Abstract
This study investigated the validity of 5 approaches to computing an accuracy score unaffected by response biases. 117 judges made predictions of four targets' self-ratings on a 10-point bipolar rating scale. Results indicated that each method is eliminating the response biases it attempts to eliminate. Examination of the response biases eliminated by each method revealed that RD3 scores were the most appropriate measure of accuracy since these scores concentrated on eliminating rating response biases.
Get full access to this article
View all access options for this article.
