GraphPad Statistics Guide

Notes of caution for statistical novices

Notes of caution for statistical novices

Previous topic Next topic No expanding text in this topic  

Notes of caution for statistical novices

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

Our goal with Prism has always been to make basic biostatistics very accessible and easy. Two-way ANOVA is pushing the limits of "basic biostatistics". Multiple comparisons after two-way ANOVA stretch this definition even more. If you haven't taken the time to really understand two-way ANOVA, it is quite easy to be misled by the results. Beware!

Two-way ANOVA is not a topic that is easy to master. In addition to reading textbooks, also consider getting help.from someone with more experience.

Prism also offers to fit a mixed model to repeated measures data. Understanding this fully is even more complicated.

Before getting lost in the many choices for multiple comparisons, first articulate clearly the scientific goals of the study. Don't articulate your goals in terms of ANOVA (looking for interactions) and avoid the word "significant" which often leads to muddled thinking. Figure out what you really want to know. Then figure out the best statistical approach to getting the answer.

In Prism, the two-way ANOVA analysis can be used when, as the name suggests, there are two factors. None, one, or both of the factors can be repeated measures.