SELECTION OF NATIONAL-SCIENCE-FOUNDATION GRADUATE FELLOWS - A CASE-STUDY OF PSYCHOLOGISTS FAILING TO APPLY WHAT THEY KNOW ABOUT DECISION-MAKING

This article describes and evaluates the procedures used by the psychology panel to select the 1990 National Science Foundation (NSF) Graduate Fellows. Two suggestions are offered. Within-rater standardization of ratings could control differences between panel members in use of the rating scale applied to NSF applicants. More important, panel members could concentrate on providing ratings of nonnumerical predictors, such as the quality of an applicant's letters of recommendation or plan of research, and leave integration of these and other predictors (GREs, undergraduate GPA) to an averaging formula. The present procedures for selecting NSF Fellows emphasize holistic judgment of each applicant, despite decades of research indicating the superiority of actuarial over clinical prediction.