ABSTRACT
Predictive models such as those described in Chapter 4 can be applied to acute toxicity data and, in many cases, to chronic lethal or sublethal effects data (e.g., Newman 2008). However, as effects become more subtle and dif cult to model, predictive models are routinely replaced by statistical tests that notionally test for the presence of a statistically signi cant effect. Also, statistical testing methods are very often applied to chronic lethal or sublethal effects data for no reason other than presumed regulatory at. This chapter will examine statistical testing as applied to chronic toxicity and sublethal indicators of stress.
