Behavioural engagement, academic dishonesty, and performance gaps: Comparing online and paper–pencil based tests in an online learning context

Abstract This study examines the growing issue of academic dishonesty in online assessments, a challenge intensified by the expansion of digital learning platforms. Specifically, it investigates the relationship between students' online engagement and their performance in online versus traditional paper-and-pencil tests. Employing a cross-sectional research design, data were collected from 1,022 students across various faculties at a state university. Regression analysis revealed that behavioural engagement significantly predicted performance in paper–pencil tests but not in online assessments. Moreover, students scored higher in online tests, yet the weak correlation between scores across both formats raises concerns about the validity of online assessments in accurately reflecting student learning. These findings underscore the necessity of enhancing online assessment integrity and leveraging engagement metrics to prevent academic dishonesty. The study suggests that educators, policymakers, and administrators should consider student engagement as a key factor in fostering a culture of academic honesty while supporting students' academic performance.