Application of Principalcomponent analysis to Secondary schools examination results in kenya

Abstract

In this project I have adopted Principal Component Analysis (peA) as data driven procedure, to study the performance of secondary school students through examinations results over ten years. The study has clearly come up with two principal components. The first has shown that, the performance is due to a combined effect of all the subjects, while the second has grouped subjects into three categories vis Languages, Business subjects and Sciences. The subjects in the same category are highly correlated to one another and languages have the highest component loadings meaning that they are highly correlated to both components, being the highest contributing factor to performance. The categories are uncorrelated to each other, hence implying a differentiated effect to perfomance by each category.