If you are doing a Driver Analysis and you get a message that the analysis cannot converge, one possible reason is that you are using filtered versions of the model and too many input variables. The small base sizes of some of these groups combined with the large number of variables prevents the model from converging.
If you are using filters, you can see this by removing the filter from one of the models, and then adding the hidden Final Category question in the Crosstab interaction box and re-run the model.
To resolve the errors you're getting, it may be necessary to exclude some of the predictor variables from the model. You can use principal component analysis in Q to reduce the large number of variables to create a smaller set of variables.
How to do principal components analysis in q
Article is closed for comments.