As our desire to understand processes in social psychological phenomena has increased, so too has the sophistication of our tools. Nevertheless, some of the field’s requirements and considerations regarding mediation analysis seem outdated. Indeed, focusing on the significance of the X fi Y relationship before or after examining a mediator might be unnecessarily restrictive. We advocate a consideration of the significance of indirect effects and examination of effect sizes accompanying those effects in theory building and hypothesis testing.
Mediation analysis in Stata
paramed (Emsley et al., 2012) 公卫论坛
ldecomp (Buis, 2010)
medeff (Hicks and Tingley, 2011) 公卫考场
gformula (Daniel, De Stavola and Cousens, 2012)