Bias and Misrepresentation of Science Undermines Productive Discourse on Animal Welfare Policy: A Case Study

SIMPLE SUMMARY: Creating good animal welfare-related laws, regulations, and policies depends on accurate knowledge. To that end, scientific reviews that explain and contextualize the relevant research can be powerful tools for informing decision-makers, assuming these reviews represent the state of the scientific knowledge accurately and objectively. In this commentary, we examine the major flaws, biases, and misrepresentations of the scientific literature in one such recent review regarding the welfare and care of captive killer whales. Such pervasive problems, in this or any review, make it impossible to determine the true state of knowledge of the relevant issues, and can ultimately result in misinformed, arbitrary, or even harmful decisions about animals and their care. ABSTRACT: Reliable scientific knowledge is crucial for informing legislative, regulatory, and policy decisions in a variety of areas. To that end, scientific reviews of topical issues can be invaluable tools for informing productive discourse and decision-making, assuming these reviews represent the target body of scientific knowledge as completely, accurately, and objectively as possible. Unfortunately, not all reviews live up to this standard. As a case in point, Marino et al.’s review regarding the welfare of killer whales in captivity contains methodological flaws and misrepresentations of the scientific literature, including problematic referencing, overinterpretation of the data, misleading word choice, and biased argumentation. These errors and misrepresentations undermine the authors’ conclusions and make it impossible to determine the true state of knowledge of the relevant issues. To achieve the goal of properly informing public discourse and policy on this and other issues, it is imperative that scientists and science communicators strive for higher standards of analysis, argumentation, and objectivity, in order to clearly communicate what is known, what is not known, what conclusions are supported by the data, and where we are lacking the data necessary to draw reliable conclusions.