Basic Analysis > Contingency Analysis > Contingency Analysis Reports > Analysis of Means for Proportions Report
Publication date: 07/08/2024

Analysis of Means for Proportions Report

In the Contingency platform, the analysis of means for proportions (ANOMP) is a multiple comparison method for testing if individual group proportions differ from the overall proportion. For more information about analysis of means methods, see Nelson et al. (2005). See also Example of Analysis of Means for Proportions.

If the Y variable has two levels, you can use this option to compare response proportions for the levels of the X variable to the overall response proportion. This method uses a normal approximation to the binomial distribution. Therefore, if the sample sizes are too small, a warning appears in the results.

The following options appear in the Analysis of Means for Proportions red triangle menu:

Set Alpha Level

Specifies the alpha level used to determine the decision limits in the Analysis of Means for Proportions chart.

Show Summary Report

Shows or hides a report that contains the response proportions and decision limits for each level of the X variable. The report also indicates whether a limit has been exceeded.

Switch Response Level for Proportion

Changes the response category that is used in the analysis.

Display Options

Contains the following options to modify the analysis of means for proportions chart:

Show Decision Limits

Shows or hides the decision limit lines in the Analysis of Means for Proportions chart.

Show Decision Limit Shading

Shows or hides the decision limit shading in the Analysis of Means for Proportions chart.

Show Center Line

Shows or hides the center line for the Analysis of Means in Proportions chart.

Point Options

Specifies the drawing style of the points in the Analysis of Means for Proportions chart. You can choose between vertical needles, connected points, and points only. By default, the chart is drawn with needles that connect the points to the horizontal line that is drawn at the average.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).