Penn study finds online peer support increases physical activity

A Penn study published recently in the journal Preventive Medicine shows that online peer networks can be powerful motivators to get people to stay committed to adding more physical activity to their lives.

Damon Centola, an associate professor in the Annenberg School for Communication, co-authored the study with Ph.D. students Jingwen Zhang, Devon Brackbill, and Sijia Yang.


The research team conducted a randomized controlled trial last year to determine what features of social media—promotional messaging or peer networks—would be more effective in increasing physical activity through participation in exercise classes.

“Our goal of the study was to really understand what could make a difference,” Centola says.

The team created a website and recruited 217 Penn graduate students to enroll in a 13-week free exercise program at the University’s Pottruck Health and Fitness Center. The study divided the students into three different groups.

The first was the control group, which received no follow-up interventions.

The second group was encouraged to attend exercise classes through conventional social networking promotional messaging, fitness tips, videos, and infographics that stressed the importance of an active lifestyle.

The third group received no messaging. Instead, group members were randomly assigned to an online network of six anonymous peers, or “health buddies,” from the program. Participants received regular online updates about the fitness activities of their peers. For example, when a health buddy signed up for a weightlifting or yoga class, all of the peers in his or her network were notified. Group members could also monitor when peers signed up for exercise classes via the study’s website.

To encourage group members to exercise more, participants only received live web updates about their peers’ positive exercise behavior, a model developed through Centola’s prior research on the dynamics of group behavior. If a peer skipped a class, for example, an online notification was not sent.

By the end of the study, the findings were clear: Social influence from anonymous online peers was far more successful than promotional messages in increasing the physical activity of study participants.

The group with program-assigned “buddies” worked out 1.6 days per week more than at the beginning of the study, about twice as much as members of the other two groups.

In the group that received the promotional messages, there was an initial spike in class attendance after the messages were sent at the beginning of the study, but as the weeks went on, the effect of the messages declined.

“Overall,” says Centola, “the study shows that online peers are not only more influential for getting people to become active, but they are also surprisingly effective at keeping them coming back for more.”

Online Peer Exercise 145