The Friedman test is useful when we want to test the null hypothesis that many groups have been drawn from the same population. The Friedman is an extension of the previously discussed Wilcoxon test. The Wilcoxon can be used with only two groups, while the Friedman can be used with three or more groups. Like the Wilcoxon, the Friedman uses related or matched groups. Either these groups are matched on the basis of some variable, or the same subjects are used for different treatments . Since the Friedman uses groups that are matched, the number of subjects in each sample is the same. Significance is measured based on the chi-square table. The requirements for this test are:
Ordinal data
Three or more groups
Related groups
Sample drawn at random from matched scores
The formula for this statistic is:
The symbols used in the calculating process of the statistic are: k = number of columns , N = number of rows, = sum of the squared rank sums, df = k - 1 degrees of freedom, equal to the number of columns minus one.