UNIVERSITY PARK, Pa. — Adolescents may get by with a little help from their friends, but, according to Penn State researchers, friend selection and friend influence, as well as gender, may all play a role in establishing friendships that can help, or possibly hurt, them.

In a study of adolescent friendship networks, the researchers found that for both boys and girls, friend selection and friend influence guided the adoption of risky behaviors, said Cassie McMillan, a doctoral student in sociology and criminology, Penn State. Gender also played a role in how selection and influence shape a few of those behaviors, including smoking and delinquency, she added.

“There are a lot of questions about whether peer influence can explain why kids tend to have similar negative behaviors to their friends, or whether it’s the result of friend selection, which is when an adolescent goes out and seeks friends who do similar things,” said McMillan. “In this study, we found that, yes, both peer influence and friend selection are shaping the same risky behavior patterns that we see in friendships.”

The researchers, who report their findings in a recent issue of Social Networks, found that, compared to boys, girls tend to be more influenced by their friends to engage in delinquent behavior, which includes skipping class and fighting. They also found that girls tend to select friends who share similar smoking habits to their own.

There are a few reasons why influence and selection guide some behaviors for girls, more so than boys, said McMillan who worked with Diane Felmlee, professor of sociology, and D. Wayne Osgood, professor emeritus of criminology and sociology, both of Penn State.

“We think that for peer influence, the gendered patterns have a lot to do with the characteristics of the behavior, rather than the character or the structure of the friendships that girls and boys have,” said McMillan. “Delinquency — especially getting into fights and stealing — is often portrayed in the media as more masculine behavior, so this may be a more normative script for boys. Boys don’t necessarily need the influence from their friends in order to get involved in these types of activities, while girls may need another force, such as influence, to get pushed into becoming involved in delinquency.”

In addition, because smoking is such a visible behavior, girls likely seek out friends who have similar smoking habits, the researchers suggest.

“Smoking is noticeable on school campuses, so kids are more apt to be more aware of who the smokers are in a school,” said McMillan. “Girls are socialized from a very young age to be quite selective about who their friends are. Perhaps that’s why girls are more likely to select friends who have similar smoking — and nonsmoking — behavior.”

Learning more about how friendship networks shape adolescent behavior could lead to better prevention strategies, especially for at-risk kids, McMillan said.

“While some earlier work suggested that boys might be immune to these influences, one of the most important takeaways for us from the study is that both boys and girls are subject to peer influence and friend selection, so prevention programs should keep teaching both boys and girls these skills,” McMillan said.

While risky behaviors were examined in this study, McMillan said that friendship networks may also shape positive behaviors, which could be used in future prevention and intervention programs. She added that the findings could also help future research aimed at tailoring those programs.

“For example, because girls are more prone to influence for delinquency, it would be good to show examples in a prevention campaign of girls resisting peer pressure when it comes to delinquent behavior,” said McMillan. “I also think there’s a lot of value in doing group-based prevention strategies that focus on changing the behaviors of groups of friends.”

The researchers used data from 13,214 students who participated in the PROSPER — Promoting School-Community Partnerships to Enhance Resilience — study. Participants attended school in one of 28 small public school districts from sixth grade to ninth grade. The school districts were located in rural communities or small cities in Iowa and Pennsylvania.

The first wave of students entered the study in 2002. Participants took self-administered surveys in the fall and spring of their sixth grade and then in the spring of each year between seventh and ninth grades. To establish friendship networks among the students, participants were asked to identify their best and closest friends in their grade.

The W.T. Grant Foundation, the National Institute of Drug Abuse and the National Science Foundation supported this work.