ABSTRACT
This chapter considers random models in which all factors are crossed. A two-factor crossed model was introduced in Example 2.2.2 where the quality control manager of a company designed an experiment to measure the sources of variability in the manufacturing of window screens. The data collected in the experiment are displayed in Table 6.1.1. The data are balanced because an equal number of observations are obtained for each of the 12 operator-machine combinations. Since there is more than one replication of each combination, the interaction effect between machines and operators can be estimated. The balanced two-factor crossed design with interaction is presented in Section 6.2.
