Abstract

This paper developed a measure of the strength of the association between populations based on ranks appropriate for the analysis of mixed effects model typed data with one observation per cell. We developed a test statistic for the proposed measure. From the result of the analysis, it was observed that  the proposed method is comparable to Kendal’s coefficient of concordance which assume the value zero (0) when there is perfect association and the value one (1) when there is no association whatsoever between the variables of interest.

Keywords: Association, ranks, test statistic, observation.