Consider a three-dimensional table containing the cross-classification of variables X, Y, and Z. The distributions of X, Y cell counts at different levels of Z can be displayed using cross-sections of the three-way table. These cross-sections are called partial tables. In the partial tables, the value of Z is held constant.
For example, Section 3.2.2 of Chapter 3 discusses health policy opinion data with variables X=stress, Y=opinion, and Z=residence. Table 16.7 displays the two partial tables of the stress × opinion cross-classification for subjects from urban and rural residences.