A study was conducted to determine whether academic performance could be predicted on the bases of the constructs need for cognition (NFC) and academic self-efficacy. Two hypotheses were generated: Positive correlations will be found between academic self-efficacy, NFC, and grade point average (GPA); and efficacy and NFC will serve as significant predictors of GPA. The path mediation technique recommended by Baron & Kenny (1986) for testing mediated relationships was also performed in order to assess the causal direction of the NFC and academic self-efficacy variables. Participants were 138 undergraduate students. The first hypothesis was generally supported in that significant correlations were found between NFC, efficacy beliefs, and GPA. In support of the second hypothesis, path analysis revealed that NFC and academic self-efficacy were significant predictors of GPA. Furthermore, the NFC-GPA relationship was shown to be mediated by efficacy beliefs.