EXPRESS: Occupational Stereotyping - Individual Differences in Susceptibility and in Response to Training

In previous studies (Finnegan, 2014; Finnegan et al., 2015a, 2015b, 2023) we investigated effects of training on the reduction of occupational stereotypes indicated by responses to the question of whether two words (e.g., uncle/builder or aunt/builder) could refer to the same person. In each study we also collected individual difference measures of sexism, sexist language, sex typing and fear of negative evaluation. Considering the individual studies, the relations between the individual difference measures and measures of stereotyping were either absent or not consistent and were not reported in the published versions of the studies. In this paper we analyse data from 299 participants across eight studies in an attempt to identify consistent patterns. We looked at five individual difference measures (ASI, ISANL-G, BSRI, EthicsQ, BFoNE) and four measures of performance: initial differences in speed and accuracy of responding to stereotypically congruent (nurse-aunt) and incongruent (nurse-uncle) word pairs and changes in those differences with training. We found that one individual difference measure, the Ambivalent Sexism Inventory (ASI; Glick & Fiske, 1996), predicted differences on both initial performance and response to training, though unexpectedly people with high sexism scores were more, not less, susceptible to training. Scores on a second measure, The Inventory of Attitudes towards Sexist and Non-sexist Language General (IASNL-G; Parks & Roberton, 2000) was also predictive, but made no contribution to either speed or accuracy in addition to that of ASI scores. We discuss the implications of these results and their relation to previous literature.