Calibration Accuracy and Self-Efficacy in Health-Eelated Fitness and Fitness Knowledge

Background/Purpose: Calibration accuracy is defined as the extent to which student judgment reflects their actual performance. Self-efficacy refers to student belief about their capability in performing a specific task. Both calibration accuracy and self-efficacy are important for metacognitive processes and self-regulation (Flavell, 1979). This research examined middle-school students’ calibration accuracy and self-efficacy in a physical fitness test (study I) and a cognitive fitness knowledge test (study II).

Method: In study I, the participants were 138 seventh graders ranging from 11 to 14 years old. The participants were asked to complete a self-efficacy scale, predict their scores on the upcoming PACER test, and then perform the actual PACER test. In study II, the participants (N=90) were also seventh graders from the same school. Because the 15 item fitness knowledge test was new, students were asked to review the knowledge test then make a prediction of their performance, and complete a self-efficacy scale. We conducted descriptive statistical analyses on predicted performance, actual performance, and self-efficacy. Paired sample t-tests were then conducted to detect the difference between the predicted and actual performance in PACER and knowledge tests. Lastly, Pearson product-moment correlation analyses were conducted to examine the association between the variables.

Analysis/Results: Overall, students’ predicted performance (M=40.60, SD=21.66) was significantly higher than their actual performance in PACER (M=37.70, SD=18.95; t=3.39, p<.05). Students’ predicted knowledge test score (M=11.33, SD=2.12) was not significantly different from their actual performance (M=11.28, SD=2.48; t=-.14, p>.05). Correlation analyses showed that students’ predicted performance was significantly correlated with their actual performance in PACER (r=.85, p<.01), but their predicted knowledge score did not correlate with their actual score (r=.05, p>.05). Students’ self-efficacy significantly correlated with their performance in PACER (r=.78, p<.01), predicted PACER performance (r=.77), and predicted knowledge score (r=.45). However, it did not correlate with the actual knowledge test score (r=.15, p>.05).

Conclusions: Consistent with findings in other domains (Hacker et al., 1998), the results showed that students had relatively high calibration accuracy for PACER (r=.85), even though they overestimated their performance. Overall, students did not overestimate their knowledge test scores. Yet they reported very low calibration accuracy (r=.05) suggesting that students had inaccurate judgment about their fitness knowledge test scores. Compared to the high calibration in PACER, the low calibration in the knowledge test and its insignificant correlation with self-efficacy could result from students’ lack of experiences in taking fitness knowledge tests in physical education.