Comparing four designs of social comparison feedback to increase physical activity


Physical activity is associated with many health benefits including reduced rates of cardiovascular disease, diabetes, hypertension, and obesity.  However, more than 50% of adults in the United States do not achieve enough regular physical activity to obtain these benefits. Many wellness programs and mobile apps use leaderboards to compare progress across individuals and groups of people, but the best way to design these social comparisons is unknown.


We designed a 6-month randomized controlled trial to test different ways of designing social comparison feedback with financial incentives to increase physical activity. We partnered with the University of Pennsylvania Health System and enrolled 280 employees. All participants used their smartphones to track activity and formed teams of four persons.  Each team was randomly assigned to one of four arms.  Teams received social comparisons to either the 50th percentile (normative feedback) or the 75th percentile (top performers or leaderboard type feedback), and either with or without forgiveness of their two lowest-performing days of the week. All arms were entered into a weekly regret lottery, and only received winnings if the team's average daily step count met the step goal.


No significant difference was found between the four arms in either this intervention or follow up period. Neither the reference point of the social comparison feedback (50th vs. 75th percentile) nor forgiveness led to a significant difference in achieving the daily step goal. Future research could test variations of social comparison feedback without financial incentives.


National Institute on Aging (NIA), NIH