Hello, I am using JMP 11 to perform discriminant analysis. I included 25 variables in X and want to predict 2 Y variables. Scoring coefficients that are used to get cannon score were 0 in a few X variables. I don't know what coefficient of 0 means. And, I would like to know how to interpret Total Structure (Pooled, or Between structure) that are either positive or negative values. Which structures are correct one to evaluate the correlation with canon1? How can I use total structure score to predict Y variable?

I would appreciate your help.