Contingent Prizes in Dynamic Contests
研究了在动态在线竞赛中，根据比赛历史发放的或有奖励如何激励参与者持续努力，并通过结构模型和随机对照试验证明其能显著改善竞赛结果。
ABSTRACT Firms and government agencies increasingly use online contests to find solutions to problems. These contests allow players to make multiple attempts, and their progress is tracked on a real‐time public leaderboard. Although the contests are dynamic , contest sponsors miss the opportunity to dynamically shape players' incentives, as they award the entire prize pool based only on players' final rankings. We propose the use of contingent prizes that depend on the game's history to incentivize players throughout the contest. We present evidence from two methodologies—a structural model and an RCT—showing that contingent prizes can significantly improve contest outcomes.