Knowledge Monitoring Calibration: Individual Differences in Sensitivity and Specificity as Predictors of Academic Achievement

Abstract

Knowledge monitoring is an important metacognitive process, which can help students improve study habits and thereby increase academic performance. Which is more useful in predicting test performance: knowing what you know, or knowing what you do not know? Two distinct constructs of knowledge monitoring calibration, sensitivity and specificity, were used along with the more traditional Goodman-Kruskal gamma correlation to predict performance on tests in an undergraduate educational psychology course. The gamma correlation provides a measure of how good one is at judging both items one knows and items one does not. Measures of sensitivity and specificity distinguish between the two. Students in an undergraduate educational psychology course completed a 50-word knowledge monitoring assessment to measure sensitivity, specificity, and gamma. These measures were then correlated with test and final exam scores in the course. It was found that sensitivity, a measure of correctly identifying known items, was the most useful in predicting overall test scores as well as final exam scores. Specificity, on the other hand, had no significant impact on exam performance. Results suggest that sensitivity and specificity may be more meaningful measures of knowledge monitoring calibration when it comes to predicting academic achievement, as well as being better adapted for missing values in any one cell of the data. Further research is recommended to determine in what other situations the measures of sensitivity and specificity may be useful. Findings presented in this study can also be used to help guide attempts to improve student metacognition and strategies.