Join us for YTH Live Global! Learn More

ETR Logo ETR Logo ReCAPP logo

etr presents YTH LIVE Global 2021: October 4-6: Virtual. Register now at

Research Summaries

All Research Summaries

Peer Effects on Adolescent Sexual Debut and Pregnancy: An Analysis of a National Survey of Adolescent Girls

Original article authored by:
Peter Bearman and Hannah Brückner, Columbia University

This summary includes the following sections:


Adolescents influence their peers by modeling behaviors and setting social norms. Peer pressure is often thought to be a negative force on adolescents, but this study demonstrates that it is more often a positive one. Using data from the National Longitudinal Study of Adolescent Health (Add Health, 1994-95), the authors of this study sought to explain peer effects on girls' sexual debut — the timing of first intercourse — and pregnancy risk — the likelihood of becoming pregnant. In order to determine the independent effect of peers on these two variables, the authors controlled for other factors associated with adolescent sexual activity and pregnancy risk.


Add Health is a nationally representative, school-based sample of adolescents in grades 7-12 attending private, religious, and public schools from urban, suburban, and rural communities across the country. Questionnaires administered to 90,000 students in 141 schools over three periods of time from September 1994 to April 1995 examined parents’ educational and occupational backgrounds, household structure, risk behaviors, future goals, self-esteem, health status, friendships, athletic and school participation, and activities with friends. Students were also asked to name their closest female and male friends.

Follow-up in-home interviews were conducted with about 20,000 adolescents and 85% of their parents in 1995, and with nearly 15,000 adolescents in 1996. The interviews provided more in-depth data on drug and alcohol use, sexual behavior, criminal activity, health status, health care utilization, family dynamics, peer networks, romantic relationships, attitudes and decision making.



The authors’ analytic model contained three components:

  1. In order to isolate the effects of peers on sexual debut and risk of pregnancy, youth were grouped according to other factors known to influence sexual debut and risk of pregnancy. These factors include: family background and structure, parental influences, socio-demographics, and individual characteristics such as cognitive ability, academics, religion, extra-curricular activities and dating.

  2. Risk categories were created for the purpose of assessing peer influence. Respondents and their peers were categorized as either high-risk or low-risk based on indicators that measured their orientation to school and participation in risky behaviors. Some specific examples of indicators were grade point average, academic aspirations, extra-curricular activities, and behaviors such as skipping school, drinking or getting drunk, getting into fights and engaging in dangerous behaviors on a dare.

  3. To best represent the complex network peers form around the time of adolescence, five levels of friendship were considered: single best friends, network of close friends, peer group or clique, the leading crowd (at the school), and the school as a whole. Peer characteristics such as risk category, age in comparison with the adolescent, and relationship with parents were examined for each level. The proportion of peers with a particular characteristic was also assessed for each level — i.e. the proportion of older peers in the adolescent’s clique or the proportion of high-risk peers in an adolescent’s network of close friends.

For all findings discussed below, "sexual debut" is defined as the likelihood that a girl will have first intercourse during the 18-month follow-up period of the study. "Pregnancy risk" is defined as the likelihood she will become pregnant during the same time period.



  • Best Friends
    Best friends are not as important as originally thought. Having a high-risk female best friend increases the likelihood of sexual debut, regardless of the respondent’s own risk status, but otherwise poses no significant additional risk of pregnancy. On the other hand, having a low-risk male best friend decreases the likelihood of pregnancy.

  • Network of Close Friends
    The composition of a girl’s circle of close friends is an important indicator for sexual debut. Having younger friends and friends who have good relationships with their parents lower a girl’s risk of sexual debut. Likewise, a circle composed primarily of low-risk male and female friends is a protective factor for both sexual debut and pregnancy. One of the more surprising findings for the authors was that a girl’s own risk status is not as important as that of her close friends. Regardless of her own risk status, a girl with high-risk male friends is at greater risk of pregnancy, and a girl with low-risk female friends finds her pregnancy risk reduced.

  • Peer Group or Clique
    The larger peer group, or clique, exerts mixed influences on sexual debut and pregnancy. As the number of high-risk members in a girl’s peer group rises, so does her risk of sexual debut. Thus, a clique composed primarily of high-risk members is considered to be a negative influence. Similar, but not equal, is the effect of the peer group on pregnancy risk. Here, as the number of low-risk members in her peer group rises, a girl’s chances of getting pregnant decrease. However, it should not be assumed that this means having high-risk members in one’s clique increases pregnancy risk; such an association was not seen in this study.

  • Leading Crowd and School as a Whole
    There were virtually no findings that peers at these two levels affect sexual debut or pregnancy. The only significant finding is that greater numbers of sexually active students at a school increase a girl’s likelihood of sexual debut.



The authors derive the following conclusions from their findings:

  1. Peer influence operates at many levels, and the network of close friends and the larger peer group have more significant effects on the female adolescent than do best friends.

  2. Peer influence is most often positive. One possible explanation for this finding is that adolescent girls often possess the necessary skills to filter the negative influences of their high-risk friends and benefit from the protective influences of low-risk friends.

  3. Some characteristics of friends appear to be of equal or greater importance as those of the individual in determining sexual debut and pregnancy risk. For instance, friends' risk status is a better determinant of pregnancy risk than the adolescent's own risk status, and the same holds true for friends' ages and sexual debuts. Further, friends’ relationships with their parents are as important as a girl’s relationship with her own parents.

  4. Male and female friends have different influences on sexual debut and pregnancy risk. Female best friends’ risk status is associated with sexual debut while male friends’ risk status is associated with pregnancy risk.

  5. Findings here indicate no effect of leading crowds on individual sexual behavior and few significant effects at the school level. This finding was surprising for the authors. Many working in peer programs believe that leading crowds shape others’ sexual behavior.


Implications for Researchers, Educators, and Parents

  • Because the network of close friends and the larger peer group are more influential than best friends, peer intervention programs to reduce risky sexual behavior may be more successful if they incorporate peers from the target’s larger peer group into their interventions.

  • Programs aiming to delay the initiation of sexual intercourse among adolescent females should consider the risk-status effects of the larger peer group and focus on groups with large numbers of high-risk youth.

  • Parents could worry less about their daughters' one or two high-risk girlfriends and increase efforts to support relationships with low-risk friends. They should also become familiar with their daughters' larger circle of friends, on the relationships those friends have with their parents, and on continuing to build a good relationship with their own daughters.

  • Future research should focus on understanding the nature and mechanisms of influence in larger peer groups.

Bearman, Peter and Brückner, Hannah. Peer Effects on Adolescent Sexual Debut and Pregnancy: An Analysis of a National Survey of Adolescent Girls. The National Campaign for the Prevention of Teen Pregnancy, April 1999.