THE ASSESSMENT OF SELF-INSIGHT IN JUDGMENT POLICIES

A widely accepted finding in research on human judgment is that people have relatively poor insight into the weighting schemes they use when they make holistic judgments. The empirical research supporting this generalization rests on indices of self-insight that are produced directly by subjects. In a recent study (Reilly & Doherty, 1989), the self-insight of accounting students was assessed by asking them to select their own policies, represented by usefulness indices, from an array of other student's policies. They were able to do so at a rate far in excess of chance. In this replication and extension ofReilly and Doherty (1989), 77 female college students made holistic judgments of the desirability of potential roommates. There were two levels of the number of attributes (6 and 12) and two levels of attribute intercorrelation (orthogonal and representative). Two weeks later a significant number of students in all four groups selected their policies from the set calculated from their and their peers' judgments. Additional analyses revealed that judgment behavior in the representative ecology differed substantially from that in the artificially created, orthogonal ecology. © 1992.