Empower adolescents to change their own behavior in school
By James Dean, Cornell Chronicle
A large-scale program that enlisted students in disadvantaged middle schools to teach younger peers reduced disciplinary problems and improved academic achievement, reports new research led by a Cornell economist.
Key to the program’s success: buy-in from challenging adolescents who crave status and don’t want to be lectured – an approach informed by the science of adolescent behavior and brain development, the researchers said.
The program already has been scaled up in Turkey, where the two-year study collected data from roughly 18,000 students in 65 schools, starting in 2020-21. And the researchers are in discussions with educators from several countries, including in U.S. schools, interested in applying the relatively low-cost intervention to a variety of contexts.
“We show that you can change behavior on a massive scale, including for the kids whose behavior you want to change and also for their friends, because of the social linkages in middle schools,” said Sule Alan, professor in the Cornell Jeb E. Brooks School of Public Policy and in the Department of Economics. “You have to allow them to change their own behavior willingly. You have to give them what they actually need, and that is social status, that is autonomy, and that is respect.”
Alan is a co-author of “Empowering Adolescents to Transform Schools: Lessons from a Behavioral Targeting,” published in the February issue of the American Economic Review, with Elif Kubilay, a senior lecturer in the Department of Economics at the University of Essex in England.
Alan investigates how school environments shape malleable character traits – such as grit, patience, self-control and belonging – that may contribute to diverging outcomes for similar students. Boosting those social and emotional attributes could benefit learning and lead to better labor market outcomes, particularly in lower-income communities.
Alan spent a year meeting with “problem” students in disadvantaged middle schools to understand their needs. They made clear, she said, that they weren’t interested in another adult telling them how to behave.
Listening to their feedback and reviewing the psychology literature about adolescents, Alan adopted a new approach. Instead of training teachers to target certain behaviors, as she had with younger children, she would ask selected seventh and eighth graders to deliver an empowerment curriculum to fifth- and sixth-grade peers.
“I said, ‘OK, I’m not going to teach you any lessons, but can you help teach the little ones in your school how to behave, how to be responsible, how to make the school better?’” Alan said. “Eventually, through self-persuasion, I thought they would start believing the messages they were teaching.”
The researchers identified senior students who scored high in emotional intelligence and were socially influential, based on friend networks the students shared. From that group, they recruited roughly 630 each in the seventh and eighth grades to be student-teachers. Including their friends, the study’s target sample totaled about 5,000 students.
In 32 randomly assigned schools, the student-teachers taught a weekly curriculum, “Our Future, Our Dream,” designed by Alan’s team. Among its nine topics were envisioning the ideal school; recognizing one’s power to shape their social environment; and understanding the perils of violence and antisocial behavior. In the remaining control group, about half of the schools did nothing differently, while the other student-teachers led basic activities such as mazes and coloring.
Results showed the program was effective in improving social, emotional and academic outcomes, the researchers said. Student-teachers and their friend networks were about 70% less likely to be flagged for disciplinary violations in the first year, and 55% less likely in the second year. Anti-social attitudes decreased. Junior students participating in the curriculum were more likely to nominate senior students as supportive peers, suggesting an improved school climate.
Most importantly, Alan said, eighth-grade student-teachers engaged in the curriculum were significantly more likely to win admission to selective high schools, potentially imparting long-term economic benefits. In 2022, 18.4% of student-teachers were admitted to selective schools – twice as many as eighth graders overall. In 2023, 21.2% of student-teachers advanced to selective schools, compared to 13% overall.
The researchers believe processes of cognitive dissonance and self-persuasion underlie the transformation.
“If I am a bully and I’m supposed to teach about the profile of a bully,” Alan said, “that creates discomfort that has to be resolved.”
Alan said the proof of concept can now be scaled up and applied to a range of behaviors, from drug use to smartphone use to sexual activity. And it can be done anywhere in the world, since adolescent development and behavior is similar across cultures.
“You need to understand how the adolescent brain works and design programs that would work for them,” Alan said. “Adolescents should embrace the program that you’re offering them to change their behavior.”
The research was supported by a European Research Council Advanced Grant.
Media Contact
Get Cornell news delivered right to your inbox.
Subscribe