Deposit? Yes, please! The effect of different modes of assigning reward-and deposit-based financial incentives on effort 319 9 Appendix C. Robustness checks for selective dropout Figure 1 shows that only very few respondents completed the experiment that were in the low payment condition and had chosen or were randomized to deposit-based incentives. In this Appendix we ran a set of test to demonstrate the robustness of our results to this selective drop-out effect. In particular, we report the following: a. A set of analyses that compare the characteristics of the respondents that showed up for the second session with low-paying deposit-based incentives to two groups. First, we compare these respondents to respondents with low-paying deposit-based incentives that dropped out of the experiment between session T0 and T1. Second, seeing as with so few observations the power of these tests may be low, we contrast the low-payment depositbased incentives respondents to the remaining respondents that showed up for session T1. The characteristics of all 3 groups are found in Table C1. It appears that any potential selection occurs at T0, as only 20 out of 67 (30%) low payment participants are within the deposit arm. In both other payment conditions the distribution between lumpsum and deposit is approximately even. Nonetheless, there are no significant differences between those 20 participants and all other respondents that showed up for session T1. b. The regression analyses reported in Table 3 were defined to take low payment as the reference-case. As such, some of the estimated fixed effects could be biased if there is selective drop-out. As such, we also reprint Table 3 with the medium payment condition as a reference-case, here shown as Table C2. Comparing Table 3 and Table C2 no major differences are observed.
RkJQdWJsaXNoZXIy MTk4NDMw