Which statement best describes the Wald-Wolfowitz runs test?

Which statement best describes the Wald-Wolfowitz runs test?

• A. It compares the means of two samples.
• B. It is a parametric test for variance.
• C. It analyzes the ranks to compute a correlation.
• D. It looks for runs of scores from the same group within the ranked data to assess whether groups differ.

Answer: (D)

The test is a nonparametric way to check if two samples come from the same distribution by looking at the order of observations. You combine the two samples and order the data by magnitude, then mark which group each observation came from. A run is a consecutive stretch of observations from the same group. If the two groups come from the same distribution, the labels should be fairly intermingled, giving many short runs. If the distributions differ, you’ll see longer runs from one group, indicating a discrepancy between the groups. This approach uses the arrangement of group labels in the ranked data to assess whether the groups differ, without assuming normality or focusing on means or variances.

The test is a nonparametric way to check if two samples come from the same distribution by looking at the order of observations. You combine the two samples and order the data by magnitude, then mark which group each observation came from. A run is a consecutive stretch of observations from the same group. If the two groups come from the same distribution, the labels should be fairly intermingled, giving many short runs. If the distributions differ, you’ll see longer runs from one group, indicating a discrepancy between the groups. This approach uses the arrangement of group labels in the ranked data to assess whether the groups differ, without assuming normality or focusing on means or variances.