Effects of Nesting and Stacking on Column Proportions Tests

The rule for column proportions tests is as follows: a separate set of pairwise tests is performed for each innermost subtable. To see how nesting affects the tests, consider the previous example, but with Labor force status nested within levels of Gender.

1. Open the table builder again (Analyze menu, Tables, Custom Tables).
2. Drag and drop Gender from the variable list into the Rows area of the canvas pane.
3. Click OK to create the table.

   

   With Labor force status nested within levels of Gender, 14 sets of column proportions tests are performed--one for each level of Labor force status for each level of Gender. The same letter keys are assigned to the categories of Marital status.

   There are a couple of things to note about the table results:

   • With more tests, there are more columns with zero column proportion. They are most common among separated respondents and widowed males.
   • The column differences previously seen among respondents keeping house seems to be entirely due to females.

   To see how stacking affects the tests:

4. Open the table builder again (Analyze menu, Tables, Custom Tables).
5. Drag and drop Highest degree from the variable list into the Rows area below Gender.
6. Click OK to create the table.

With Highest degree stacked with Gender, 19 sets of column means tests are performed--the 14 previously discussed plus one for each level of Highest degree. The same letter keys are assigned to the categories of Marital status.

There are a few things to note about the table results:

• The test results for the 14 previously run sets of tests are the same.
• People who have less than a high school degree are more common among widowers than among married, divorced, or never-married respondents.
• People with some post-high school education tend to be more common among those people who are married, divorced, and never married than among widowers.