2-1-1-2 modeling PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Li Lu posted on Wednesday, April 24, 2013 - 4:39 am
We are running a multilevel two step mediator model, and I kind of integrated the codes from multilevel part and moderated mediation path analysis, but it seems not working. Can someone give me some suggestion? Thanks!
Here is the output:
title: 2-1-1-2 mediation model
data: FILE IS "xxx";
variable:
NAMES = team simr1 simr2 simr3 manicode mbl1 mbl2 mbl3 exin1 exin2 exin3 exin4
simrsum gsimrsum simrfed exptsum gexptsum expt1 expt2 expt3 expt4 expt5 exptfed meanInno
dpstotscre combscore;
USEVAR = Team simrsum exptsum mbl1 mbl2 mbl3 exin1 exin2 exin3 exin4 meanInno dpstotscre;
MISSING are .;
BETWEEN are dpstotscre;
CLUSTER IS Team;

Analysis:
TYPE IS TWOLEVEL RANDOM;
processors = 2;
MODEL: !model specification follows
%WITHIN%
exinW by exin1 exin2 exin3 exin4;
MblW by mbl1 mbl2 mbl3;
exinW on simrsum mblW;

%BETWEEN%
MblB by mbl1 mbl2 mbl3;
exinB BY exin1 exin2 exin3 exin4;
exinB on simrsum(gamma1) mblB(gamma2);
dpstotscre ON exinB(b);

MODEL CONSTRAINT:
LOOP(mod,-2,2,0.1)
NEW(indb);
indb=b*(gamma1+gamma2*mod);
OUTPUT: TECH1 TECH8 CINTERVAL;

*** ERROR in MODEL command
Unknown variable(s) in an ON statement: (GAMMA1)
 Linda K. Muthen posted on Wednesday, April 24, 2013 - 12:24 pm
I think you need a space between the label and the variable name. Also you should have only one label per line.

exinB on simrsum(gamma1) mblB(gamma2);

should be

exinB on simrsum (gamma1)
mblB (gamma2);
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: