Race and gender stereotyping in evaluative language

点击翻译按钮获取中文摘要

In everyday life, we often are asked to provide assessments or evaluations of others' abilities. Stereotypes can subtly shape these evaluations and judgments, even among those who view themselves as non-prejudiced. This can be very consequential in certain contexts; for example, hiring and admissions decisions can be based in part on the evaluative language used in letters of recommendation, and the language used may be influenced by gender and racial stereotypes. Further, audiences may also interpret this language with reference to those negative stereotypes about women and ethnic minorities. The proposed research addresses how evaluative language content varies depending on the gender and race of the person being described, and how in turn others interpret and use that content. This research has broad implications for understanding bias in real world, evaluative settings. Further, this research will increase our understanding of barriers that women and members of underrepresented groups face when considering educational and career opportunities in STEM (science, technology, engineering and mathematics) fields. Understanding how and when stereotyping can occur in evaluative contexts is a critical step towards reducing bias, prejudice, and discrimination. The proposed research predicts that stereotypes activate different standards of judgments for members of different groups; therefore, evaluations (adjectives) mean different things depending on the person described. For example, in a masculine work domain where women are stereotyped as less competent, "good" for a woman may mean something objectively less good than "good" for a man. Dr. Monica Biernat, University of Kansas, will examine actual letters of recommendation written for male and female, racial majority and minority applicants to a variety of graduate programs at a public university. These letters will be coded for content including positivity, complexity, and use of ability versus effort terms, to examine differences based on gender and race across disciplines that vary in their gender/race representation. The research will address whether these content features predict admission decisions. The findings of this research could be applied to a broad array of real-life settings where evaluations occur.