Using online assessment to replace invigilated assessment in times of a natural disaster: Are some online assessment conditions better than others?

Abstract

As a result of the Canterbury earthquake on 4 September 2010, and associated aftershocks on 22 February 2011 and 13 June 2011, final examinations in the two first-year Economics papers at Canterbury University were cancelled at short notice in Semester 1, 2011. The final examination weightings were spread over the remaining assessments to obtain a final grade for students. This paper attempts to establish how different online assessment conditions affected final grade distributions when online assessments were substituted for an invigilated final examination. Pearson correlation coefficients and Spearman rank-order correlation coefficients were used to show that there was a greater correlation between online quizzes and invigilated assessments when those quizzes were only available for a restricted period of time compared with the whole semester. We found that online quizzes were more closely correlated with invigilated assessments when the first attempt at a quiz was recorded, as opposed to the higher of two attempts. We also found that using the first attempt leads to less grade disruption when compared with a ‘normal’ semester that includes a final examination. Finally, the actual effect on student grades when online quizzes are substituted for a final examination is discussed.