[A comparison of two models of gender-related schematic processing].

An experiment was conducted to compare two models of gender-related schematic processing: Bem's gender schema theory (1981) and self-schema theory by Markus, Crane, Bernstein, and Saladi (1982). While the former argues for a construct that is unidimensional, the latter proposes separate masculine and feminine schemata. Two hundred and thirty-six college students (111 males and 125 females) completed Bem Sex Role Inventory, and then estimated subjective percentages of males and females who had each trait of BSRI. The subjective percentages were used to compute sex-typing scores. After the questionnaires, the subjects attempted free recall of the inventory items, and clustering scores were computed from the results. Data analyses, including those of covariance structure, yielded the following results: 1) The fit of the gender schema model was better than the self-schema model. 2) A strong gender schema was related to a high-congruent self-concept score, a low sex-incongruent self-concept score, a high sex-congruent sex-typing score, a high sex-incongruent sex-typing score, and a high sex-congruent clustering score. As a whole, the results could be summarized to support the gender schema theory.