How To Override Q's Default Statistical Testing Settings

Introduction

By default, Q automatically takes into account the structure of the data when determining what statistical test to perform and how to run it. You can use the method below to override Q's default selections.

Method

There are a number of situations where it can be useful to override these defaults:

1. When it is desired to show Column Comparisons (pairwise testing) rather than Exception Tests.
2. Controlling which specific columns should be compared. For example, see:
   • How to Specify Comparisons for ANOVA-Based Tests
   • How to Specify Columns to be Compared
   • How To Test Against The NET/Total/Average.
   • How To Create Banner Tables Using Pick One - Multi Or Grid Questions.
3. When attempting to reproduce results computed from other programs. See Why Results in Q Are Different to Those From Another Program.
4. With very small sample sizes it can be useful to turn off Multiple Comparison Corrections and/or increase the Overall significance level.
5. Where there are serious missing data problems in repeated measures data. See How to Perform Column Comparisons with Missing Repeated Measures Data.