Sex differences on the Rotter I–E scale.

Abstract

Studies using the Rotter I-E scale have generally reported similar means and standard deviations for male and female subjects. Factor-analytic studies with the scale have also revealed similar factors across sex of respondent. Yet, when male and female subjects are asked to respond to the Rotter scale according to stereotypical sex roles, differences are striking, suggesting that sex role bias could be operating. A number of investigators also report that the Rotter scale predicts differentially for males and females, especially in studies of achievement behavior. The present study involved a factor analysis and item analysis of the matched scores of 200 male and 200 female subjects who completed the Rotter scale according to the usual instructions. The factor analysis revealed factors that were similar to those derived in prior analyses of the scale. However, males and females responded to different items within the factors, and significant sex differences in responding were found on 8 of 23 keyed items on individual item analyses. Thus, total scores and some factors on the Rotter scale may represent different meanings for male and female respondents. Implications for test construction and validation are discussed.