The aim of this review is to examine how the choice of test species and study design employed in the use of in situ approaches in ecological risk assessment can maximize the ecological relevance of data. We provide a framework to define and assess ecological relevance that permits study designs to remain focused on the ecological question being addressed. This framework makes explicit the linkages between effects at lower levels of biological organization and higher-order ecological effects at the population, community, and ecosystem levels. The usefulness of this framework is illustrated by reference to specific examples from aquatic ecotoxicology. The use of models as both interpretive and predictive tools is discussed, with suggestions of appropriate methods for different protection goals.