Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We conducted a field experiment with a charitable group to investigate whether giving the donor an option to write a personalized holiday card to the recipient influences giving behavior. Over 1500 households were approached in a door-to-door campaign and randomized to either a treatment group, in which donors were presented with the option to write their own card for the recipient, or a control group, in which donors were not given the option to write their own card for the recipient. We predict that treatment should increase contributions through making the gift more meaningful, but may also decrease contributions by increasing the transaction and social costs of donating. We find evidence in favor of the negative effects of costs from treatment, and no evidence of increased giving. We also observe that our treatment crowds out small donors (donors giving $5 or less).