The relationship between postsecondary education and adult health behaviors

Abstract

Nearly 80% of American adults between the ages of 33-44 have at least some postsecondary education, which ranges from vocational training to a doctorate or professional degree. However, in education-health studies, postsecondary credentials are often grouped into a limited number of categories. This is an important omission as it obscures differentiations between the various types of postsecondary credentials. This study provides the first comprehensive analysis of disparities in health behaviors across detailed levels of postsecondary education. Data comes from Wave 5 of the 2018 National Longitudinal Study of Adolescent to Adult Health (Add Health). A covariance-weighting technique is used to produce behavioral index scores that identify the full spectrum of health behaviors influenced by postsecondary educational attainment. Estimates are initially produced in aggregate for the total sample population, with interaction models subsequently being used to test differences across gender and race/ethnicity population subgroups. The aggregate results indicate that adults with at least a bachelor's degree exhibit healthier lifestyles; however, no difference is observed among adults with lower-level postsecondary credentials, compared to high school graduates. Women experience steeper gradients at higher levels of postsecondary education, compared to men. Both White and Hispanic American adults exhibit comparable health lifestyles across levels of postsecondary education; however, Black Americans were found to experience no returns except at the doctorate or professional degree level. These findings have important implications particularly as adults in their thirties and forties continue to exhibit troubling health and mortality trends.

Keywords: Adults; Gender; Health behaviors; Postsecondary education; Race; United States.

Fig. 1

Predicted Health Behavior Index Scores. Estimates from Table 3 Model 5 are converted into predicted behavioral index scores, while holding the independent control variables constant at typical values, their average proportions or means (Fox & Andersen, 2006).

Fig. 2

Predicted Health Behavior Index Scores by Gender and Racial Identity. Estimates from Table 4 Model 2 for each group are converted into predicted behavioral index scores, while holding the independent control variables constant at typical values, their average proportions or means (Fox & Andersen, 2006).

Appendix Fig. 1

Predicted Probabilities of Individual Health Behaviors. Estimates from Appendix Table 4 are converted into predicted probabilities for each behavior, while holding the independent control variables constant at typical values, their average proportions or means (Fox & Andersen, 2006).