Estimating the Payoff to Attending a More Selective College: An Application of Selection on Observables and Unobservables

Abstract

, and colleagues at the Mellon foundation for helpful discussions. We alone are responsible for any errors in computation or interpretation that may remain despite their helpful advice. This paper makes use of the College and Beyond (C&B) database. The C&B database is a "restricted access database." Research who are interested in using this database may apply to the Andrew W. Mellon foundation for access. The views expressed herein are those of the authors and not necessarily those of the National Bureau of Economic Research.

The effect of university attended on graduates’ labour market prospects: A field study of Great Britain

“…Alternatively, in the US, two studies suggest that better-quality universities do not result in higher earnings (Dale and Krueger, 2002;2014). Exemptions exist for black and Hispanic students, and for students who come from less well-educated families (Dale and Krueger, 2014).…”

“…Studies emphasize that convincingly estimating the economic returns based on the university attended requires overcoming the biases arising from the fact that attendance at a betterquality university is likely to be correlated with unobserved characteristics that will themselves affect future earnings (Hoekstra, 2009;Broecke, 2012). Unobserved characteristics in relation to students' motivation, commitment, ability, skills, and personality characteristics, may be rewarded in the labour market (Brewer et al, 1999;Hoekstra, 2009;Dale and Krueger, 2002). That is, more able students studying in more reputable and selective universities may have the desirable characteristics that firms require.…”

“…That is, more able students studying in more reputable and selective universities may have the desirable characteristics that firms require. In addition, students' socioeconomic backgrounds, such as family characteristics, parental support and networks, could affect their labour market outcomes, and thus some students may have greater employability and salary capacity regardless of the university attended (Brewer et al, 1999;Hoekstra, 2009;Dale and Krueger, 2002). The literature concludes that, based on these difficult-to-control unobserved characteristics, firm conclusions on the relation between university attended and labour prospects cannot be drawn.…”

Accounting for institutional variation in expected returns to higher education

“…In both cases, test scores were significant predictors of student ROI and/or post-graduation earnings (Davies & Guppy, 1997;Fox, 1993;Long, 2010;Paglin & Rufolo, 1990;Thomas & Zhang, 2005;Zhang, 2005). Studies of institutional quality that have investigated the relationship between SAT scores (among other measures of selectivity and quality) and graduates' earnings find a positive relationship between graduate earnings and test scores (Black & Smith, 2004Dale & Krueger, 2002, 2014Long, 2008Long, , 2010. Because results vary by student characteristics such as race/ethnicity and student background, it is necessary to cleanse student test scores of other student attributes that also influence ROI (Dale & Kruger, 2014).…”

Demonstrating Use of Natural Language Processing to Compare College of Engineering Mission Statements

“…Prior literature suggests none (Dale & Krueger, 2002) to little (Pascarella & Terenzini, 2005) difference in pay-off or occupational status, between students attending a more selective college versus a public. However, based on the analysis in this research, and if mission statements are truly representative of the underlying principles and intents of an institute, there does exist a reasonable difference in the priorities to various topics or values.…”