From: xinrong on
I am trying to do SEM with SAS, my problem right now is that over half
of my exogenous variables are nominal/rank variables. I know that PROC
CALIS or PROC TCALIS requires normal distribution, any suggestions
about how should I deal with this situation?