Figure 7. Marriage Health Gap: Results by Gender.

Note: Solid lines show estimated marriage health gaps β(a) from Equations (1), (2), and (3) respectively, estimated separately on the samples of males and females. The regression is fitted to the PSID. The dependent variable is an indicator variable that takes a value of one if the individual is healthy. Control variables include black and college dummies, income, dummies for 0–3, 4–12, and 13–18 year-old children at home, and year of birth dummies. The regressions also estimate α(a), α(a, η_gi), and φh_it−1 + α(a) respectively. The horizontal axis indicates age. In estimation, five-year age bins (20–24, through 60–64) are considered. The center point of the bin is represented in the figure. Weights are used in estimation. Confidence bands of ± two standard errors around point estimates are computed following the survey design of each database. Standard errors are clustered at the household level in the PSID.