Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We use experiments to test the incentive compatibility of the “random problem selection” payment mechanism, in which only one choice out of many is randomly chosen for payment. We find that the mechanism is not incentive compatible when all decisions are shown together in a single list. But when the rows of the list are randomized and shown on separate screens, incentive compatibility is restored. This causes more apparent intransitivities in choice (“multiple switching”), but, since the experiment is incentive compatible, these intransitivities must be inherent in subjects’ preferences.