How do we deal with social learning task when examining existence of caring conformism?
Situation
In this task, there is no fundamental preference clash. Preferences are aligned. The only reason to conform in this task is learning about the true state of the world. Accordingly, we do not expect any caring conformism. If I care about the others I learn from them or not but I do not conform and then return back.
Solution
Include following variables in regression
- dummy for jar
- interaction term for jar and conforming
Interpretation
Coefficient of interaction term tells me whether conforming history has different effect in the jar task. It should. In fact, the coefficient should be negative and large enough to erase the effect that we observe for the other tasks. This is a sort of robustness test to see whether we are really catching caring conformists.
Edited by Wendelin Schnedler