Diagnostic and Prescriptive Bias in School Psychologists' Reports of a Learning Disabled Child

Abstract

Twenty-four experienced school psychologists independently assessed a simulated learning disability case of a fourth grade girl. Bias due to the child's race, socioeconomic status, and the reading level of the child's home school was analyzed with both an unprompted report written by each subject (recall), and a checklist of descriptive and prescriptive statements about the case (recognition). Only 17 of the 744 possible diagnostic main effects and 2-way interactions which might indicate bias were significant at p<.01, while seven such differences would be expected purely by chance. No statements about intelligence, classroom behavior, or social relationships had statistically significant differences. However, the simulated low SES black child did tend to be classified as “not eligible for special education” much more often than either the white or upper class black child on both the written reports and on the checklist, while the upper SES child was labelled learning disabled more frequently than the low SES child on the recognized checklist. Other main effects and interactions were also noted in some statements about developmental history, arithmetic achievement, perceptual-motor integration, and excessive dependence.