Updating impressions: The differential effects of new performance information on evaluations of women and men

Madeline Heilman, Francesca Manzi, Suzette Caleo

Research output: Contribution to journalArticle


In three experimental studies we investigated whether changes in performance would have different consequences on the competence perceptions and performance evaluations of women and men whose earlier performance had been unmistakably successful or unsuccessful. We reasoned that the ambiguity created by new performance information that was inconsistent with previous performance information would facilitate stereotype-based gender bias. The results provided support for this idea. Whereas no differences emerged between reactions to men and women when performance remained the same, differences emerged when performance changed. Moreover, regardless of the nature of the change in performance, in male gender-typed domains women were evaluated more negatively than men: an improvement in performance had a less beneficial effect for women than for men (Study 1)and a decline in performance had a more detrimental effect for women than for men (Study 2). These effects were shown to be moderated by the gender-type of the field. Women were evaluated more negatively than men whether performance improved or declined only when the field was male gender-typed; when the field was female gender-typed, men were evaluated more negatively than women (Study 3). These findings are consistent with the idea that gender stereotypes and the performance expectations they produce can influence responses to new information about men's and women's performance.

Original languageEnglish (US)
Pages (from-to)105-121
Number of pages17
JournalOrganizational Behavior and Human Decision Processes
StatePublished - May 1 2019



  • Competence perceptions
  • Gender bias
  • Gender stereotypes
  • Lack of fit
  • Performance evaluation

ASJC Scopus subject areas

  • Applied Psychology
  • Organizational Behavior and Human Resource Management

Cite this