Bias in Expert Judgment Within Large‐Scale Science and Technology Delphis: What Do We Know (and Not Know) and Can Utilization of Panelists' Rationales Help Reduce Potential Bias?

In this paper, we offer a critique of Sokolov, Grebenyuk, and Urashima's paper, “Biases in expert judgments in large‐scale S&T Delphi surveys: how to cope with them?”, published in the journal Technological Forecasting & Social Change in 2025. Although their paper makes a valuable contribution by drawing attention to the potential existence of bias in expert judgment, the authors misunderstand both the purpose of Delphi applications and the concept of bias, and, further, their analysis of secondary data lacks rigor. Together, these shortcomings negate the validity of their inferences on the prevalence of expert bias in large‐scale Science & Technology (S&T) Delphi applications as well as the relevance of their proposed remedial interventions. We argue that future empirical work should adopt a more conceptually grounded and methodologically robust approach, one that distinguishes between (i) legitimate expert/non‐expert divergence in opinions and (ii) genuine cognitive or motivational bias in expert judgment. We recommend the utilization of panelists' rationales as a key component of both future practice and underpinning laboratory‐based research.