|
|
 |
| Auxilary Vars. to make data MAR |
 
|
| Message/Author |
|
|
|
|
Graham (2003) and Enders(2006) suggest code to make data MAR:
USEVARIABLES are stage1-stage10 eversmk1 livwsmkr racewhte;
MISSING = all (-1234);
Analysis:
TYPE = MISSING;
Model:
i s| stage1@0 stage2@.5 stage3@1 stage4@1.5 stage5@2 stage6@2.5 stage8@3.5 stage9@4 stage10@4.5;
i on eversmk1;
[stage1-stage10@0];
[i s];
stage1-stage10;
i s;
!auxiliary variables;
livwsmkr racewhte;
!correlations among AVs;
livwsmkr with racewhte;
!correlate with IVs;
livwsmkr racewhte with eversmk1;
!correlate with DVs;
livwsmkr racewhte with stage1-stage10;
Output:
sampstat standardized;
This message appears for binary or continuous aux vars.
WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES.CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE LIVWSMKR.
Advise? |
|
|
|
|
| If the variable referred to in the error message is an auxiliary variable, I think you can ignore the message. |
|
|
| Back to top |
|
|
