Moderated Mediation
Message/Author
 Hana Johnson posted on Monday, January 20, 2014 - 10:58 pm
Hello,

I'm conducting a multi-level moderated mediation analysis. My IV, mediator, and DV are at level 1, and the moderator is at level 2 and it is stage 1 moderation. I also have 5 control variables. My syntax and variables are below. Does everything look correct?

Thank you so much!

id = grouping variable
c1, c2, c3, x2, x3 = control variables
x1 = IV
w= moderator
m=mediator
i1 = interaction

VARIABLE: NAMES ARE
id c1 c2 c3 x1 x2 x3 w i1 m y;
USEVARIABLES ARE
id c1 c2 c3 x1 x2 x3 w i1 m y;
BETWEEN IS w;
CLUSTER IS id;
ANALYSIS: TYPE IS TWOLEVEL RANDOM;
MODEL:
%WITHIN%
c1 c2 c3 x1 x2 x3 m y;
sa1 | m ON x1;
sb1 | y ON m;
sab1 | m ON i1;
%BETWEEN%
x1 w m y;
m ON x1(aa1);
m ON w;
m ON i1(ab1);
y ON m(bb);
y ON x1;
y ON w;
y ON i1;
y ON c1;
y ON c2;
y ON c3;
y ON x2;
y ON x3;
sa1 WITH x1 w m y;
sb1 WITH x1 w m y;
[sa1](a1w);
[sb1](b1w);
MODEL CONSTRAINT:
NEW(a1 b1 indb1 wmodval);
wmodval = .53583;
a1=aa1+a1w;
b1=bb+b1w;
indb1=(a1+ab1*wmodval)*b1;
OUTPUT: TECH1 TECH8 CINTERVAL;
 Linda K. Muthen posted on Tuesday, January 21, 2014 - 6:57 am
The best way to know if the input is correct is to run the analysis and see if you get the results that you expect. It is difficult to eyeball and input and make that judgement.
 Hana Johnson posted on Tuesday, January 21, 2014 - 8:14 am

WARNING: THE MODEL ESTIMATION HAS REACHED A SADDLE POINT OR A POINT WHERE THE OBSERVED AND THE EXPECTED INFORMATION MATRICES DO NOT MATCH. AN ADJUSTMENT TO THE ESTIMATION OF THE INFORMATION MATRIX HAS BEEN MADE. THE CONDITION NUMBER IS -0.268D-06.

THE PROBLEM MAY ALSO BE RESOLVED BY DECREASING THE VALUE OF THE MCONVERGENCE OR LOGCRITERION OPTIONS OR BY CHANGING THE STARTING VALUES OR BY USING THE MLF ESTIMATOR.

THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES MAY NOT BE TRUSTWORTHY FOR SOME PARAMETERS DUE TO A NON-POSITIVE DEFINITE FIRST-ORDER DERIVATIVE PRODUCT MATRIX. THIS MAY BE DUE TO THE STARTING VALUES BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE CONDITION NUMBER IS 0.230D-15. PROBLEM INVOLVING PARAMETER 42.

THE MODEL ESTIMATION TERMINATED NORMALLY

A MATRIX COULD NOT BE INVERTED DURING THE H1 MODEL ESTIMATION. THE ESTIMATED BETWEEN COVARIANCE MATRIX IS NOT POSITIVE DEFINITE AS IT SHOULD BE. COMPUTATION COULD NOT BE COMPLETED.

THE VARIANCE OF I3 APPROACHES 0. FIX THIS VARIANCE AND THE CORRESPONDING COVARIANCES TO 0, DECREASE THE MINIMUM VARIANCE, OR SPECIFY THE VARIABLE AS A WITHIN VARIABLE.

THE H1 MODEL ESTIMATION DID NOT CONVERGE. CHI-SQUARE TEST AND SAMPLE STATISTICS COULD NOT BE COMPUTED.
 Linda K. Muthen posted on Tuesday, January 21, 2014 - 10:24 am