Norms and the relationship of Boston Naming Test performance to vocabulary and education: A review

Abstract

Background: Performance on the Boston Naming Test (BNT) reflects influences from demographic and ability variables that were not well understood at the time of the publication of the test. Means for normal samples with limited educational backgrounds, or limited vocabularies, that fall as much as two standard deviations below the original norms have been reported (Hawkins et al., 1993). Failure to recognise these influences will lead to substantial levels of misdiagnosis of dysnomia. These influences have, however, been obscured by the publication of studies reporting lesser, or no, relationship between vocabulary, or educational level, and BNT performance. Aims: The present paper aims to review studies yielding BNT norms, and to identify factors that obscure the relationship between vocabulary, education, and test performance. Main Contribution: It is established that few studies generating BNT norms have been adequately representative of the population, with the vast majority featuring a disproportionate representation of highly educated subjects. In combination with the psychometric properties of the BNT, these imbalances obscure the relationship between vocabulary, or education, and BNT demonstrated in studies utilising more representative samples. Some attempts to account for education have been distorted by a disproportionate representation of subjects of higher intellectual ability than would be typical for their stated years of schooling. Conclusions: BNT norms should be finely stratified by education. Whenever possible, the clinical interpretation of BNT scores should be further moderated by estimations of premorbid vocabulary.