What Are the Most Important Predictors of the Earnings of College Graduates? Data From the College Scorecard

• In Event: 16.061 - Roundtable Session 4
  In Roundtable Session: 16.061-12 - Institutional and International Contexts

Abstract

This study investigated the most important predictors of the 6-year post-graduation income of college graduates who used student aid from the U.S. Department of Education during their time at college. The latest data publicized by College Scorecard was used. Specifically, 1,429 cohorts of graduates from three years (2001, 2003, and 2005) were included in the data analysis. Three algorithms, including forward and backward stepwise regression, and the genetic algorithm, were applied to the predictor selection from 31 relevant factors. This study found that the genetic algorithm outperformed both forward and backward stepwise regressions. The best selected predictor subset included 18 factors, based on which the heterogeneity of labor market returns to higher education is discussed.

Authors

• Ewan Wright, The University of Hong Kong

• Qiang Hao, Western Washington University

• Joshua Michael Rosenberg, The University of Tennessee - Knoxville