Anchoring Bias in Value Function Elicitation Within Multiattribute Value Theory.

This article investigates the impact of anchoring bias—the human tendency to rely heavily on initial information—on the value function elicitation step within the multiattribute value theory (MAVT), focusing specifically on the midvalue splitting procedure. Through an experiment involving 320 participants from six European countries, the study demonstrates that starting points provided by analysts serve as anchors that systematically influence decision makers' judgments, altering attribute-specific value functions and consequently affecting overall decision outcomes. The research further shows that two debiasing strategies, avoiding anchors and employing counter-anchors, effectively mitigate these anchoring effects by producing value functions and decision outcomes that fall between those elicited under low and high anchor conditions. The findings highlight the susceptibility of MAVT and similar decision analysis methods to cognitive biases and underscore the importance of methodological design and debiasing interventions to enhance decision reliability.