Group-based trajectory modeling to identify effect of badges in online Q&A communities
研究通过群体轨迹建模分析2,055名用户数据，发现徽章激励对用户行为的影响随时间变化，且对低参与度用户调整徽章数量比提升整体水平更有效。
Purpose The study aims to investigate how badge incentives influence systems shape longitudinal behavioral patterns in online knowledge communities. While prior research based on the goal-gradient hypothesis has documented temporary increases in user contributions preceding badge attainment, the mechanisms underlying sustained behavioral evolution remain underexplored. Design/methodology/approach We employed group-based trajectory modeling to identify distinct behavioral pattern (N = 2,055) through longitudinal analysis of user activity records. Users’ posting frequency before and after badge acquisition was systematically tracked to observe the steering effect across global and subgroup levels. Multinomial logistic regression analyses examined co-evolutionary relationships between behavioral trajectories and badge incentives. Findings The results reveal three key insights: 1) Badge attainment exerts time-dependent steering effects that differentially influence engagement trajectories across user subgroups. 2) The evolution of user behavior has a close positive correlation with badge incentive, in addition to being influenced by user knowledge characteristics. 3) Strategic adjustment of the number of badges rather than the overall level shows greater potential for activating low-engagement users. Originality/value This research advances incentive mechanism literature by 1) developing a dynamic framework for modeling behavioral trajectories in reward systems, 2) demonstrating heterogeneous badge effects across user segments, especially for low-engagement users and 3) proposing evidence-based guidelines for optimizing badge architectures. The findings provide actionable insights for platform designers seeking to balance engagement boosting and quality maintenance through targeted incentive engineering.