I need to compare 11 groups of data, for the same relationship.
My first thoughts were to create 11 regression equations (one for each), and to compare parameters between groups.
Problem is that depended variable is a dummy variable and I need to compute a logistic or probit model. In SPSS everything is okay.
But AMOS give such a nice visualization of the model (s) (relatively complicated), and I want to compute each of logistic (or probit) regression equation in AMOS. Is this possible? If yes, please give me some places where to look for the commands and steps for this kind of analyses
Yes, sort of, but only using the Bayesian approach.
Use lavaan, an R package, and you can do this.
You can also do this using an interaction term (or set of interaction terms) using the logistic regression function in SPSS.
- Solved – How to create item parcels in SPSS Amos
- Solved – Linear Probability Model, Probit and Logistic Models gives different significance level for a variable
- Solved – Comparing three groups on one categorical dependent variable
- Solved – Correlation with binary response and Likert scale response
- Solved – Categorical demographic variables as control variables in SEM