Naive application of permutation testing leads to inflated type I error rates.

Failure to account for family structure within populations or in complex mating designs via uninformed applications of permutation testing will lead to inflated type I error rates. Careful consideration of the design factors is essential since some situations allow several valid permutation strategies, and the choice that maximizes statistical power will not always be intuitive.