The positive association between education and health is well establis
hed, bur explanations for this association are not. Our explanations f
all into three categories: (1) work and economic conditions, (2) socia
l-psychological resources, and (3) health lifestyle. We replicate anal
yses with two samples, cross-sectionally and over time, using two heal
th measures (self-reported health and physical functioning). The first
data set comes from a national probability sample of U.S. households
in which respondents were interviewed by telephone in 1990 (2,031 resp
ondents, ages 18 to 90). The second data set comes from a national pro
bability sample of U.S. households in which respondents ages 20 to 64
were interviewed by telephone first in 1979 (3,025 respondents), and t
hen again in 1980 (2,436 respondents). Results demonstrate a positive
association between education and health and help explain why the asso
ciation exists. (1) Compared to the poorly educated well educated resp
ondents are less likely to be unemployed, are more likely to work full
-time, to have fulfilling, subjectively rewarding jobs, high incomes,
and low economic hardship. Full-time work, fulfilling work, high incom
e, and low economic hardship in turn significantly improve health in a
ll analyses. (2) The well educated report a greater sense of control o
ver their lives and their health, and they have higher levels of socia
l support. The sense of control, and to a lesser extent support, are a
ssociated with good health. (3) The well educated are less likely to s
moke, are more likely to exercise, to get health check-ups, and to dri
nk moderately, all of which, except check-ups, are associated with goo
d health. We conclude that high educational attainment improves health
directly, and it improves health indirectly through work and economic
conditions, social-psychological resources, and health lifestyle.