The comparability of bioassessments: a review of conceptual and methodological issues

Differences in the raw data used in bioassessments and choices regarding how those data are analyzed and summarized can affect inferences regarding the status of ecological resources and, thus, the degree to which we can trust individual ecological assessments, compare assessments across different programs and regions, or share data when developing or refining new endpoint indices. Progress in addressing these issues has been hindered by lack of consensus regarding what a general definition of comparability should be in the context of bioassessments and what measures of comparability are appropriate for ecological data. In this paper, we review the state of knowledge regarding the comparability of assessments as affected by differences in raw data (composition and relative abundance of taxa), derived measures (biotic metrics and endpoint indices), and assessment levels (condition classes). We specifically address the extent to which the comparability of assessments can be compromised by systematic differences in data, discuss the factors known to affect data comparability, and consider the techniques available to evaluate and improve comparability. Rigorous assessment of data comparability should be a standard aspect of quality assurance when developing and applying biological indices.