Predicting Student Performance Using Past Data during COVID-19 Pandemic: A Validation Study

Abstract

COVID-19 has been a time of test in almost all sectors of life, and education is not an untouched sector. The online teaching-learning process emerged to be the only viable and effective mode of carrying on instructional processes in HEIs. However, it became difficult to ensure that learning happened the way it is expected in the offline mode of teaching-learning. The guiding underlying rationale behind the present study is the impact of teaching online mode on students in terms of learning and evaluation. Student testing and evaluation without the assurance of meaningful learning is an emergent problem owing to poor online connectivity and the digital divide denying democratic access to all learners. Through the present study, the authors want to propose that some indigenous alternative strategies should always be tested and tried in advance to evaluate students. This can solve the problems of unforeseen, unfavourable and unpredictable times, such as the COVID-19 pandemic, with regard to student evaluation. The study aims to suggest a viable student evaluation strategy using the statistical technique of prediction. The present paper discusses this effective and innovative strategy tried out using the available examination result database of the PG programme students from Mizoram University for three consecutive sessions spanning from 2015-2019 and the prediction of the final semester result of the 2018-2020 session. Pearson productmoment correlation, regression analysis, and Student t-test were used to analyze the data. The result of the analysis suggests that statistical prediction must be adopted as an alternative strategy for evaluating students.