A big obstacle use of structural equation modeling(SEM) is handling categorical variables. When you want to measure the impact of one or more variables on a categorical variable such as gender, you have trouble using SEM statistical software.

Suppose you want to examine the effect of training and performance on the win of playoffs games. You can define Win latent variable that measures via the Result measurement variable(See Introduction to SEM-PLS). You give the Result variable the numbers 0 or 1. This means that the Result variable is 1 if the football team wins and 0 if it loses.

Now, in this case, with this definition of the Result variable, the results obtained will be completely wrong because the WIN variable must be qualitative. It would be best to use Logistic SEM-PLS to examine the effect of training and performance on the Win variable.

The SEM-PLS software allows you to perform the Logistic model using structural equation methods based on the PLS method(See GLM Regression and Path Analysis).