Objective: To compare the discriminating ability of online assessment tools in the module examinations during the COVID-19 pandemic. Study Design: Descriptive study. Place and Duration of Study: Department of Anatomy, CMH Multan Institute of Medical Sciences (CIMS), Multan, from 22 June to 09 July 2021. Methodology: In the academic year 2020, the rst module examinations of the 2nd-year MBBS anatomy course was conducted on-campus via multiple-choice questions, short answer questions, and viva-voce. Owing to the COVID-19 lockdown, the following two module examinations were held online. The e ectiveness of di erent assessment tooilns the three module examinations was compared by calculating the discrimination indices and the area under the curve (AUC), using the receiver operating characteristic (ROC) curve analysis. Results: SAQs showed the highest discrimination index (0.38) compared to MCQs and viva-voce in the on-campus module examinations but dropped to the lowest in the subsequent online modules (0.24 and 0.18). In contrast, the discriminating power of the viva-voce increased from marginally (0.23) to very good (0.47 and 0.49) as the mode of assessment shifted to online examinations. The ROC AUC also showed the same pattern. In the second and third module examinations, theviva-voce had signi cantly higher (p <0.05) AUC than MCQs alone and both the MCQs and SAQs, respectively. Conclusion: Viva-voce had a signi cantly higher discriminating index than MCQs and SAQs in online examinations. On-campus, SAQs had the highest discriminatory index. Using this statistical approach, the e ectiveness of di erent components of the online examinations can be monitored to improve the quality of online examinations amidst the COVID-19 crisis.