College quality and early adult outcomes

Abstract

This paper estimates the effects of various college qualities on several early adult outcomes, using panel data from the National Education Longitudinal Study. I compare the results using ordinary least squares with three alternative methods of estimation, including instrumental variables, and the methods used by Dale and Krueger [(2002). Estimating the payoff to attending a more selective college: An application of selection on observables and unobservables. Quarterly Journal of Economics, 117(4), 1491–1528.] and Black and Smith [(2004). How robust is the evidence on the effects of college quality? Evidence from matching. Journal of Econometrics, 121, 99–124.]. I find that college quality does have positive significant effects on most outcomes studied using OLS. While there is some evidence of positive selection bias in the OLS results, the alternative methods rarely produce findings that are significantly different from the OLS estimates. Furthermore, alternative methods have their own limitations, which are discussed. Across methods of estimation, there is solid evidence of positive effects of college quality on college graduation and household income, and weaker evidence of effects on hourly wages.