Table 2.
Variable | M | Linearized SE | Min. | Max. |
Drinking behavior | ||||
Drinks per day | 0.598 | 0.014 | 0 | 5.5 |
CAGE score | 0.426 | 0.012 | 0 | 4 |
Income and demographic characteristics | ||||
Household income, in U.S. $10,000 | 5.28 | 0.15 | 0 | 130 |
Works full-time | 0.586 | 0.008 | 0 | 1 |
Years of education | 12.5 | 0.089 | 0 | 17 |
Age | 55.5 | 0.042 | 51 | 61 |
White | 0.877 | 0.007 | 0 | 1 |
Female | 0.547 | 0.005 | 0 | 1 |
Married | 0.760 | 0.007 | 0 | 1 |
Health and cognitive status | ||||
Fair/poor health | 0.152 | 0.007 | 0 | 1 |
Cognition scorea | 13.473 | 0.137 | 0 | 40 |
CESD score | 0.708 | 0.024 | 0 | 8 |
Preferences and social supports | ||||
Risk aversion | 3.18 | 0.02 | 0 | 4 |
1 = least risk averse | ||||
4 = most risk averse | ||||
0 = missing | ||||
Financial planning horizon | 2.98 | 0.02 | 0 | 5 |
1 = next few months | ||||
5 = longer than 10 years | ||||
0 = missing | ||||
Rarity of religious service attendanceb | 1.93 | 0.02 | 0 | 5 |
1 = more than once/week | ||||
5 = not at all | ||||
0 = missing | ||||
Importance of religionc | 1.84 | 0.01 | 0 | 2 |
0 = not too important | ||||
2 = very important | ||||
Frequency of socializing with neighborsd | 1.98 | 0.02 | 0 | 5 |
0 = no friends in neighborhood | ||||
5 = daily or almost every day | ||||
0 = missing | ||||
Relatives live nearby | 0.208 | 0.006 | 0 | 1 |
Friends live nearby | 0.436 | 0.007 | 0 | 1 |
Life events | ||||
Retired | 0.496 | 0.007 | 0 | 1 |
Became disabled | 0.033 | 0.002 | 0 | 1 |
Got married | 0.055 | 0.004 | 0 | 1 |
Got divorced | 0.046 | 0.003 | 0 | 1 |
Death of a spouse | 0.124 | 0.004 | 0 | 1 |
Death of a sibling | 0.343 | 0.007 | 0 | 1 |
Hospitalized | 0.731 | 0.015 | 0 | 7 |
New diabetes diagnosis | 0.149 | 0.004 | 0 | 1 |
New cancer diagnosis | 0.119 | 0.004 | 0 | 1 |
New stroke | 0.063 | 0.003 | 0 | 1 |
New psychiatric diagnosis | 0.099 | 0.004 | 0 | 1 |
Observations, n | 6,787 |
Notes: Means calculated using survey weights, clustering, and strata. Min. = minimum; max. = maximum; CESD = Center for Epidemiological Studies—Depression scale.
Fewer observations because of missing values (n = 6,324);
fewer observations because of missing values (n = 6,260);
fewer observations because of missing values (n = 4,852);
fewer observations because of missing values (n = 6,336).