Skip to main content
. 2022 Dec 7;13:1054566. doi: 10.3389/fpsyg.2022.1054566

Table 3.

Regression models of mediating effect.

(1) Wellbeing (2) ReliActFreq (3) Wellbeing
ReligionType: Institutional 0.131* 1.023*** 0.102
(0.053) (0.107) (0.054)
ReliActFreq 0.028*
(0.013)
RegionType: Rural 0.175** 0.027 0.173**
(0.061) (0.125) (0.061)
Age 0.008*** 0.013** 0.008***
(0.002) (0.004) (0.002)
Gender: Female −0.094 0.099 −0.096
(0.050) (0.102) (0.050)
Marriage: Married −0.106 0.155 −0.110*
(0.054) (0.110) (0.054)
Job: FullTime 0.160** −0.062 0.161**
(0.056) (0.113) (0.056)
EduYear 0.009 0.000 0.009
(0.008) (0.017) (0.008)
SelfClass 0.125*** 0.074* 0.123***
(0.017) (0.034) (0.017)
(Intercept) −0.012 2.891*** −0.008
(0.088) (0.226) (0.089)
Marginal R2 0.078 0.079 0.080
Conditional R2 0.096 0.124 0.100
AIC 4,106.591 6,179.907 4,110.605
BIC 4,164.874 6,238.190 4,174.186
Num. obs. 1,478 1,478 1,478
Num. groups: Region 6 6 6
Var: Region (Intercept) 0.018 0.190 0.019
Var: Residual 0.897 3.671 0.895

1. Baselines for the qualitative variables are as follows: Diffused(ReligionType), Urban(RegionType), NotMarried(Marriage), NotFull (Job) and Male(Gender).

2. Unstandardized regression coefficients are displayed, with standard errors within parentheses. *p < 0.05. **p < 0.01. ***p < 0.001.