Cultural bias in the WISC

Abstract

It is often argued that IQ differences between groups of American Blacks and Whites are the result of IQ tests being culturally biased instruments. The present study attempts to determine the existence of cultural bias in the Wechsler Intelligence Scale for Children (WISC) by comparing Black and White children on: (1) loadings of the first principal component on the WISC subtests; (2) the rank order of item difficulty; (3) measuring the contribution to the total variance of the Race by Item interaction obtained by ANOVA; (4) and by attempting to stimulate race differences by within-race age differences. The results indicate that: (1) there is no evidence of specific factors peculiar to either racial group (the groups differ on what is common to all subtests); (2) the rank order of item difficulties is similar in both racial groups; (3) ANOVA reveals a significant Race by Item interaction, but one which accounts for less than five percent of the total variance, less than one percent in the case of age offset analysis; (4) the items which best discriminate between Blacks and Whites at any age level are also the items which best discriminate between age groups within-race. The data support the view that race differences are differences in mental maturity rather than in artifact of biased testing instruments.