Two-Way Analysis of Variance

The marginal views of the data in Figure 2 are fine so long as the two factors don’t interact. Experimental factors interact if the effect of one factor depends on the level of the second factor. That is clearly happening in these data. Customers in the South, for instance, respond differently to price partitioning than customers in the Midwest. The question that remains is whether the differences found in Table 1 and Figure 1 are statistically significant.

In keeping with the tabular layout of Table 1, the standard analysis of data from an experiment with two factors is known as a two-way analysis of variance. The method for fitting a two-way analysis of variance resembles that used to fit a one-way analysis of variance ...

Get Statistics for Business: Decision Making and Analysis, 3rd Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.