This paper uses quantile regression techniques to analyze heterogeneous patterns of returnto education across the conditional wage distribution in four transition countries. We correctfor sample selection bias using a procedure suggested by Buchinsky (2001), which is basedon a Newey (1991, 2009) power series expansion. We also examine the empiricalimplications of allowing for the endogeneity of schooling, using the control function approachproposed by Lee (2007). Using household data from Bulgaria, Russia, Kazakhstan andSerbia in 2003, we show that the return to education is heterogeneous across the earningsdistribution. It is also found that accounting for the endogeneity of schooling leads to a higherrate of return to education