CORRECTING FOR SAMPLING BIAS IN EDUCATION PRODUCTION FUNCTIONS

Sample selectivity bias arises in the estimation of education production functions when the process of selecting the sample is not independent of the value of the dependent variable. In this study, the dependent variable was a student's retention of microeconomics knowledge after a period of 6 weeks following the final examination. Potential sampling bias arises from a student's decision to participate or not to participate in the research. Results obtained from the Heckman selection model indicated minor sampling bias. Student ability and the male dominance in economics performance were overestimated in the OLS model. Student effort was underestimated. Copyright (c) 2007 The Author. Journal compilation (c) 2007 Economic Society of South Africa.