Table 2.

Regression-estimated effects of our treatment on donated incentives in study 2

	Was the incentive donated? (1 = yes, 0 = no)
	Model 1	Model 2
Treatment	0.010* (0.004)	0.011** (0.004)
Cash earned	−0.029*** (0.001)	−0.017*** (0.001)
Male	0.009* (0.004)	0.020*** (0.004)
Sex unknown	−0.011 (0.017)	−0.010 (0.016)
Visits (during incentive period)		−0.059*** (0.004)
Treatment* Visits		0.010* (0.005)
StepUp program fixed effects	Yes	Yes
Observations	17,968	17,968
Adjusted R2	0.111	0.131

This table reports the results of two ordinary least squares regression models predicting whether a StepUp participant in our experiment chose to donate their incentives in study 2 (1 = yes, donated; 0 = no, did not donate). Model 1 shows the main effect of our treatment while model 2 shows moderation. Predictors in model 1 include an indicator for being assigned to the treatment condition, fixed effects for a participant’s StepUp program, the amount of cash earned by the participant, and indicators for the participant’s sex (female omitted). Model 2 includes two additional predictors to test for moderation. These additional predictors are the (mean-centered) number of times the participant visited the gym during the StepUp program and an interaction between this variable and assignment to our treatment condition. Robust SEs are in parentheses. *, **, and *** denote significance at the 5%, 1%, and 0.1% levels, respectively.