No matter how diligent teachers and administrators are, it’s easy for bullying to happen under the noses of adults at school. In the bathrooms, the hallways, and on social media, students are often the only ones around to police themselves.
That’s why researchers at Princeton, Rutgers, and Yale universities are analyzing middle schoolers’ social networks to find the students most likely to change their classmates’ attitudes around bullying. They are finding that bullying is generally driven not by a few bad apples but by a majority of students within the overall culture of a school. Shifting alliances and cycles of harassment and retribution can all play into that culture, and undercut adults’ anti-bullying campaigns.
“Adult-identified leaders are often very different from student-identified leaders,” said Hana Shepherd, an assistant sociology professor at Rutgers University. “Adults look at traditionally defined ‘popular’ kids, the ‘good’ kids, while kids who are leaders of smaller groups might not be on the social radar of adults, but often are [influential] too.”
During the 2012-13, school year, Shepherd, Elizabeth L. Paluck, a Princeton psychology professor, and Peter M. Aronow of Yale University repeatedly surveyed more than 24,000 students across 56 middle schools about the students they respected most and liked spending time with online and in person, out of a list of every student in their schools. They also asked students to list peers they had conflicts with, and the social norms in each school around behaviors shown to increase conflict, such as retaliating on behalf of a friend who has been bullied.
Not Just ‘Popular’ Kids
The researchers used the data to create network maps of student friendships in each school, identifying not just the most popular students or those whom teachers considered leaders, but the students who are most influential to different peer groups throughout the school. Of those so-called “seed” students, the researchers randomly invited half to participate in the Roots program, an anti-bullying program intended to support students in recognizing and finding ways to improve their own school climate around bullying.
In a study published late last fall, the researchers found schools using Roots had 30 percent fewer discipline reports on student conflict than similar schools not using the program.
Through 10 sessions over the course of the 2012-13 school year, Paluck said a “breakfast club” of influential students in different cliques in each school met to think out their own responses to bullying and discuss ways to reduce peer conflicts. Even the wording mattered. Rather than discussing “bullying"—a term that prior research has shown is linked more with stereotypes of physical intimidation and lunch-money theft—the students typically referred to conflicts as “drama.” Students talked through exercises among themselves about how they would respond if they either saw or heard about conflicts among students.
Critically, the seed students also discussed how their peers would react to their responses, and how they could influence their classmates better. In each school, students came up with their own projects, such as creating positive GIFs, or looping animations, for Instagram or handing out wristbands to reward a student who is seen de-escalating a fight or supporting a bullying victim.
“We treated students as politicians, campaigning for a better school. The theory was that their public behaviors and statements could change norms,” Paluck said.
Results in Brief
It seems they did. Across the nearly 12,000 students in schools using the program, the total number of discipline incidents fell from 2,695 to 2,012 in one school year.
By the end of the 2012-13 school year, the researchers found the rate of reported disciplinary incidents for student conflicts—such as bullying or harassment—was 0.2 times per student in schools that did not participate in Roots, but .06 times per student in the 23 middle schools that used the program. Students in Roots schools were also significantly more likely to report they had talked with friends about ways to reduce conflict. In schools where at least 20 percent of the influential students opted to join the program, the average rate of discipline reports dropped by 60 percent.
Peer pressure has a long history as a lever in interventions, but most such strategies have focused on direct interventions by adults: role-playing with students on how to intervene if they see bullying, or displaying statistics about classmates’ views on harassment or drug use to convince students that it is not widely condoned, for example. Allowing the students to decide their own approaches to changing school culture has also enabled the researchers to dig into how peer influence really works in adolescence—the most socially focused period of child development.
The researchers are continuing to dig into what they say is now the largest set of longitudinal data on student social networks and bullying, looking at the roles of gender and teacher opinions. The New Jersey education department included the program in a teacher-training meeting this summer, but it’s unclear how many schools are using it this year.