Permuation tests (also called randomization or re-randomization tests) have been around for a long time, but it took the advent of high-speed computers to make them practically available. They can be particularly useful when your data are sampled from unkown distributions, when sample sizes are small, or when outliers are present.

R has two powerful packages for permutation tests – the coin package and the lmPerm package. In this post, we will take a look at the later.

The lmPerm package provides permutation tests for linear models and is particularly easy to impliment. You can use it for all manner of ANOVA/ANCOVA designs, as well as simple, polynomial, and multiple regression. Simply use lmp() and aovp() where you would have used lm() and aov().

Example

Consider the following analysis of covariance senario. Seventy five pregnant mice are divided into four groups and each group receives a different drug dosage (0…