Relative Comparison and Category Membership: The Case of Equity Analysts
Although audience perception is critical to the theory that classification affects rewards, such as ratings or sales, tests of the classification–rewards link occur without directly measuring audience perception. As a result, although a great deal is known about the mean level of rewards as a function of classification, little is known about the individual evaluations that underlie them and that contribute to the variance. I advance a process-based explanation for evaluative outcomes. Individuals make evaluations as a result of relative judgments on a subset of objects, comparing each object under consideration against a small set of others. Categorical boundaries matter to individuals perhaps because of personal preferences but also, importantly, because fit within boundaries determines how strictly to apply performance results. Simply by different individuals examining different subsets of objects, evaluative outcomes can vary dramatically, such that the same object may have different evaluations by audience members. Using recommendations by analysts at U.S. brokerages, I find support for the hypothesis that lower performance of a stock relative to other stocks already rated by a given analyst is associated with a lower likelihood of a high rating by an analyst, but this effect applies only to those stocks that fit clearly into industry boundaries. In general, the results suggest that the positive effect of category membership on evaluative outcomes, well established in prior literature, is contingent on the evaluative processes of individual audience members.