Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We analyse educational choices and earnings of individuals at two different levels in the Portuguese educational system. At each exit we consider two decisions: the decision to continue studying and the employment decision. We find empirical support for the existence of selectivity bias as the errors of the earnings functions are in general correlated with the disturbances of the decision functions. The obtained rates of return to education vary between 2.4 and 9.4 percent, depending on whether or not selectivity and quantity effects are excluded from the computations. Note that selectivity represents three quarters of this difference. Copyright Springer-Verlag Berlin Heidelberg 2003