Multicollinearity in SEM

[Anonymous]

Hi,

I would like to understand in detail how SEM handles multicollinearity? In case of path model or structural model, if the variables are correlated, does this have bearing on results? Are coefficient and fit indices robust if we have correlated indicators or SEM in the back end handles it. Please throw some light on this confusion.  

[Phil Wood]

I'd be interested in learning a bit more about your question. Briefly, collinearity affects structural models the same way as it does for regression. The difference, however, is that highly collinear predictors are often summarized by means of a latent variable which then is used to predict a criterion (or criterion factor). As a result, collinearity in manifest variables in regression may find non-significant results, whereas a corresponding latent variable model uses precisely the high degree of correlation as evidence for the existence of a factor. Does that help? Phil

On Thu, Nov 15, 2018 at 9:25 PM Anonymous <pankaj.s...@gmail.com> wrote:

Hi,

I would like to understand in detail how SEM handles multicollinearity? In case of path model or structural model, if the variables are correlated, does this have bearing on results? Are coefficient and fit indices robust if we have correlated indicators or SEM in the back end handles it. Please throw some light on this confusion.  

--
You received this message because you are subscribed to the Google Groups "Structural Equation Modeling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structural-equation-...@googlegroups.com.
To post to this group, send email to structural-equ...@googlegroups.com.
Visit this group at https://groups.google.com/group/structural-equation-modeling.
For more options, visit https://groups.google.com/d/optout.

[pankaj singhi]

Hi,

Thanks for the response. A bit clear now on how in general collinearity is handled. I will still read some materials shared in the link and will get back if there is any confusion.

Here is my question in particular to lavaan only. After going through lot of questions in forum i got to know that its not pretty straightforward to understand Non positive definite warning of covariance matrix. My concern is when i run path model with lavaan, i never get this warning but when i run a proper structural model i very often get this warning. I believe(might be wrong) one of the reason for this warning is multicollinearity between variables. But not able to understand how multicollinearity is handled differently in case of path model and structural model.

Kindly help.

[Phil Wood]

Can you give us any examples of what you're dealing with?