Exploring Student Confidence and Calibration Accuracy in College Algebra: A Mixed-Methods Explanatory Sequential Study

Calibration, or students’ ability to predict and postdict their performance on an academic task, has important implications for academic performance. It can influence students' decision making processes concerning learning activities, as well as their levels of motivation and self-regulation (Kolovelonis, 2023). The literature on calibration is clear, often indicating that students are not great at these judgements (Lin et. al, 2001). While this phenomenon is well documented, it is not an easy one to explain. Therefore, a two-phase explanatory-sequential mixed method design was used to examine undergraduate students’ calibration judgements on a college algebra exam. To further contribute to the literature, a key goal of the present study was to conduct one on one video recorded interviews to gain additional insights into students’ rationale for their calibration judgements. Quantitative findings indicated that higher achieving students were more accurate in their predictions; whereas lower achieving students were less accurate and tended to overestimate their performance. There were no significant differences in postdiction accuracy. These are both robust patterns of results (e.g., Hacker et al. 2000; Bol & Hacker, 2012). Qualitative results from follow up interviews revealed that students across achievement levels commonly experienced math related anxiety and stress particularly in testing situations. Students often based their confidence on emotions such as anxiety, doubt, or optimism rather than clear evidence of performance. These results underscore the complex interplay between emotion and self-awareness in students’ confidence judgments and calibration accuracy.