In reviewing research in the field of stereotype processing, it was noted that as definitive measures of automatically activated implicit stereotypic knowledge are not established, exploring the relationships between measures purporting to measure such knowledge is critical. Once relationships between measures are established, theoretical explanations of the cognitive processes involved will be better informed.
While measuring implicit stereotypic knowledge indicates, by definition, that the knowledge is automatically activated, it may be the case that the knowledge activated is similar to knowledge of stereotype generated in a more controlled manner. As little research has been conducted to determine the relationship between the content of implicit stereotypic knowledge, and that of general knowledge of stereotypes, this study will employ both types of measures. The measure of stereotype knowledge to be employed in this study, the Others Views Checklist, will be treated as a possible correlate of the implicit knowledge measures.
Based on the features of paradigms commonly employed to measure implicit stereotypic knowledge (Dovidio et al., 1997; Foddy, 1997; Wittenbrink et al., 1997), a category-trait priming paradigm with a subsequent word association task has been selected as a potential measure of implicit knowledge. The previously described Sentence Completion Task (von Hippel et al., 1997) will also be employed as a proposed measure of implicit stereotypic knowledge.
To ascertain explicit beliefs this study will employ .two measures of attitudes towards women, and one measure of stereotype-related beliefs about male and female characteristics.
To determine whether the level of stereotype endorsement is related to expectations for competence, competence-related behaviour in a mixed-sex, task-related setting will be measured, and following this, its relationship to stereotypic beliefs will be ascertained.
Owing to the fact that in task-related settings with other differentiating characteristics controlled, expectations for competence are higher for males than for females, men will be more likely to reject the influence of their female counterpart than accept it (Foddy & Smithson, 1996; Pugh & Wahrman, 1983; Wagner, Ford & Ford, 1986). However, as it has also been found that different types of women are perceived differentially with respect to competency (Glick et al., 1997), it may be the case that certain types of women will engender lower rates of influence rejection in males than will other types of women. Therefore, in this study two principal subtypes of the female stereotype will be incorporated into the design. In this, the male participants, who will be lead to believe they are working jointly on a task with a female partner, will be exposed to written information describing either a traditional (feminine) subtype, or a modern (career oriented) subtype. This type of exposure may serve to prime stereotype-related information, and concomitantly increase the saliency of gender as a status characteristic within the setting (Beckett & Park, 1995). This feature of the design will enable further assessment of the relationship between stereotype endorsement and behaviour perceptions of women's competency.
As it is argued that females are generally less motivated to accept negative stereotypes of women, it is likely that the role played by implicit knowledge and/or explicit attitudes of stereotypes in behavioural outcomes will be more evident in a male sample than in a female sample (Foschi, 1996). For this reason, only male participants will be tested in this study, and accordingly the predictions made will apply only to men.