Table 2.
Estimated regression coefficients from a simple mediation analysis with two independent variables.
| Consequent |
||||||||
|---|---|---|---|---|---|---|---|---|
| Stress |
Mental health |
|||||||
| B | SE | p | B | SE | p | |||
| Stigma | a1 | 0.075 | 0.023 | 0.002 | c1 | − 0.306 | 0.042 | < 0.001 |
| Stress | – | – | – | b | − 0.810 | 0.130 | < 0.001 | |
| Hardiness | a2 | − 0.310 | 0.051 | < 0.001 | c2 | 0.487 | 0.100 | < 0.001 |
| Constant | 25.751 | 1.455 | < 0.001 | 52.649 | 4.202 | < 0.001 | ||
| R2 = 0.205 | R2 = 0.510 | |||||||
| F(2, 184) = 23.701, p < 0.001 | F(3, 183) = 63.476, p < 0.001 | |||||||
B = unstandardized coefficient; SE = standard error.
a1, a2, b, c1, and c2: unstandardized regression coefficients; e1 and e2: unexplained parts of stress and mental health, respectively.
Two regression models: stress = constrant1 + a1·stigma + a2·hardiness + e1; mental health = constrant2 + c1·stigma + c2·hardiness + b·stress + e2.