Who Stays Who Leaves? Attrition Among First-Term Enlistees

Three variables observable at the time of enlistment are found to have a strong negative effect on attrition: positive expectations of further education, months in the Delayed Entry Program, and employment stability, which discriminate a wide range of attrition likelihood.

Abstract

Typically, 30 percent of an enlisting cohort leave before completing the first term, and attrition rates remain near this level even though over 90 percent of recent cohorts are high school graduates. This paper uses individual-level data to analyze attrition among high school graduates who enlisted in spring 1979. To control for selectivity bias, enlistment is analyzed jointly with attrition. Three variables observable at the time of enlistment are found to have a strong negative effect on attrition: positive expectations of further education, months in the Delayed Entry Program, and employment stability. As a set, these variables discriminate a wide range of attrition likelihood. No evidence of selectivity bias is found for this cohort; hence, results are applicable not only to enlistees but also to prospects. Policy implications are discussed.