Forgive what may turn out to be a series of very dumb questions. I'm using MPlus for the first time to test an SEM for my dissertation, and I'm not as computer or stats savvy as I should be. I have 7 variables to be tested in the model - 1 independent (a); 3 independent/mediating (b, c, d); 3 dependent (e, f, g). The goal is to test to what extent the 3 mediating variables mediate relationships between the independent variable and each of the 3 dependent variables.
Here are the issues/questions I have getting started.
1) I'm used to SPSS, where all variables are labeled above the column. In MPlus, I see some examples were variables are labeled only by letter (f1, y1, etc.) and some with actual names (age, sex, etc.). How should my variables be labeled in the data file so that MPlus can use them?
2) Do I need to list every single variable that's in the data file (many are not used in the model) before the USEVARIABLES command? If so, should I just create a separate data file with only the variables I need?
2. If you want the data to be read correctly, you must give a name for each column of the data set. The USEVARIABLES option can be used to select a set of variables for the analysis. It is not necessary to create a special data set.