This paper examines whether failure to control for family background, aptitude, high school quality, high school curriculum, and community characteristics leads to bias in estimates of the economic return resulting from postsecondary education. High school specific fixed effects were used to control for all observed and unobserved characteristics common to students from a given high school and an instrumental variables procedure was used to deal with measurement error in reported education. The estimation was conducted using panel data on several persons from each of a large number of high schools sampled in the National Longitudinal Survey of the High School Class of 1972. Controlling for family background and aptitude and achievement measures led to a substantial reduction in estimates of the rate of return to education. Measurement error was not responsible for the reduction. However, the use of fixed effects to control for high school and community characteristics had only a modest effect on estimates of the rate of return to education, and controlling for high school curriculum made little difference. It is noted that this is an important finding because few data sets permit researchers to control for these factors when estimating the effect of education on wages. Included is an appendix of means and standard deviations. Contains 12 references. (Author)