Test Anxiety and Self-Efficacy in a Computer-Based Test Environment

Many university courses are moving from traditional paper-based assessments to computer-based tests with automated grading. This change could lead to increased student test anxiety but because the ease of grading facilitates frequent smaller-stakes assessments, it may reduce test anxiety particularly over the course of a semester as students become familiar with the test platform. This study examines the roles of test anxiety, self-efficacy, and academic performance, leveraging an in-person, supervised, computer-based testing approach. We measure each of these variables for 173 CS1 students over five unit quizzes and one final exam. We find no significant correlation between test anxiety and performance for the men or the women, while self-efficacy correlates moderately with performance for the men. Despite higher anxiety and lower self-efficacy, the women achieved comparable academic outcomes to the men. These results challenge traditional views on the impact of test anxiety on academic performance, highlighting the need for deeper investigation into these phenomena.