Differential Prediction of Preparatory and Performance Self-Efficacy Judgments

Abstract

This study examined the effects of self-efficacy in preparatory and performance contexts. We conducted a longitudinal study of 63 students' exam preparation and performance. Three waves of data were collected in conjunction with course exams. Within each wave, self-efficacy was assessed at 2 time points, 1 week before the exam (preparatory self-efficacy) and within the 24 hr preceding the exam (performance self-efficacy). We expected preparatory and performance self-efficacy to predict exam preparation and performance, respectively. Multilevel analysis revealed that the effects of preparatory self-efficacy depended on the level of analysis, but performance self-efficacy had positive effects on exam performance at the between- and within-person levels of analysis. Our results suggest that timing of self-efficacy assessment matters as the effects of self-efficacy depend on the context.

Notes

¹One participant's data were excluded from the analysis of study hours for two exams because they reported a value of 100 hr, which was thought to be an unreasonable value for a 1-week period, and it was +3 SD above the mean for study hours.

²Because the results obtained in the HLM analysis for self-efficacy were consistent with those obtained with expected grade, we show only the results related to self-efficacy in Tables 2 and 3. There were only three major differences in the results obtained with expected grade compared to the results obtained with self-efficacy: First, self-efficacy did not change over the course of the semester, but preparatory expected grade did reliably decrease. Second, H3, which stated that goal grade would be positively related to exam preparation, was partly supported with the measure of self-efficacy in that within-person preparatory self-efficacy was positively related to metacognitive activity. This hypothesis was not supported for preparatory expected grade, which was not significantly related to either metacognitive activity or study hours at the between- or within-person levels. Third, H5, which stated that exam preparation would be positively related to performance self-efficacy and exam performance, was not supported with measures of self-efficacy (neither study hours nor metacognitive activity were related to performance self-efficacy). This hypothesis was somewhat supported with expected grade. That is, metacognitive activity was significantly related to performance expected grade. In sum, our major findings—that preparatory self-efficacy influences goals, which influence exam preparation, and performance self-efficacy influences exam performance—remain unchanged with both measures. Results with expected grade are available, upon request, from the second author.