Skip to main content
. Author manuscript; available in PMC: 2016 Feb 11.
Published in final edited form as: Obesity (Silver Spring). 2015 Mar 25;23(5):1105–1111. doi: 10.1002/oby.21052

Table 4.

Model testing mediation of the association between obesity and perceived weight discrimination by psychological wellbeing (see Figure 2)

Coeff. SE p* Bootstrap 95% CI Effect ratio
Total effect (path c) 0.112 0.006 <.001 - -
Direct effect (path c’) 0.106 0.006 <.001 - -
Indirect effect (via mediators) 0.0063 0.001 <.001 [0.004; 0.009] 0.056
  Indirect effect (via quality of life) 0.0059 0.001 <.001 [0.004; 0.009] 0.053
  Indirect effect (via life satisfaction) −0.0009 0.0005 .072 [−0.002; −0.0001] −0.008
  Indirect effect (via depressive symptoms) 0.0013 0.0007 .067 [0.0001; 0.003] 0.012

Model uses z-scores for all psychological wellbeing variables.

Model is adjusted for age, sex, and wealth.

Coeff. = coefficient; SE = standard error; CI = confidence interval.

*

p values shown for indirect effects are derived from the Sobel test for consistency with total and direct effects, however bootstrap 95% confidence intervals provide a more robust indication of significant mediation (see Method for more details).