Assumptions in SEM PreviousNext
Mplus Discussion > Structural Equation Modeling >
 JMC posted on Wednesday, June 12, 2013 - 10:45 am

Thank you in advance for your help. Thus far I have established my assumptions using observed scales, but my understanding is that these will not be entirely reflective of my latent scales. Is it appropriate to check assumptions using observed variables in SPSS before running the latent analysis? If they should be examined in a latent manner how can I check these (e.g. Normality, Multicollinearity, Linearity, Homoscedasticity) in MPlus. In my data the most important of these by far is multicollinearity because I suspect a problem with this.

I have tried to used some of the tech options, but have not been able to use them with my code.

Many thanks,
 Bengt O. Muthen posted on Wednesday, June 12, 2013 - 3:31 pm
That's a big, general topic which is suitable for SEMNET. The related question of how Mplus can be used to connect classic test theory and modern psychometrics is described in the 2011 Raykov-Marcoulides book "Introduction to Psychometric Theory".
 JMC posted on Friday, June 14, 2013 - 5:39 am
Thank you!
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message