The fragile basic anchoring effect



The anchoring bias, the effect of uninformative anchor numbers on judgments, is a robust finding. In experiments yielding the anchoring bias, typically participants are explicitly asked to compare the anchor and target. A logical question is whether any manner of considering a number will bias peoples' judgment. Wilson et al. (1996) showed that merely presenting a number to people will bias their judgments, a result they termed basic anchoring. The absence of any published studies that followed up on Wilson et al.'s work prompted us to examine the basic anchoring effect in three experiments. The three experiments (N = 881, 205, and 117) suggest that basic anchoring is a weak effect limited to the precise manipulations used by Wilson et al. Trivial changes such as altering the order of a series of anchor numbers, or using different anchor numbers, eliminated the bias. Our findings suggest that basic anchoring has a much narrower scope of impact than traditional anchoring effects. Copyright © 2002 John Wiley & Sons, Ltd.