The Role of Evaluation Order in Anchoring Effects: A Reverse Design Investigation

This study investigates the cognitive mechanisms underlying the evaluative moderation of the anchoring effect. Previous findings indicate that the anchoring effect is significantly reduced or eliminated when participants judge an anchor as a “Bad” estimate. Our prior work suggests that this evaluative judgment is not arbitrary but is systematically determined by the consistency between the anchor and the participant’s pre-existing knowledge, measured via their unanchored baseline estimate. The primary purpose of this research is to determine whether this elimination of anchoring is an artifact of the “Good vs. Bad” evaluation question, or if it is a more general phenomenon that persists even in a standard anchoring task. To test this, we employ a "reverse order" design. Participants will first complete a standard anchoring task (Stage 1), and one week later in a subsequent session (Stage 2), they will provide a baseline unanchored estimate and then evaluate whether the anchor they originally saw in Stage 1 was a “Good” or “Bad” estimate of the target quantity. We expect to demonstrate that the moderation effect persists even when the evaluation occurs later. Specifically, we expect to see that anchors later identified as “Bad” in Stage 2 will be associated with a weaker or eliminated anchoring effect in Stage 1, with anchoring measured in relation to the baseline.