Crossed categorization effects on the formation of illusory correlations
Article first published online: 19 JUL 2001
Copyright © 2001 John Wiley & Sons, Ltd.
European Journal of Social Psychology
Volume 31, Issue 4, pages 443–466, July/August 2001
How to Cite
Meiser, T. and Hewstone, M. (2001), Crossed categorization effects on the formation of illusory correlations. Eur. J. Soc. Psychol., 31: 443–466. doi: 10.1002/ejsp.55
- Issue published online: 19 JUL 2001
- Article first published online: 19 JUL 2001
- Manuscript Accepted: 10 JAN 2001
- Manuscript Received: 17 MAR 2000
- Alexander von Humboldt Foundation
- William and Flora Hewlett Foundation
This study investigates how stereotypes are formed and whether stereotype formation is reduced by the prevalence of multiple categorizations. Illusory correlations between the desirability of behaviours and two dimensions of social categorization, both containing a majority and a minority category, were assessed in single categorization and crossed categorization conditions. In the single categorization conditions, the usual illusory correlation in favour of the majority category was obtained. In the crossed condition, the combination of the two majority categories was positively discriminated from the remaining three combinations, while no differences were found among the latter. A source-monitoring analysis of assignment frequencies replicated earlier findings that illusory correlations are due to an evaluative guessing bias, rather than to enhanced memory for individual instances of behaviour. The results show inconsistencies with a distinctiveness-based and a social categorization account of illusory correlations, but they can be explained in terms of information loss. Copyright © 2001 John Wiley & Sons, Ltd.