two mediators in one equation

[MKA-4]

Hi listers;
I now try make a mediaton test with multiple regression analysis. When I make individual mediation tests these partial mediation effects revealed: Emotional intelligence->job satisfaction->organizational commitment->organizational citizenship behavior. That is job sat. is partial mediator between EI and org. commit.       Org commit. is a partial mediator between job sat. and org. cit.    I wonder there is a way in SPSS to test the mediation effects of two variables (job satisf. and org. commitm.) between Em. Intell. and org. citizenship in one analysis rather than separate individual mediation tests.
If it is possible to test the mediation effects of these two variables (job sat and org. commit) in one analysis, In which order I should enter these variables and their interaction variables (cross products) into the equation:

Step 1                        Dependent variable: Org. Citizenship  
EInt.

Step 2
JSat.
OCom.

Step 3
EInt X Jsat.
EInt X Ocom.
Com X JSat

Step 4
EInt X Jsat. X OCom


Mehmet KARAKUS

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Sara House]

Mehmet -
 
I'm not sure if there is a way with regression, but you could definitely test multiple mediators using structural equation modeling.  You would just create a path model like the one you described:
 
Emotional intelligence->job satisfaction->organizational commitment->organizational citizenship behavior
 
and then test that with SEM.  Unfortunately, I have not done SEM with Amos (I'm a LISREL girl myself), so I don't know what the syntax would look like.  But I'm sure someone out there knows.  If you want to talk about how to do this analysis in LISREL, feel free to email me off list.
 
Sara
 
Sara M. House, M.A.
Adjunct Faculty
Loyola University Chicago, Psychology Department
Email: [hidden email]
Teaching: Research Methods, Psychology & Law

>>> MKA <[hidden email]> 9/3/2008 2:24 PM >>>
Hi listers;
I now try make a mediaton test with multiple regression analysis. When I make individual mediation tests these partial mediation effects revealed: Emotional intelligence->job satisfaction->organizational commitment->organizational citizenship behavior. That is job sat. is partial mediator between EI and org. commit.       Org commit. is a partial mediator between job sat. and org. cit.    I wonder there is a way in SPSS to test the mediation effects of two variables (job satisf. and org. commitm.) between Em. Intell. and org. citizenship in one analysis rather than separate individual mediation tests.
If it is possible to test the mediation effects of these two variables (job sat and org. commit) in one analysis, In which order I should enter these variables and their interaction variables (cross products) into the equation:

Step 1                        Dependent variable: Org. Citizenship  
EInt.

Step 2
JSat.
OCom.

Step 3
EInt X Jsat.
EInt X Ocom.
Com X JSat

Step 4
EInt X Jsat. X OCom


Mehmet KARAKUS

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD