UPCOMING WEBINAR: From Chat to Agents: The New Phase of AI in Market Research Sometimes, significance testing can change when the structure of the underlying questions change. Often, this is the result of Multiple Comparison Corrections. When multiple comparison corrections are being used on tables, this causes the results of the statistical tests to change when the table is modified and the number of comparisons in the table changes.
See Multiple Comparisons (Post Hoc Testing) for more detailed information on multiple comparison correction.