I am running a path model with a categorical predictor. Specifically, I have fne (social anxiety scores), a mood variable (condition) and the interaction term (fnexcond) as predicting two mediators (tension reduction and enhancement cognitions) and two measures of alcohol urges (pos or neg reinf. urges).
I run the model with the following code:
IAT_TR IAT_EN ON cfne_m ccond fnexcond; NEGreinf POSreinf ON cfne_m ccond fnexcond; NEGreinf ON IAT_TR IAT_EN; POSreinf ON IAT_EN;
Cfne_m WITH ccond fnexcond; Ccond WITH fnexcond; IAT_TR WITH IAT_EN; POSreinf WITH negreinf;
And get this error:
THE MODEL ESTIMATION TERMINATED NORMALLY
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES MAY NOT BE TRUSTWORTHY FOR SOME PARAMETERS DUE TO A NON-POSITIVE DEFINITE FIRST-ORDER DERIVATIVE PRODUCT MATRIX. THIS MAY BE DUE TO THE STARTING VALUES BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE CONDITION NUMBER IS 0.167D-15. PROBLEM INVOLVING PARAMETER 31.
According to TECH1, parameter 31 is the condition variable. Initially it was coded using -.5 and .5, so I tried it with a 0 and 1 coding as well and received the same error. I know the variables are not skewed or kurtotic.