Impact of incentives on behavior after incentives are removed
| Ate a serving | Number of servings eaten | Number of servings discarded | |
|---|---|---|---|
| Incentive | 0.314** [0.020] | 0.261** [0.018] | −0.203** [0.029] |
| Incentive | 0.234** [0.032] | 0.171** [0.023] | −0.166 [0.083] |
| Incentive* (fraction of school with free lunch) | 0.229* [0.084] | 0.259* [0.097] | −0.108 [0.194] |
| Incentive | 0.305** [0.019] | 0.251** [0.018] | −0.199** [0.027] |
| First two weeks after | 0.102** [0.024] | 0.067* [0.025] | −0.050 [0.034] |
| Next two weeks after | 0.013 [0.025] | −0.014 [0.027] | 0.053 [0.026] |
| Mean (baseline) | 0.332 | 0.457 | 0.494 |
↵Notes: The data in this table are based on data from the second field experiment. The unit of analysis is the student day. The first two panels are restricted to the baseline and incentive period (N=21,594). The third panel includes the baseline, incentive, and post-incentive period (N=35,851). The regressions include school and day of week fixed effects and controls for the child’s grade and gender and whether only vegetables were offered that day. Standard errors are clustered at the school level. The fraction of the school with free lunch has been centered at the 17 percent. The ** and * symbols indicate statistical significance at the 1 percent and 5 percent levels respectively.