People overweight shared comparison dimensions even when unique dimensions matter more

Evidence

People also can ease their burden of calculation by comparing alternatives only on the dimension that they have in common. This is fully rational if that dimension is the most important one: but the method will be employed whether or not this is the case, as is brought out by an experiment in which subjects are asked to compare pairs of students. … In their evaluations, subjects weighed the common dimensions more heavily: the student with the higher English skill was likely to be rated as superior overall, even if the gap on this dimension was slight, and the student who lagged here did extremely well on the unique dimension. Furthermore, neither cautioning subjects about the effect nor giving them feedback as to the “correct” answers changed their method. Interestingly, when the subjects were questioned after the experiment, they denied that they had given extra weight to the common dimension (Slovic, 1975; Slovic & MacPhillmay, 1974). — John T. Jost and Jim Sidanius, Political Psychology, 2571

Categorization

• XREF: Connects to Kahneman’s attribute substitution, Tversky’s feature matching, and the general similarity-comparison literature.
• SEED: This common-dimension weighting bias is a clean behavioral economics finding worth an essay on how it distorts policy and product comparisons.