Abstract Recently, researchers have been interested in studying positive psychological capital or simply PsyCap at the group-level of analysis. All studies have focused on the effect of group-level PsyCap on work group outcomes. There were no studies have conducted on the effect of group-level PsyCap on student group-related outcomes. The objective of this study was to examine the effect of group-level PsyCap on group effectiveness of student groups working together in the subject group assignment. This study also investigated the effect of group-level PsyCap on group effectiveness when using different data aggregation models of group-level PsyCap. The participants were 1,349 undergraduate students (303 student groups) from 10 universities in Thailand in the academic year 2013. Structural equation modeling (SEM) was employed to test the proposed models. Results showed that the models of the effect of group-level PsyCap, using the direct consensus model and the referent-shift consensus model, on group effectiveness were fitted to empirical data. Group-level PsyCap, based on the direct consensus model and the referent-shift consensus model, had significantly positive effect on group effectiveness.