On partial and conditional association measures for ordinal contingency tables

This paper proposes new non-model based partial/conditional association measures for ordinal contingency table where the main interest is between two variables of interest, adjusting effect covariate(s). We first develop a type scores variables, subcopula scores, taking into account ordering categories and derive regressions describing relationships The proposed are obtained from which uniquely links joint distribution with their marginal distributions. then define to be correlations sets them adjusted by response on theoretically investigate properties corresponding estimators. Finally, performance evaluated simulation studies real-data examples.