Prev: DI studio add index to dataset
Next: finding the mode
From: sounpra on 25 Mar 2010 11:46
Hi All -

I'm using PROC CALIS is do path analysis where some of my endogenous
and exgenous variables are categorical. Given that one of the
necessary conditions in a path analysis is that all endogenous
variables should be assessed on an interval or ratio-level of
measurement, I’m proposing to perform a path analysis based on a mixed
matrix of Pearson correlation and polychloric/polyserial
correlations.

While the maximum likelihood estimation procedure appears to be fairly
robust against moderate violations of normality, will my results from
CALIS be valid given my mixture of continuous and categorical
variables in a path analysis?

Thanks,

Song
 | 
Pages: 1
Prev: DI studio add index to dataset
Next: finding the mode