I'm new to this site. I have a question that is so basic I'm embarrassed to ask it here: I am attempting to use loglinear analysis in SPSS for Windows to analyze sequential data from an observational study I am conducting. My problem is: I am having trouble setting up my data in the data editor, so that they will run in the loglinear program. Here are the details: I have a categorical independent variable--amount of help a child requires prior to teacher intervention (much, some, none). I want to test the probability of certain teacher responses (direct, indirect, or observe) following each of these child behaviors. I am trying to conduct a lag sequential analysis using loglinear analysis.

My problem: I cannot (or don't know how to) prepare my data so it may be entered into the loglinear program in SPSS for Windows 14.0. I have made several attempts: I have set up the child variable as a categorical variable (much, some, none), then set up a teacher response variable that is continuous and includes the frequencies of responses to each of the child categories. When I try to run the program, it does not show the nominal variable at all--only the continues variable, teacher response. I have tried entering my case-by-case observations, both as nominal variables, with the thought that the loglinear program would convert the data to frequencies, construct contingency tables, and then compute desired probabilities, determine significance, etc. Nope. When I do this, the window shows no variables (and there is a message saying the variables--both nominal--are not appropriate for this procedure).

Sorry to go on and on, but my main question is: Does anyone know how to set up a loglinear analysis in SPSS for Windows that might be willing to help me?

Thanks!