Peruvian secondary school students' perceptions of their peers' sexual activity is one of the most consistent predictors of their own behavior, according to results of a survey conducted in nine cities.¹ For example, males who said that many of their friends had had sex were more likely than those who thought that none had done so to be sexually experienced and to have had multiple partners in the past three months; however, they also were more likely than others to have used a condom at first intercourse. Self-esteem and socioeconomic status also play an independent role in determining whether young people will engage in risky or protective sexual behavior, but students' knowledge of reproductive health issues is not a factor.

The survey was conducted in late 1998 as part of an evaluation of Peru's national sexuality education program, launched in 1996. In all, 6,962 students from 38 schools participated. Schools and students were selected through two-stage cluster sampling in Lima and eight other cities that represent the country's major ecological regions (coast, mountains and jungle).

Female students outnumbered males by almost two to one because a disproportionate number of all-female schools were included; the researchers note that since they analyzed females and males separately, this disparity does not affect the results. One-third of participants were from Lima, and the rest were distributed about equally among the other regions. Two-fifths of students (mostly 13-14-year-olds) were in their first two years of secondary school; the remainder (who were typically 15-18 years old) were in their last three years.

Respondents completed a questionnaire that explored their knowledge of and attitudes toward issues covered in the sexuality education program, as well as a wide array of background and psychosocial characteristics. All students were asked if they had ever had sex; those in the upper three grades also were asked the number of partners they had had in the last three months and whether they had used condoms at first and last intercourse. The researchers conducted bivariate analyses to identify factors that were significantly related to students' sexual behavior and condom use, and then used a variety of multivariate techniques to isolate the inde- pendent effects of these factors.

The bivariate analyses yielded several striking findings--particularly that students' knowledge of reproductive health risks and how to avoid them, perceptions of gender roles, and frequency of attendance at religious services were not associated with their sexual behavior or condom use. According to the researchers, the lack of an association for the first two of these factors probably reflects that participants uniformly reported themselves to be highly knowledgeable about reproductive health and quite "modern" with respect to gender role norms; the finding on religion, the investigators add, is more difficult to explain.

Sexual Behavior

Seventeen percent of all respondents--32% of males and 7% of females--had ever had sex. Results of a hazard regression analysis indicate that students' perceptions of their peers' behavior (regardless of the accuracy of those perceptions) were a strong predictor of whether they had had intercourse: Participants who said that a few or many of their friends had had sex were at least twice as likely as others to be sexually experienced themselves (hazards ratios, 2.0-2.3 for males and 2.5 for females); for males, the perception that many friends had been involved in a pregnancy also raised the likelihood of sexual experience (1.7).

Self-esteem was another key factor determining young people's sexual behavior: For both males and females, respondents who felt that they were always or almost always important to those they lived with were less likely to have ever had sex than were those who thought this was never the case (hazards ratios, 0.5-0.8). In addition, the likelihood of being sexually experienced was elevated for young men who said that they were almost always leaders among their friends (1.3) and was reduced among young women who always told friends when they disagreed with them (0.7).

Most background characteristics were not associated with students' having initiated sexual intercourse. However, women of high socioeconomic status were less likely than those in the lowest stratum to be sexually experienced (0.6), and men who lived in jungle areas were considerably more likely than residents of Lima to have had intercourse (2.6).

Overall, 5% of respondents (10% of males and 2% of females) had had intercourse in the three months before the survey. More than half of these young men (54%) had had two partners or more during that time, but nearly all of the women (91%) had had only one. The number of females with recent sexual experience was too small for further analysis, but the researchers assessed the determinants of multiple partners among males, using tobit regression analysis. Once again, peer norms and self-esteem played a substantial role in men's behavior: Males who perceived their friends to be sexually experienced or to have been involved in a pregnancy were more likely than others to have had multiple partners in the past three months (coefficients, 1.1-2.2), as were those who were almost always leaders among their friends (0.5).

Young men from the highest socioeconomic level and those who worked had an increased likelihood of having had more than one partner (coefficients, 0.6-0.8), and the likelihood of this behavior rose with age (0.8). Males living in the jungle were at greater risk of having had multiple partners than were their counterparts from Lima (1.9).

Condom Use

Among sexually active students, 38% of males and 26% of females said that they had used a condom at first intercourse; 63% and 42%, respectively, reported having used one at last intercourse. The researchers used logit regression analysis to examine the determinants of condom use, again excluding women because of the small number with sexual experience.

Men from the middle and high socioeconomic tiers were significantly more likely than those with low socioeconomic status to have used a condom at first intercourse (odds ratios, 1.6-2.2), and those who lived with neither parent were significantly less likely than those from two-parent households to have done so (0.3).

Only one factor related to perceptions of peers' behaviors influenced condom use at first intercourse: Men who reported that many of their friends were sexually experienced were significantly more likely than those who reported no sexually experienced friends to have used a condom at first sex (1.8). The investigators point out that this finding of a protective influence stands in contrast to the findings on sexual activity, which suggest that males' perceptions that their friends were sexually active led to risky behavior. A plausible explanation, they note, is that young males anticipating their first intercourse were advised to use a condom by more experienced peers.

Finally, one measure of self-esteem was associated with increased odds of having used a condom at first intercourse. Students who said that they almost always needed guidance when they had a problem were significantly more likely than those reporting that they never needed guidance to have used a condom the first time they had sex (1.6).

Findings regarding condom use at most recent intercourse were essentially the same as those for use at first intercourse. A notable exception is that whereas a young man's perception that many of his friends were sexually experienced was associated with increased odds of use at first sex, it was associated with decreased odds of use at last sex. In the researchers' view, this result may point to changing effects of peer norms as young men become more sexually experienced, or it may reflect that the number of males reporting recent intercourse (250) was too small to yield reliable results.

Conclusion

In light of their findings, the researchers emphasize that programs directed solely at improving young people's knowledge of reproductive health "may not be sufficient to avert teens' risk-taking in settings presenting a large number of other risk factors." Rather, they suggest, expanding programs' focus "to target some of the contextual factors that influence adolescent behavior is likely to enhance [their] impact."--D. Hollander

REFERENCE

1. Magnani RJ et al., Correlates of sexual activity and condom use among secondary-school students in urban Peru, Studies in Family Planning, 2001, 32(1):53-66.