Investigating stereotype structure with empirical network models

In this work we present empirical network models as a new approach in the investigation of stereotype structure. We will argue that empirical network models can provide more insight into stereotype structure because they do not suffer from the inherent constraints of factor analysis and multidimensional scaling (e.g., group features interpreted homogeneously only on the basis of their shared variance, impossibility to adequately represent cognitive schemas, difficulties to make inferences on the basis of dimensions potentially overlapping). In the present research we show how empirical network models can represent stereotypes as dynamic cognitive structures clustered in different substructures. These structures will be based on both the stereotype content and the co-occurrence of features in each group target. Additionally, this research shows how using empirical networks can contribute to broadening the interpretation of stereotypes representing them in the framework of prejudice or intergroup attitudes.