Std. error p-values vs. difference test PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Naomi Friedman posted on Friday, October 16, 2009 - 11:51 am
I have run across several cases where the p-values given in the output do not always match what I obtain when I use DIFFTEST (in the case of WLSMV), or with the corrected chi-square difference test (in the case of MLR). In many cases this happens with twin models, but not always (sometimes with growth or SEM). I'm wondering why this would happen and which p-values I should trust.
 Bengt O. Muthen posted on Saturday, October 17, 2009 - 12:29 pm
I don't see why DIFFTEST or chi-square difference tests should give the same p values as any given H0 model. The H0 model is tested against a completely unrestricted H1 model, whereas difference testing typically uses a more specific H1 model - a model that is just a little bit more general than H0. Maybe I am misunderstanding your question.
 Naomi Friedman posted on Monday, October 19, 2009 - 12:37 pm
Hi. Sorry I was unclear. The p-values I am referring to are for a particular parameter, not the overall model fit. So, in my specific case, I am running an ACE model, and the output z-test for the C loading suggests it is highly significant, but setting this loading to zero results in a non-significant DIFFTEST result.
 Bengt O. Muthen posted on Tuesday, October 20, 2009 - 4:10 pm
The C component may be near the 0 boundary - in that case the SE is not reliable and the Difftest is probably better.
 Naomi Friedman posted on Tuesday, October 20, 2009 - 4:21 pm
Unfortunately I don't think that's it, because actually, the C loading is .563, SE=.191, p=.003. But the DIFFTEST (df=1) is 2.172, p=.1406.
 Linda K. Muthen posted on Tuesday, October 20, 2009 - 4:23 pm
Then you need to send the relevant files and your license number to support@statmodel.com. There is not sufficient information to answer this question.
 Yvan posted on Friday, July 06, 2012 - 6:49 am
Dear all,
I am running following model:

missing = all (99);
ANALYSIS:

Estimator = MLR;
Type = General;

MODEL:
f1 BY SW3 SW4;
f2 BY ....

f8 ON f2 f6;
f2 ON ...
Output: SAMPSTAT MODINDICES (0) STANDARDIZED
RESIDUAL TECH1 TECH2 TECH3 TECH4
TECH5 FSCOEF FSDET CINTERVAL PATTERNS;


but I do not get the two-tailed p-value...I get Estimates S.E. Est./S.E. Std StdYX. Is this normal? Or what is my mistake?
Thank you so much for your help. I am appreciate this.
Regards
 Linda K. Muthen posted on Friday, July 06, 2012 - 11:13 am
With WLSMV we do not give standard errors for standardized coefficients when the model includes covariates.
 Yvan posted on Monday, July 09, 2012 - 12:41 am
Dear Mrs Prof Muthen,

thank you so much for you fast and helpful answer.
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: