• the Purpose of the article
• a description of the Methods used
• a summary of the Findings, and
• Implications for Practice

This study was conducted to understand the role of peer norms in determining whether young adolescents will initiate sexual intercourse.

Some 1400 6th graders from an urban public school district in the northeast U.S. completed questionnaires twice — once at the beginning and again at the end of their school year. The questionnaires included demographic information (e.g., age, gender, race/ethnicity), behavior information (e.g., ever had sexual intercourse), information about intentions (e.g., likelihood that a participant will have sex in the next year), and perceptions of peer norms (e.g., perceived prevalence of sexual intercourse among participants' peers).

Statistical analyses were conducted to determine what influenced 6th graders to initiate intercourse during the school year. Therefore, the focus of the analysis was on comparing the "never" group — those who had not had sex either at Time 1 (the beginning of the school year) or Time 2 (the end of the school year) — with the "initiated" group — those who first had sex sometime between Time 1 and Time 2.

Half the participating students were male and half were female. There were 55% African-Americans, 22% whites, 12% Latinos, and 3% Asians. Most participants were 11 or 12 years old.

As is often found in studies of risk behavior, the "initiated" group differed from the "never" group demographically. Those in the "initiated" group were more likely to be older, male, African-American, attend schools with high poverty, and come from neighborhoods with a large percentage of single-family homes.

The two groups differed dramatically in terms of their intentions. Half of the participants who would go on to initiate sexual intercourse reported at Time 1 that they were likely to do so. In contrast, just 11% of the "never" group reported at Time 1 that they were likely to have sexual intercourse in the next year.

A statistical model confirmed this finding: Even more than demographic variables and other risk behaviors, the single best predictor of whether participants would initiate was their reported intention to do so.

This finding then begs the question, what determines intention? Another statistical model showed that the strongest predictor of intention was the perception of peer norms related to initiating sexual intercourse. In fact, those who believed at Time 1 that most of their peers had already had sexual intercourse were 2.5 times as likely to initiate sex themselves between Time 1 and Time 2. Drinking alcohol and believing that older peers drank alcohol were also associated with high intention to have sex, perhaps because these are markers for students who are "social innovators" — individuals who are interested in trying new, adult behaviors.

The study reported additional findings that may be of interest, including a role for the perception of social gain and stigma in influencing sexual initiation. Adolescents who perceived that there was a social stigma associated with early sex were less likely to initiate.

• The finding that half of the 6th graders who initiated sex during the school year had a high intention to do so at the start of the year indicates that many early sexual initiations are planned events. Young people who have not yet had sex but plan to do so represent a high-risk group who could be targeted for interventions to help them delay and/or practice safer sex.
• The strong influence that peer norms have on intentions to initiate sex suggests that interventions to alter perceptions of peer norms may be very powerful. Because we know from other studies that peer norms are frequently misperceived, programs that correct misperceptions may reduce young people's intentions to have early sex, which in turn should reduce the likelihood that they will do so.
• Beyond correcting misperceptions of peer norms, programs that target peer leaders or entire cohorts of teens may be more powerful than programs focusing on individual behavior.