Abstract
The effects of sex referents in test items and subjects' perceived gender roles were assessed in an attempt to explain sex differences in mathematics performance. The arithmetic subtest of the Wechsler Adult Intelligence Scale (WAIS) was administered to 50 male and 145 female college students. Test items were experimentally manipulated so that their orientation was either masculine, feminine, or neutral. Participants were randomly assigned to each group. No significant differences in performance were found. The Bern Sex-role Inventory was used to classify subjects as either masculine, androgynous, or undifferentiated. Irrespective of biological sex, masculine-oriented subjects performed significantly better on the WAIS arithmetic subtest than feminine oriented subjects. No other significant differences were noted. It was concluded that, for adults, personality variables appear to be a more salient factor than biological sex in explaining differences in basic mathematical skill performance.