Estimating years of education using the current population survey after 2014

摘要

In 2015, the Current Population Survey (CPS) eliminated three questions related to educational attainment. These now-eliminated questions continue to be used by the National Bureau of Economic Research (NBER) to calculate their years of education variable in their Merged Outgoing Rotation Groups (MORG) extracts. We demonstrate that the NBER imputation method after 2014 results in missing values for imputed education for 27.5% of the observations and average years of education of about 1.1 fewer years. We provide coding that can be used with the current CPS questions to allow comparison of years of education over time. (c) 2020 Elsevier B.V. All rights reserved.