| Abstract: | Reports an error in the original article by R. M. Dawes (American Psychologist, 1972[Aug], Vol 27[8], 773-774). On page 774, the following statement appeared: "Rather than ignore this finding, I used it to illustrate the very first principle I discussed: that linear composites based on paramorphic representations are superior to those based on actuarial analysis". The statement should read: "Rather than ignore this finding, I used it to illustrate the very first principle I discussed: that linear composites based on actuarial predictions are superior to clinical judgments when both are based on the same codable input. Nowhere in the article did I state or imply that linear composites based on paramorphic representations are superior to those based on actuarial analysis". (The following abstract of this article originally appeared in record 1990-56997-001.) Responds to A. G. Weinstein's (1972) comments on R. M. Dawes's (see record 1971-25701-001) findings supporting "bootstrapping" in Dawes's study of graduate admissions. Bootstrapping can be used when actuarial analysis is impossible (e.g., criterion information is lacking or unavailable until after decisions are made). It is not implied that linear composites based on paramorphic representations are superior to those based on actuarial analysis. (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
