A longitudinal study on relationships of Chinese postgraduates’ ought-to L2 self, self-efficacy and engagement

Ought-to L2 self, self-efficacy, and L2 engagement play a key role in postgraduate students’ academic English learning. However, longitudinal research on the joint development of these three factors in this setting remains limited. This study explored the relationships among ought-to L2 self, self-efficacy, and L2 engagement in 174 Chinese master’s students across one semester. The data were gathered at three stages: early semester (T1), mid-semester (T2), and late semester (T3). Latent growth curve modeling and cross-lagged panel modeling were applied. The growth models indicated no significant mean linear change in ought-to L2 self, whereas latent-basis models yielded significantly negative mean growth factors for self-efficacy and engagement. The observed means of the latter two constructs declined at mid-semester and partly recovered at the final wave. Individual differences in change were evident for ought-to L2 self and engagement. Both ought-to L2 self and engagement demonstrated significant autoregressive stability, indicating continuity in their relative standing. The significant cross-lagged paths showed the following pattern: T1 ought-to L2 self positively predicted T2 engagement but negatively predicted T2 self-efficacy; T1 self-efficacy significantly predicted higher T2 engagement. In the latter half of the semester, T2 self-efficacy and T2 engagement reciprocally predicted each other at T3, showing positive prospective associations in both directions over that interval. These patterns are consistent with reciprocal relationships between capability beliefs and participation, but do not establish causal effects or the internalization of obligations. The findings suggest pairing clear expectations with achievable tasks and progress-focused feedback throughout the semester.