Ethnicity and Language Dominance — Influence on the Prediction of Achievement Based on Intelligence Test Scores in Nonreferred and Referred Samples

Abstract

An urban sample of referred and nonreferred Black, Hispanic, and Anglo pupils were assessed with the WISC-R and K-ABC cognitive and achievement scales. Approximately 38% of the Hispanic sample was identified as Limited-English Proficient (LEP). Analysis of various procedures revealed differences in cognitive and achievement performance due to pupils' ethnicity, referral status, and LEP status. Predictive bias was examined using Potthoff's (1966) simultaneous test of regression slopes and intercepts. For both WISC-R and K-ABC IQ measures, a number of regressions on the K-ABC Total and Arithmetic achievement measures evidenced bias across ethnic groups for both referred and nonreferred samples. Bias due to language dominance also was found for WISC-R and K-ABC composite scales. Implications of predictive bias for assessment and placement of minority and LEP children in special education classes are discussed.