Nonparametric Tests

Nonparametric tests make minimal assumptions about the underlying distribution of the data. The tests that are available in these dialogs can be grouped into three broad categories based on how the data are organized:

• A one-sample test analyzes one field.
• A test for related samples compares two or more fields for the same set of cases.
• An independent-samples test analyzes one field that is grouped by categories of another field.

One-Sample Nonparametric Tests

One-sample nonparametric tests identify differences in single fields using one or more nonparametric tests. Nonparametric tests do not assume your data follow the normal distribution.

What is your objective? The objectives allow you to quickly specify different but commonly used test settings.

• Automatically compare observed data to hypothesized. This objective applies the Binomial test to categorical fields with only two categories, the Chi-Square test to all other categorical fields, and the Kolmogorov-Smirnov test to continuous fields.
• Test sequence for randomness. This objective uses the Runs test to test the observed sequence of data values for randomness.
• 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.

Independent-Samples Nonparametric Tests

Independent-samples nonparametric tests identify differences between two or more groups using one or more nonparametric tests. Nonparametric tests do not assume your data follow the normal distribution.

What is your objective? The objectives allow you to quickly specify different but commonly used test settings.

• Automatically compare distributions across groups. This objective applies the Mann-Whitney U test to data with 2 groups, or the Kruskal-Wallis 1-way ANOVA to data with k groups.
• Compare medians across groups. This objective uses the Median test to compare the observed medians across groups.
• 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.

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.