Reflections on two decades of research on teen sexual behavior and pregnancy.

During the past 20 years, both researchers and program developers made great progress in their efforts to reduce adolescent unprotected sex and prevent teen pregnancy. Research studies are now more likely to employ experimental designs with random assignments, to have large sample sizes with adequate statistical power, to measure actual sexual and contraceptive behaviors, to measure longer term effects, to employ proper statistical methods, and to report results in an unbiased manner. As a result of this body of research, large advances have occurred in our understanding of: 1) the incidence of teen pregnancy, and its consequences; 2) the effects of improving adolescent knowledge, increasing access to contraception, and improving parent/child communication; and 3) the characteristics of effective programs. The on-going evaluation of sex and HIV education programs coupled with creativity and perseverance on the part of program developers led to two groups of effective programs--sex and HIV education programs that reduce sexual risk-taking behavior, and youth development programs that reduce teen-age pregnancy and childbearing.