The Effects of Actual Behavior vs. Mere Labelling in Attribution of Sex Differences

Abstract

A recurrent problem in interpreting research showing differential reactions to the sexes is that such results may reflect real behavior differences among those observed or differential expectations of those observing. This study used a methodology which allowed these two competing explanations to be tested by assessing whether subjects' ratings of taped interactants whose gender was ambiguous would be determined mainly by what gender the subjects thought the interactants were or by the interactants' behavior. The results showed that the causes of the sex differences in assessments differed depending on what type of assessment was being made. Specifically, behavior differences and not observers' expectations about such differences largely determined observers' ratings of specific attributes of the interactants, but observers' expectations and not behavior differences determined observers' general ratings of masculinity/femininity.