Hi, I am doing a study for my masters and in that I performed linear regressions to check if all the assumptions were met. All the regressions came out to be significant. Even the mediators(I have 4 mediators) to dependent variable ones. But when I performed the mediation, the indirect effect was not significant but the total effect was. I don't understand why that is the case. Please help
As far as I can see, the last one of your indirect effects is statistically significant at the .05 level (p = .043).
It is entirely possible though for a total effect to be significant even when no indirect effect is significant. The total effect is the sum of the direct effect plus all indirect effects. If the direct effect is significant, this can drive the significance of the total effect, even when no indirect effect is significant. In addition, significance tests for the indirect effect that are based on its standard error (Sobel method) can be less powerful since the indirect effect is the product of multiple regression coefficients and tends to not be normally distributed.
Thank you so much for your insight! I also just want to know if we can say that the 4 factors play a mediating role or only the last one can be considered to be mediating?
According to your results table, only PSY_T would be considered a "statistically significant" mediator. Notice that this significance test, which seems to be based on the standard error and z statistic, may not be the most powerful test. Mediated effects can be more effectively tested with asymmetric bootstrap confidence intervals which account for the non-normal distribution of the indirect effect. See
MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate Behavioral Research, 39(1), 99-128.
- Badges
- Science topic