A question about SK tests PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 milan lee posted on Saturday, December 07, 2013 - 10:23 pm
Skewness and Kurtosis tests are recommended in Dr.Muthen's chpt published in 2004. However, as mentioned in the section of "Statistical checking", the SK tests cannot be applied to longitudinal data with missing data or with outliers. So, if I want to keep the missing data using ML, I cannot use the SK tests?
On the other hand, supposedly that the dataset contains no missing data but a high non-normality in the outcomes. Can I still apply SK tests?

Thank you!
 Bengt O. Muthen posted on Sunday, December 08, 2013 - 10:52 pm
Yes and yes. But I've found that the SK testing often isn't that useful so I wouldn't rely very heavily on it.
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: