Modelling the Impact of Continuous Assessment Tests on Final Grade in a Course

Abstract

This study is an application of multi-equation system modelling to evaluate the effect of Continuous Assessment (CA) tests on academic performance. Understanding and consequently having good grades in statistics is generally presumed to be a difficult task for many students. Many factors contribute to academic performance, but the concern at the moment is on CA and other extraneous factors. The model was a 2-equation system. The first structural equation expresses final grade as a function of CA, and the second expresses CA as a function of the frequency of conducting CA and class attendance. Two service courses in statistics were used for the study. The results showed the frequency of CA and attendance to be highly significant (P<0.01) in explaining the variation in CA, while, as expected, CA significantly affects the final grades. Estimation of a simple regression of final grade on attendance revealed that attendance was significant (P<0.01) but not when final grade was regressed on both CA and attendance. This disagreement could be attributed to the multicollinearity of CA and attendance. Thus, multi-equation modelling enables the consideration of variables that could further help to improve on CA and, consequently, on the final grade without drawing a wrong inference.