Winning by Learning? An Exploratory Study of Knowledge Sharing and Usage on Crowdsourcing Platforms

A crowdsourcing platform enables solution seekers to use contributions from a group of online users, usually by organizing crowdsourcing contests through which online users contribute to the solution generation process. Knowledge sharing on a crowdsourcing platform could play an important role in the process of crowdsourcing contests and contestants generating high-quality solutions. On one hand, more knowledge resources could lower the participation cost and help improve crowdsourcing outcomes. On the other hand, the shared knowledge may also interrupt contestants’ independent solution search processes and limit contestants’ creativity. Specifically, knowledge sharing can affect crowdsourcing outcomes via the parallel path effect and stimulating effect in crowdsourcing contests. This study provides an in-depth analysis on how on a crowdsourcing platform the knowledge sharing process impacts contestants' innovation process and contest outcome quality. We first examine how different dimensions of the shared knowledge affect crowdsourcing contests. We find that high knowledge sharing originality negatively impacts the outcome by decreasing the parallel path effect and stimulating effect, while high knowledge sharing quality positively impacts the outcome by increasing the stimulating effect. Subsequent analyses using a finite mixture model indicate that contestants vary in their ways of using knowledge and hence are influenced by knowledge sharing differently.