*3.2. Tests for the Mediation Model and Estimated Coefficient Paths*

The first step of the SEM estimated the factor loadings through CFA (Table 2). The results of the reliability test are also presented, which indicated an acceptable range of reliability. After adjusting for age and gender, the multiple mediator model was used to estimate the indirect and direct effects, and the estimated path coefficients are illustrated in Figure 2. We found that an indirect effect at a value of −0.06 reached statistical significance (Sobel test: Z = −4.05; *p* < 0.05), and this was based on the product terms of the path from perceived social support to risk perception (β = −0.13, *p* < 0.001) and the path from risk perception to active coping with COVID-19 (β = 0.49, *p* < 0.001). On the other hand, the mediating effect of confidence on the path between perceived social support and active coping with COVID-19 was not significant (Sobel test: Z = 0.99; *p* = 0.32). Moreover, the direct effect from perceived social support to active coping with COVID-19 was not statistically significant. The significance of the path analysis did not change after adjusting for age and gender.

These results confirmed the mediating effect of risk perception on the association between perceived social support and active coping with COVID-19. Based on the model fit index, the hypothesized model had an adequate model fit index for RMESA (0.068), GFI (0.927), AGFI (0.902), and SRMR (0.069), indicating the good fit of our hypothesized mediation model.

**Figure 2.** Final model of mediating effect indicating the estimated coefficients of the paths.

#### **4. Discussion**
