By default in Q, NETs are excluded from Column comparisons. This article describes how to include the NET column when performing column comparison stat testing.
The following table shows column comparisons where the main NET column (the first column below) has been included in the testing.
- A table with Compare columns selected in the Show significance menu and a main NET or Total column is present.
- Go to Edit > Project Options > Customize > Statistical Assumptions if you want to make the change across all tables in your project or Edit > Table Options > Statistical Assumptions if you want to make the change at the table level. IMPORTANT: Please read through Statistical Assumptions to understand what each function performs before making updates to the statistical assumption settings in your project.
- Untick Recycle column letters in the Column Comparisons section.
- In the Overlaps menu in the Column Comparisons section, select Independent or Dependent as desired, see Overlaps on our wiki for what this setting does.
- In the Test Type section (also called Statistical tests for categorical and numeric data prior to Q v5.14) it is also recommended that you:
- Change the Proportions setting to Survey Reporter Proportions or Quantum Proportions.
- Change the Means setting to Survey Reporter Means or Quantum Means.
- OPTIONAL: Change Weights and significance to Kish's approximation in the Significance levels section.
- OPTIONAL: Un-tick Within row and span under Column comparisons if you want the Multiple comparison correction to take into account all columns in the table with its adjustment.
- Click OK.
- Configure what columns to test. Right-click one of the column headings and select Comparisons to specify which groups of columns should be compared:
- Select Compare all columns and tick Include NETs,
- OR select Custom and specify which groups of columns should be compared. See How to Specify Columns to be Compared in a Table for more on this.
How to Specify Columns to be Compared in a Table
How to Change Significance Levels for Column Comparisons
How to Interpret Column Comparisons
Article is closed for comments.