Quasi-asymmetry model for square tables with nominal categories

For an <italic>R</italic>×<italic>R</italic> square contingency table with nominal categories, the present paper proposes a model which indicates that the absolute values of log odds of the odds ratio for rows <italic>i</italic> and <italic>j</italic> and columns <italic>j</italic> and <italic>R</italic> to the corresponding symmetric odds ratio for rows <italic>j</italic> and <italic>R</italic> and columns <italic>i</italic> and <italic>j</italic> are constant for every <italic>i</italic>><italic>j</italic>><italic>R</italic>. The model is an extension of the quasi-symmetry model and states a structure of asymmetry of odds ratios. An example is given.