O. Ashenfelter et Dj. Zimmerman, ESTIMATES OF THE RETURNS TO SCHOOLING FROM SIBLING DATA - FATHERS, SONS, AND BROTHERS, Review of economics and statistics, 79(1), 1997, pp. 1-9
Data on brothers and on fathers and sons from the National Longitudina
l Survey are used to consider the impact of omitted variables and meas
urement errors on the economic returns to schooling. The analysis sugg
ests that the upward bias in estimated returns due to omitted variable
s is likely offset by an equal downward bias resulting from measuremen
t errors in reported schooling. Controlling for both of these potentia
l sources of bias yields results comparable to conventional regression
estimates of the economic return to schooling.