We study gender differences in social norms as an explanation for gender differences in economic behavior in a variety of contexts. We also study beliefs about gender differences and whether these beliefs are correct.
Experimental design
Variants of a dictator game, a trust game, and a lying game with spectators; variation of spectator & interaction partners' gender
Design details
Randomization method
Random assignment of participants to experimental condition in Qualtrics.
Randomization unit
Assignment to experimental condition is randomized at the individual level.
Clusters
No clustering: the experiment is one-shot and there is no interaction before choices are made.
Observations
Dictator Game: 800 individual observations as spectators, 400 male and 400 female participants.
Trust Game: 800 individual observations as spectators, 400 male and 400 female participants.
Lying Game: 800 individual observations as spectators, 400 male and 400 female participants.
Before sampling spectators, we sample 20 male and 20 female participants as players in each role in each 2x2 gender condition in the DG and TG, resulting in 20 player pairs in each condition. In the one-player lying game, we sample 40 male and 40 female participants. If we do not observe at least one norm violation in each condition (i.e. not sharing anything, not returning anything, and lying), we continue to sample 10 extra pairs (or individual observations in the lying game) until we observe a norm violation, which we can then show to the spectators.
In a between-subjects design, each participant only participates once, in one game and one condition.
(Updated 19/02/2021: Before conducting the Lying Game, we increase the registered sample size for the LG to the sample sizes of the DG and TG, which enables us to investigate main effects with the same power in all three games.)
Arms
Dictator Game: 100 male and 100 female participants as spectators in each of the four conditions: MM, MF, FM, and FF.
Trust Game: 100 male and 100 female participants as spectators in each of the four conditions: MM, MF, FM, and FF.
Lying Game: 200 male and 200 female participants as spectators in each of the two conditions: M and F.
Condition names reflect gender composition in the player pairs, see "Planned Number of Observations" for number of participants as players in the respective conditions.
(Updated 19/02/2021: Before conducting the Lying Game, we increase the registered sample size for the LG to the sample sizes of the DG and TG, which enables us to investigate main effects with the same power in all three games.)
Primary outcomes
We explore gender differences in:
- social norms with respect to altruism, trustworthiness/positive reciprocity, and honesty
- emotions, especially surprise and anger, towards norm violations
- empirical expectations / beliefs about behavior
and whether beliefs are correct.
Cited by
No citation data available for this registration yet.