Greetings, I am working on survey data for my dissertation and I completely using SAS on demand studio for conducting my analyses. My main aim is to evaluate a mediational relationships. After reading different sources I found out that the best approach is to examine the indirect effect of the independent variable on the dependent variable via the mediator.   I read about the Proc Causalmed statement and tried to work on it. However, the statement does not look to be working with me. I cannot do anything with it and therefore cannot perform the procedure.   Another question is about the application of survey weights, strata, and clusters for the analysis. I am not sure if this is applicable or not?   Please I need your help to conduct the analysis and if there is another approach for a mediation analysis besides the Baron and Kenny approach (three regression models) that would be very helpful.   Thanks, ... View more