This project involves a randomized field experiment. The investigators worked with the vending contractor for the City of Philadelphia to randomize vending machine locations (covering \~260 beverage and snack machines) to one of the four different message conditions: 1) single green traffic light labels (on healthy foods); 2) multiple traffic light labels (red traffic light labels applied to less healthy foods, yellow applied to moderately healthy foods, and green on healthy foods); 3) physical activity labels (calorie content displayed in terms of physical activity equivalents); 4) posters reminding consumers of the Philadelphia sweetened beverage tax and encouraging healthier choices. The City's vending machine contractors provided monthly aggregated sales data from each machine for analysis. Customer intercepts were also collected at the point of sale, after customers completed their purchase, to evaluate factors that moderate message influence. They were asked to answer questions about their perceptions of the different labels.
