Bias is an ambiguous term, defined in different ways. In conventional usage, it indicates unwarranted prejudice. However, in health research, the notion that bias is invariably bad is biased. Although research bias is an error that is always harmful, researcher bias is a tendency to think in a particular way that may obscure or illuminate attempts to address research questions. Researcher bias begins with pre-judgements whose continuing evaluation infuses the subjectivity of researchers as persons who are socially situated in health sciences focusing on human subjects. Two sets of conditions can make this bias in health researchers useful. The first is volume control. Researchers can vary the loudness of their own and other voices in different research environments. The second condition is smart working. It balances researcher bias against analytic thinking to work creatively with irregularity and uncertainty. Thus, health researchers need to bring their biases to consciousness. A dialectical approach can then engage the biases as conversational partners to innovate health policy that is informed by principles including transparency, good faith and tolerance. Less critical than whether researchers are biased is whose interests their bias serves given their positionality and role.
Keywords: bias; error; prejudice; researcher.
© 2021 John Wiley & Sons, Ltd.