Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We experimentally disentangle two potential sources for endogenous social interactions effects. By comparing groups where the aggregate behavior is publicly observable with those where it is not we can measure the size of any endogenous observation effect. By comparing connected with disconnected groups we can measure the size of any endogenous contagion effect. Results are provided for both a coordination game and social dilemma. We find strong evidence of an endogenous observation effect in the coordination game but not social dilemma. We find no evidence of an endogenous contagion effect in either game. While our results point towards a conformity effect we argue that information on group behavior primarily acts as a coordinating device which may be reflected in changes in beliefs.