Mplus vs. SPSS PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Tanya Vachar posted on Monday, March 28, 2011 - 12:59 pm
I am running simple linear regressions in both Mplus and SPSS to cross-validate my code and results. In general, the regressions are replicating in the two programs, except for just one of the models, and I think I've been able to isolate the variable that seems "problematic". Specifically, there is a signficant positive effect of A -> B in SPSS (p = 0.00) but the effect becomes negative and non-significant in Mplus. However, the dataset used in both programs is identical, and the reported means in both programs are identical too.

Any ides for troubleshooting? Or, is there any particular reason why there could be this discrepancy between just the one particular variable?

Here is my code:
TITLE: read in raw data;
DATA: file='c:\users\desktop\dyad.dat';
VARIABLE:
names = var1 var2;

usevar = var1 var2;

MODEL:
var2 ON var1;

OUTPUT: standardized tech1 tech4 modindices(0);
 Linda K. Muthen posted on Monday, March 28, 2011 - 2:04 pm
I would need to see the SPSS output in a format I can read, the Mplus output, and your license number at support@statmodel.com to say why this difference occurs.
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: