Related-Samples Nonparametric Tests
Identifies differences between two or more related fields using one or more nonparametric tests. Nonparametric tests do not assume your data follow the normal distribution.
Data Considerations. Each record corresponds to a given subject for which two or more related measurements are stored in separate fields in the dataset. For example, a study concerning the effectiveness of a dieting plan can be analyzed using related-samples nonparametric tests if each subject's weight is measured at regular intervals and stored in fields like Pre-diet weight, Interim weight, and Post-diet weight. These fields are "related".
What is your objective? The objectives allow you to quickly specify different but commonly used test settings.
- Automatically compare observed data to hypothesized data. This objective applies McNemar's Test to categorical data when 2 fields are specified, Cochran's Q to categorical data when more than 2 fields are specified, the Wilcoxon Matched-Pair Signed-Rank test to continuous data when 2 fields are specified, and Friedman's 2-Way ANOVA by Ranks to continuous data when more than 2 fields are specified.
- Custom analysis. When you want to manually amend the test settings on the Settings tab, select this option. Note that this setting is automatically selected if you subsequently make changes to options on the Settings tab that are incompatible with the currently selected objective.
When fields of differing measurement level are specified, they are first separated by measurement level and then the appropriate test is applied to each group. For example, if you choose Automatically compare observed data to hypothesized data as your objective and specify 3 continuous fields and 2 nominal fields, then Friedman's test is applied to the continuous fields and McNemar's test is applied to the nominal fields.
This procedure pastes NPTESTS command syntax.