Approaches to predictive studies: Possibilities and challenges

Abstract

This article investigates methodological issues associated with predictive studies related to selection and access to Higher Education. These issues are discussed in the context of tests designed and administered by the Alternative Admissions Research Project (AARP) at the University of Cape Town. The aim of the project is to design tests that broaden access to talented students who are not easily identified through the High School Senior Certificate examination system. The tests attempt to provide a mechanism for selection based upon whether the writers have the potential to succeed in the University environment. The article comments on the difficulties and limitations of various approaches to predictive studies. The article argues that a methodology grounded in the survival analysis approach holds particular promise for predictive studies and that it can provide a useful insight into the complex processes of student persistence, retention and attrition rates.

South African Journal of Higher Education Vol. 19(6) 2005: 1160-1176