Chi-Square

Chi-Square statistical testing capability is available for banner columns and is selected in Setup|Banners|Edit Banner|Statistical Testing. The Chi-Square test is used to test the independence of a pair of variables, one defining a set of banner columns and the other defining a set of rows. The Chi-Square test reports the level at which the test is significant (sometimes referred to as the "p-value").

You can have the Chi-Square test report the complement of the p-value by choosing the Show confidence level option. The Degrees of Freedom can be reported by selecting the Show degrees of freedom option. The Degrees of Freedom will automatically be adjusted when an entire blank row or column is encountered.

Options:

Show confidence level - displays the confidence level. The confidence level is calculated by (1-significance value) multiplied by 100.
Show significance value - displays the significance level (p-value).
Show significance indicators in banner - displays the significance annotation under each banner column.
Show degrees of freedom - displays the degrees of freedom under the Chi-Square.

IMPORTANT: Statistical testing must be selected for a banner AND selected for a table.

Note: A footnote is printed for Chi-Square tests if any cells have an expected value of less than 1 or if more than 20% of cells have an expected value of less than 5. You can suppress the statistical footnote using the Layout menu within the Banner Editor - Banner # dialog box. Deselecting the Show Statistical Footnote Below Table option will suppress the display of the statistical footnote.