|
<< Click to Display Table of Contents >> Navigation: Reference Manual > Tables > Tables overview |
This section describes the various types of tables (cross tabulations) that can be produced using Tables setup.
Tables produced generally have rows and columns with totals to the top and left. The cells of the table show the number of records that fall into that row and that column.
IMPORTANT: The figures in the total row and total column are not the sum of the figures in the body of table; they are the number of records that qualify for that row or column.
A standard table is a normal table which can have rows, columns, or usually both.
It can be filtered: not include all records based on whether a filter definition is passed or not.
It can be respondent weighted if a weight is set globally.
It can be quantity weighted if a weight is set on the table itself.
Various formatting options can be applied to it.
The rows can be shown immediately below the previous table using format CON so that it looks like part of the same table.
See Formats overview for details of how to control the layout of tables.
See also Table statistics, Significance testing and distinction analysis in Data drill down window
An overlaid table is a standard table that will be added to the previous table(s).
A banked table is the name used when overlaid tables are built from independent rows laid side by side.
A grid table is a banked table that is produced using a single table definition and relies on Grid variables with the correct texts and entry names.
An arithmetic table is a table that is built using arithmetic on other tables or constants. If only adding tables together then overlays should be used instead.
Each table used in arithmetic must have the same number of rows and columns.
A syntax table contains some Command Language (CL) script that will be included in the run. It is used to set things that are not available using the menus.