Skip to main content
. 2014 Jul 1;70(6):948–956. doi: 10.1093/geronb/gbu080

Table 2.

OLS Regression Coefficients for Predictors of Depression Scorea (n = 325)

Predictor Model A Model B Model C Model D
Unstandardized B SE Unstandardized B SE Unstandardized B SE Unstandardized B SE Standardized β
Intercept 57.230*** 4.019 61.640*** 6.305 57.518*** 6.555 67.588*** 6.952
Age (years) 0.026 0.063 −0.082 0.063 −0.057 0.064 −0.058 0.064 −0.050
Sex = male −4.866*** 1.111 −3.799*** 1.042 −3.963*** 1.039 −4.450*** 1.146 −0.436
BMI (kg/m2) −0.390* 0.177 −0.362* 0.176 −0.400* 0.176 −0.116
Disability scoreb 1.263*** 0.178 1.210*** 0.179 1.139*** 0.182 0.328
Subsistence involvement scorec −1.028* 0.473 −1.066* 0.471 −1.126* 0.469 −0.129
Social conflict scored 1.129* 0.524 1.288* 0.525 0.126
Spanish fluency (fluent vs. none) 3.822 2.128 0.374
Distance to town (km) −0.035 0.025 −0.072
Adjusted R 2 0.050 0.188 0.197 0.205

Notes. BMI = body mass index; OLS = ordinary least squares.

aRaw depression score (summed 16 items) was converted to % maximum possible score (summed items/64×100). Unstandardized betas represent change in % maximum possible depression score.

bHigher score indicates greater degree of physical limitations.

cHigher score indicates greater involvement in daily subsistence tasks.

dHigher score indicates greater reported unresolved conflict.

*p < .05. ***p < .001.