This study investigates the dynamics of solver engagement in online creative crowdsourcing competitions through the lens of social comparison between entries. Using data from ZBJ.com, a leading Chinese crowdsourcing platform, the authors examine how competitors’ prior entries, reviewed by seekers and incorporating both positive and negative feedback, influence subsequent solver behavior.
Drawing on data from the logo design crowdsourcing competitions held on ZBJ.com in 2020, this study categorized solvers based on their exposure to competitors’ entries with positive feedback, negative feedback, or both. Logistic regression and sentiment analysis, using Naive Bayes, CNSenti, and SnowNLP models, along with a fusion model, were employed for the analysis, with various heterogeneity and endogeneity checks to ensure result reliability.
The results demonstrate that the number of competitors’ visible entries receiving either positive or negative feedback from seekers is positively correlated with the likelihood of solvers submitting a second entry. The distance between the seeker-reviewed entries and the solvers’ second entries weakens these positive effects. When simultaneously exposed to both positive and negative feedback entries, solvers show greater attention to the positive feedback entries, indicating a “positive bias”.
This study deepens our understanding of social comparison dynamics and the phenomenon of entry visibility in creative crowdsourcing competitions. It reveals how the combination of publicly visible entries and their associated feedback influences solver engagement, as well as the moderating effect of the entry location, providing novel insights into the motivations of solvers.
