Grouping columns are columns that you want to use to classify your data into categories of information. They can have character, integer, or even decimal values, but the number of unique values should be limited. Grouping columns are either nominal or ordinal.
Note the following:
• If grouping columns are nested, Tabulate constructs distinct categories from the hierarchical nesting of the values of the columns. For example, from the grouping columns Sex with values F and M, and the grouping column Marital Status with values Married and Single, Tabulate constructs four distinct categories: F and Married, F and Single, M and Married, M and Single.
• You can specify grouping columns for column tables as well as row tables. Together they generate the categories that define each table cell.
• Tabulate does not include observations with a missing value for one or more grouping columns by default. You can include them by checking the Include missing for grouping columns option.
• To specify codes or values that should be treated as missing, use the Missing Value Codes column property. You can include these by checking the Include missing for grouping columns option. For more information about Missing Value Codes, see Missing Value Codes in Using JMP.