Extended moderated mediation (Path analysis)

[kkim]

Hello all,

I'm conducting a psychological study on gender and cognitive activation of networks. I have a question about analysing a data set collected for the study.

Particularly, I would like to see whether there is gender difference in the relationships between several variables (please see the model 1 attached) – i.e., whether there is gender difference in the path, “Threat” -> “Density” -> “Accuracy” -> “Negotiation.” The path represents an indirect effect, and in the model 1, the path is highlighted in red.

I would like to run a path-analysis (without specifying latent variables) for this Model 1 on Stata, and more importantly I want to examine “whether the indirect effect (Threat -> Density -> Accuracy -> Negotiation Likelihood)” differs for men and for women.

I think basically, this is an extension of moderated mediation models, but I have not run it on Stata.
Any insight or information on running this model (i.e., the path-model, and the indirect effects) would be appreciated. Thank you in advance!

Best,
Karen

[Dave Airey]

Here is a nice start at the ATS UCLA site: http://www.ats.ucla.edu/stat/stata/faq/modmed.htm.

[Dave Airey]

Here's another page at their site with a categorical moderator: http://www.ats.ucla.edu/stat/stata/faq/modmedcat2.htm.