I am trying to construct a Path model using one exogenous variable which is nominal. It is basically a 5-category variable indicative of different ways of religious participation. The problem is, it is hard to use that 'single' variable in the model as its nature is different from ordinal or continuous exogenous variables. What I did was that I created four dummy variables on the basis of the first four categories. Ideally this should treat the last category as the 'reference group' so that the coefficients displayed in the .out file show the contrasts between each dummy and the omitted category, just like in a regression analysis. So my first question is whether it is OK to create four dummies to represent the 5-cat variable? I myself am really not sure about it. Meanwhile I have also tried to find articles using nominal exogenous variables, but failed to find any. Therefore I would also like to know if you can recommend me some research studies that use/discuss nominal exogenous variables in the SEM. Thank you very much.