gibbon lab posted on Tuesday, September 16, 2014 - 12:10 pm
I am running a path analysis with a binary outcome. The outcome variable is claimed as categorical in the Mplus program. From the output, it looks to me that Mpus is using a probit model where the residual error is assumed to have the standard normal distribution N(0,1). My question is regarding the STDY output. What is the assumption used for that output? Isn't it the same: a N(0,1) distribution for the residual error? Thanks.