Regression and matching estimates of the effects of elite college attendance on educational and career achievement

This paper adopts a potential outcome approach to identify and estimate the average treatment effect of attending an elite college on educational and career achievement. A central purpose is to compare the estimates yielded by regression and matching methods of adjusting for the endogeneity of elite college attendance. The analysis follows a high school graduation and college entry cohort across four decades of labor force participation, and estimates elite college effects on educational attainment, occupational socioeconomic status at early-, mid-, and late-career, and wages at mid- and late-career. The findings suggest that attending an elite college yields an advantage with respect to educational achievement and occupational status; results for wages are mixed. One prominent pattern is that the returns to attending an elite college for those who did attend are small by comparison to those that would have been achieved by otherwise equivalent students who attended non-elite institutions.