The use and interpretation of the Friedman test in the analysis of ordinal-scale data in repeated measures designs.

The purpose of this paper is to review the use and interpretation of the Friedman two-way analysis of variance by ranks test for ordinal-level data in repeated measurement designs. Physical therapists frequently make three or more repeated measurements of the same individual to compare different treatments, or to assess the effect of a single treatment over time. When the measurements are ordinal-scaled, such as some ratings of functional status and muscle strength, statistical significance may be determined by the Friedman test. We illustrate the use of the Friedman test and a post hoc multiple comparison test with data from 27 subjects whose performance on a lifting task was rated on three occasions by use of an ordinal scale. We discuss the interpretation of ordinal-level data and recommend that therapists understand the limitations a measurement scale imposes on the inferences that can be made from these tests.