GraphPad Statistics Guide

Summary of multiple comparisons available (two-way)

Summary of multiple comparisons available (two-way)

Previous topic Next topic No expanding text in this topic  

Summary of multiple comparisons available (two-way)

Previous topic Next topic JavaScript is required for expanding text JavaScript is required for the print function Mail us feedback on this topic!  

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.

Goal

CI?

Method

Compare every mean to every other mean

Yes

Tukey (preferred)

Bonferroni

No

Holm (preferred)

Newman-Keuls

Compare every mean to a control mean

Yes

Dunnettt

No

Holm