Daniel Arad posted on Saturday, October 03, 2009 - 2:07 pm
Dear Sir, Madam
I am using MPlus for an economic model with MIMIC configuration. It is similar to example 5.8 but with a single continuous latent variable and only categorical indicators("y"s). "x"s are both categorical and continuous.
My query is: if all indicators are categorical; how does Mplus deal in such a case with the normalization of the indicator? What is the simplest way to define this in Mplus?
The metric of a factor is set by fixing the first factor indicator to one or fixing the factor variance to one. This is done the same way irrespective of the scale of the factor indicators. Factors are continuous even when factor indicators are not.