Statistics > Nonparametrics > Wilcoxon Signed Rank Test

This utility performs calculations for the Wilcoxon Signed Rank Test, which is a non-parametric hypothesis test for the median of a single sample or of two samples of matched pairs. The null hypothesis H0 of a claim is median = median0, where median0 is the hypothesized median. The alternative hypothesis H1 can be one of the following: median < median0, median > median0, or median ≠ median0.

Let x be a value in a single sample or the difference of matched pairs in two samples. For each value of x, the absolute value of the difference between x and the hypothesized median median0 is computed. Differences of 0 are discarded. Let n be the number of nonzero differences. Ranks from 1 to n are then assigned to each x based on the ascending order of the differences. Mean of ranks are assigned to tied values. A negative (-) sign is assigned to a rank if its corresponding x is below median0 , and a positive (+) sign is assigned if x is above median0.

The rank sums are calculated as follows:

If n is less than or equal to 30, the exact p-value and critical value for the rank sum is calculated. The test statistic R is the rank sum corresponding to the type of test described above. The p-value for a one-tailed test (left-tailed or right-tailed) is P(X ≤ R) where X is a random variable representing the rank sum. The p-value for a two-tailed test is 2P(X ≤ s).

If n is greater than 30, normal approximation is used. The test statistic is

The p-value for a one-tailed test (left-tailed or right-tailed) is P(z ≤ zw) where z is a random variable representing the z-score. The p-value for a two-tailed test is 2P(z ≤ zw).

To use the utility, follow these steps: