High dropout rates and low pass rates are prevalent problems encountered by online learning platforms, which greatly hinder the development of online education. Drawing upon the theory of attention allocation, this study aimed to investigate the factors influencing the effectiveness of Massive Open Online Courses (MOOCs), as well as the potential moderating effects. To address the limitation of using course completion rates as an overall measurement, this study endeavours to measure MOOCs learning effectiveness by examining dropout rates and pass rates as separate outcome indicators. We use secondary data analysis to investigate our research questions. Specifically, we collect 8602 courses from a Chinese MOOC platform (Zhihuishu) using data-crawling techniques, and employ regression analyses to examine our research hypotheses. The findings indicate that course quality, content richness and interactivity significantly influence course dropout rates and pass rates. Besides, content richness moderates the relationship between course quality and learning effectiveness. Furthermore, frequent online interaction is associated with lower pass rates in high-quality courses, but the moderating effect of online interaction on dropout rates is insignificant. This study contributes to the extant literature by examining course-level factors that affect learning effectiveness. It also offers new theoretical insights and provides valuable suggestions for the design of MOOCs.