upon the very behavior that the programs were designed to change, namely the sexual and contraceptive behaviors of youth. It was even less common to ask questions about sexual behavior in school settings, and still less common to do so with federal funds. When the Center for Health Promotion and Education (the forerunner of the Division of Adolescent and School Health) at the Centers for Disease Control (CDC) funded a group of us to measure the impact of sex education programs on student sexual attitudes and behavior, many people clearly stated that it simply could not be done.
And indeed, it almost could not. People opposed to using taxpayer dollars in this manner obtained copies of our proposed questionnaires, selected questions about attitudes and behavior that they felt were particularly egregious, and sent them with a highly critical cover letter to every Congress person in Washington. Shortly thereafter, I received a concerned call that the Center had received a Congressional inquiry -- this seemed unprecedented for that Center, and clearly of grave concern. And then there was a second inquiry, and then more, and then many, many more. Fortunately, the Center, and CDC more generally, remained committed to the study and resisted the pressure to terminate the study or to abstain from measuring sexual behavior.
Today, questions about adolescent sexual behavior are routinely asked in surveys, even in school settings. Questions are commonly asked about whether the students have ever had sex, their frequency of sexual activity, their number of sexual partners, their use of condoms, their use of other forms of contraception, whether they have ever been pregnant and whether they have ever had an STD. Some more recent studies have even refined these questions to assess the behaviors that are most highly related to pregnancy, HIV, and STD (ie, frequency of sex without any kind of contraception, frequency of sex without condoms, and number of sexual partners without condoms). Indeed, some of these questions about sexual and contraceptive behavior are asked not only of high school students, but of middle school students. Studies asking these questions now provide reasonably accurate estimates of student sexual and contraceptive behavior in many individual schools, communities, states, and even the entire United States. Studies employing these questions have also given us a better understanding of the behavioral impact of different programs.
This very positive picture should, however, be balanced by the recognition of efforts to thwart and limit survey research. For example, there are efforts to make it more difficult to ask students any personal questions and it is difficult, if not impossible, in most schools to ask questions about sexual orientation, even though such questions could help us better understand how to meet the needs of some youth at higher risk of HIV. …