The Impact of Emotional Intelligence on Academic Performance of College Students: The Moderating Effect of Self-Efficacy

Abstract

As a non-cognitive resource that supplements general cognitive ability in explaining academic outcomes in higher education, emotional intelligence is widely discussed, yet reported effect sizes vary widely across samples, instruments and performance criteria. We examine whether academic self-efficacy operates as a boundary condition on the association between emotional intelligence and academic performance among college students. A composite emotional intelligence index is constructed from the four dimensions of the Wong and Law framework using reliability-derived weights, academic self-efficacy is expressed as a mean-scaled latent score, and academic performance is normalized from institutional grade point average records. On a sample of 612 undergraduates drawn from three comprehensive universities, a hierarchical moderated regression model with a mean-centred product term is estimated, and the interaction is probed through simple slope decomposition and the Johnson-Neyman technique. A significant positive main effect on academic performance emerges from emotional intelligence, academic self-efficacy exhibits a stronger main effect, and the product term is significant and positive, indicating that the predictive value of emotional intelligence rises as efficacy beliefs strengthen. At low levels of self-efficacy the conditional effect of emotional intelligence is weak and statistically indistinguishable from zero, while it reaches roughly nine times that magnitude at high levels. The Johnson-Neyman transition point falls at a standardized self-efficacy value of about negative 0.67, above which the conditional effect attains significance, covering approximately three quarters of the sample. The interaction coefficient is found to be robust when considering alternative covariate sets and under bootstrap resampling. These results indicate that emotional competence converts into measurable academic gain mainly when students hold sufficiently strong beliefs about their own capability, and this implies that emotional skills training should be paired with efficacy-building instructional practice rather than delivered in isolation.