Contact Name
Barbara Dietsch
562-799-5126
bdietsc@WestEd.org

By Carolyn Tucker Halpern

Individual differences in cognitive abilities are often mentioned as contributors to the timing of sexual debut. However, few studies have explicitly investigated this link. This is an interesting gap in the research literature, because early sexual debut often clusters with behaviors like truancy or delinquency, which have been empirically associated with differences in intelligence. In fact, there is a generic, inverse relationship between intelligence and rule-breaking that is evident for a wide range of behaviors. Research on adolescent sexual behavior has focused more on grades in school or educational aspirations as predictors, the hypothesis being that adolescents who make higher grades or have higher goals postpone intercourse because they want to protect their future. These academic predictors are related to cognitive abilities, but also tap other dimensions such as motivation and perception. If sexual postponement does reflect primarily a desire to safeguard future plans, then we would expect to see a relationship between intelligence and intercourse, and perhaps heavy petting. However, we would not expect relationships with early, noncoital activities like kissing and light petting, which do not entail obvious social or health risks.

Using data from the National Longitudinal Study of Adolescent Health (Add Health) and from the Biosocial Factors in Adolescent Development Projects, we explored two questions: 1) is there a relationship between intelligence and sexual postponement; and 2) if so, what are the processes underlying the relationship? Sexual postponement was examined in terms of both noncoital and coital behavior. Two separate data sets were used for analyses because they offered different and complementary strengths. Add Health provides a large, nationally representative sample of adolescents who were in the 7^th to 12^th grades in 1994. The Biosocial Factors study was much smaller, but provided intensive, detailed measures of coital and noncoital sexual activity for male and female adolescents who were about age 13 at study entry, and who were followed over a 3-year period beginning in 1986 and 1989, respectively.

Confidential, self-administered surveys (audio-CASI in ADD Health) were used to assess sexual experience, and the Peabody Picture Vocabulary Test (PPVT) was used as the measure of intelligence. The latter instrument correlates highly with standardized intelligence tests and can be properly administered under field conditions by interviewers without highly specialized training. The PPVT was modified for use in Add Health and the modified form (Add Health Picture Vocabulary Test — AHPVT) was standardized on that sample.

Analyses revealed that adolescents with average AHPVT scores were 1.5 to 5 times more likely to have had sex compared to teens who scored very well (one or more standard deviations above the mean) or who scored very poorly (one or more standard deviations below the mean). The relationships were the same for Black and White adolescents, but differences were more pronounced for females than for males and for older (15 years and up) adolescents. Further, higher test scores were not just associated with a postponement of sexual intercourse, but also with a postponement of early, noncoital sexual behaviors like holding hands, kissing, and light petting. However, higher intelligence was not related to less sexual interest, just with a postponement of acting on that interest.

Through statistical modeling, we explored multiple paths through which intelligence might be linked to sexual postponement. For example, we examined differences in motivation, values, situational opportunities, anticipated consequences of sexual activity, and expectations about future educational or career goals. Our analyses suggested that these factors are part of the story for adolescents who scored well on our intelligence measure, but they did not fully account for the relationship between intelligence and postponement. When this long list of possible mediators was included in statistical models of coital timing, differences for adolescents who scored highly on the intelligence instrument were diminished but still evident. For example, even after controlling for all of the above, plus pubertal maturation and attractiveness, a 15 year-old adolescent girl who scored 130 on the AHPVT was still half as likely to have had sex as a girl who scored 100 (the average score). The association between intelligence and "early" sexual behaviors like holding hands, kissing, and light petting, which have no obvious negative consequences, also suggests that other factors are at work. An idea that adolescents worry that kissing and holding hands will put them on a "slippery slope" to intercourse is hard to take seriously.

In sum, our findings are consistent with those in the delinquency literature, and indicate that there is a generalized postponement of all types of sexual activity among adolescents who score at the upper end of this intelligence measure. Postponement of coital activity was also evident for adolescents who scored poorly on the AHPVT, but we explained little of the variation in coital activity for this group, and we were unable to test relationships with noncoital behavior. There are undoubtedly many factors and processes underlying these associations, and they are probably different for adolescents at the two ends of the intelligence distribution. Conventional values and expectations are part of the story for adolescents who score above average in intelligence, but are not very informative for those who score below average. We expected that this latter group might be more vulnerable to manipulation by others, but perhaps these adolescents benefit from the protective efforts of various gatekeepers.

These findings highlight the fact that we know relatively little about adolescent decision-making processes in sexual situations. When a teen is confronted with the reality of a romantic partner under the pressures of "real time" and physical excitement, does he or she really engage in probabilistic thinking about relative risk and alternative outcomes? How does intelligence fit into the formation of romantic relationships, and into the interpersonal processes that lead to different sexual outcomes for different adolescent couples? A better understanding of these factors and processes could make a positive contribution to sexuality programs directed at both males and females.

Carolyn Tucker Halpern is Assistant Professor, Department of Maternal and Child Health, School of Public Health, University of North Carolina, Chapel Hill, NC.