Hi -

Why log-linear models can reflect a significant partial association
between factors  when none of the cells are significant?.

Thank you very much

Miguel Verdu

Reply via email to