Predicting success in higher education:: predictive validity, attainment at school level and its relationship to degree class

Abstract

This study examined the predictive validity of certain variables (Prior Achievement, Home Background, aspirational level) against a criterion variable i.e. Degree Class. The measures came from the ALIS (A-level Information System) in England. The main source of data were 1167 students who agreed to complete questionnaire sent as part of the ALIS project. Simple correlation, multiple and stepwise and regression analysis were performed. The following conclusion were drawn from the findings:1. All the predictor variables except home background were significantly predictive of academic success in terms of scholastic performance, the predictive power was very low for average O level (AVO), socio economic status of Head of Household (HOH), moderate for the likelihood of staying in Education (LSE) and moderately high for A-level. 11. Total A-level was the best predictor of success at the Degree Class. The low predictability of the AVO may be attributed to a weakness to the power of the preparatory programme. The low values for HOH may be because decision about future study had already been made perhaps on the basis of home background. The Total A-level variable correlated 0.37 with degree class but it seems likely that this figure would be higher if it were possible to look at individual courses at particular universities.