Item Bias in the Assessment of Children's Anxiety: Race and Sex Interaction on Items of the Revised Children's Manifest Anxiety Scale

To evaluate the potential bias (racial and sexual) of items in the Revised Children's Manifest Anxiety Scale, a scale used in the diagnosis and evaluation of emotional disturbance, a race-by-sex-by-items ANOVA was calculated from the responses of more than 5000 children with a Bonferroni-type adjusted follow-up test to identify specific biased items. Numerous indications of item bias were detected, though the direction of the bias tended to be balanced across groups. The race-by-sex-by-items interaction term, while highly statistically significant (p < .001), accounted for less than 1 percent of score variance.