Three Essays on Behaviors in Crowdsourcing Communities

Over the last decade, the interest in studying crowdsourcing platforms has dramatically increased among researchers in the information systems (IS) area. My dissertation includes three essays on crowdsourcing. The first essay provides an extensive literature review on crowdsourcing and shows several gaps in the literature. The second essay highlights the gaps in the literature on innovation contest platforms and draws on the organizational learning literature to fill the gaps. In this research, I draw on the organizational learning literature to study the undelying dynamics of specialization and variation in innovation contest platforms. Using data of nearly 35000 users\u27 participation history in competitions hosted on Kaggle, I find that both specialization and variation play key roles in individuals\u27 performance in innovation contest platforms and reinforce each other. My findings reveal that their influence on performance is contingent on contest characteristics. The third essay investigates idea quality and the number of submitted ideas on idea crowdsourcing platforms for new product development (NPD). My review of the literature in this area reveals that the relationships between interacting with peers’ ideas through idea evaluation performance and cognitive effort on feedbacks to peers\u27 ideas and the quality of past ideas on idea quality and the number of submitted ideas that have not been studied in the literature. To fill these gaps, I draw on the literature on idea crowdsourcing, creativity, attention, and goal attainment to develop hypotheses regarding the role of these factors on idea quality and the number of submitted ideas in idea crowdsourcing platforms for NPD. Based on the data collected from Lego Ideas, I show that users with better idea evaluation performance generate higher quality ideas and submit fewer ideas. I also find that cognitive effort on feedback to peers’ ideas has a significant positive impact on the number of submitted ideas. Furthermore, the results indicate an inverted U-shape relationship between the average quality of past ideas and idea quality and the number of submitted ideas. Besides filling several gaps in the literature, my research findings offer several implications for crowdsourcing platform sponsors to design better platforms