Two rows or two columns

If the data table only has two columns (or two rows), then Prism compares the two values at each row (column) and uses either the Bonferroni or Holm method to correct for multiple comparisons.

More than two rows or columns

If there are more than two rows and columns, then you first need to choose how to define each family of comparisons in the Experimental Design tab. Then you need to choose how to correct for multiple comparisons within each family by making choices in the Options tab. The choices for two-way ANOVA depend on two decisions:

•Your goal. Which comparisons do you want to make?

•Do you want confidence intervals (CI) included in your results? Not all multiple comparisons tests can compute confidence intervals.