Re: Selecting Covariates at SPSS Mediation Analysis?

[Hans Chen]

Dear Lister

I was told that when conducting mediation analysis using SPSS, demographic variables (such as gender, age, and profession) should be used. Could you please advise which demographic variables I should use , the demographic variables which have direct impact on the dependent variable or demographic variables which have direct impact on other independent variables and mediating variables or all of them?

As we know, there are several indictors in Model Summary (R, R square and Adjusted R-square) from SPSS mediation analysis, could you advise which one should I choose?

These questions are from my friend.Thanks for your help!

--
Han

[Ryan]

Dear Han,

You ought to consider using Andrew Hayes' PROCESS custom dialog when examining mediation in SPSS:

http://www.afhayes.com/introduction-to-mediation-moderation-and-conditional-process-analysis.html

As for the questions you posed, I suggest you seek one-on-one consultation before proceeding with the analysis. Alternatively, you may consider learning the fundamentals of mediation analysis on your own through online sources (e.g., http://davidakenny.net/cm/mediate.htm) and perhaps through published sources (e.g., http://www.guilford.com/cgi-bin/cartscript.cgi?page=pr/hayes3.htm&dir=research/MSS_series). 

Best,

Ryan

On Fri, Apr 4, 2014 at 6:52 PM, Han Chen <[hidden email]> wrote:

Dear Lister

I was told that when conducting mediation analysis using SPSS, demographic variables (such as gender, age, and profession) should be used. Could you please advise which demographic variables I should use , the demographic variables which have direct impact on the dependent variable or demographic variables which have direct impact on other independent variables and mediating variables or all of them?

As we know, there are several indictors in Model Summary (R, R square and Adjusted R-square) from SPSS mediation analysis, could you advise which one should I choose?

These questions are from my friend.Thanks for your help!

--
Han