Within-Subjects SSCP Matrices

Figure 1. Within-subjects SSCP matrix
Within-subjects SSCP matrix

This table displays the hypothesis and error sum-of-squares and cross-products (SSCP) matrices for testing the within-subjects factors. Since there are four levels of the within-subjects factor and one measure, each matrix has (4-1)*1=3 columns and 3 rows. For example, the 3x3 matrix associated with Intercept in the table is the hypothesis matrix for testing the significance of Week, the within-subjects factor. The matrix associated with PROMO in the table is the hypothesis matrix for testing the significance of the interaction between the within-subjects factor Week and the between-subjects factor Promotion. The matrix associated with MARKETID in the table is the hypothesis matrix for testing the significance of the interaction between Week and Market ID. The matrix associated with PROMO*MARKETID is used for testing the three-way interaction of Week, Promotion, and Market ID.

The error matrix is used in testing each effect. The particular way in which the SSCP matrices are used to test the within-subjects effects depends upon whether the multivariate or univariate tests are used.

Next