ABSTRACT
When used to test hypotheses, standard parametric statistics such as analysis of variance (ANOVA) require the samples and data involved to adhere to at least one restrictive assumption. If data fail to meet the conditions laid down in such assumptions, any conclusions drawn from the analyses can be suspect. Randomization methods can also be used to test hypotheses but require fewer assumptions. Given this extra ¥exibility, it is surprising that there is not a greater awareness of randomization tests. In this chapter we will examine randomization or permutation testing and its potential value to ecological and žsheries research.
