November 9, 2010
Recommendation Letters May Cost Women Jobs, Promotions
Women described in social terms that hurt likelihood of being hired
A recommendation letter could be the chute in a woman's career ladder, according to ongoing research at Rice University. The comprehensive study shows that qualities mentioned in recommendation letters for women differ sharply from those for men, and those differences may be costing women jobs and promotions in academia and medicine.Funded by the National Science Foundation, Rice University professors Michelle Hebl and Randi Martin and graduate student Juan Madera, now an assistant professor at the University of Houston, reviewed 624 letters of recommendation for 194 applicants for eight junior faculty positions at a U.S. university. They found that letter writers conformed to traditional gender schemas when describing candidates. Female candidates were described in more communal (social or emotive) terms and male candidates in more agentic (active or assertive) terms.
A further aspect of the study involved rating the strength of the letters, or the likelihood the candidate would be hired based on the letter. The research team removed names and personal pronouns from the letters and asked faculty members to evaluate them. The researchers controlled for such variables as the number of years candidates were in graduate school, the number of papers they had published, the number of publications on which they were the lead author, the number of honors they received, the number of years of postdoctoral education, the position applied for and the number of courses taught.
"We found that being communal is not valued in academia," said Martin, the Elma Schneider Professor of Psychology at Rice. "The more communal characteristics mentioned, the lower the evaluation of the candidate."
A follow-up study funded by the National Institutes of Health is under way and includes applicants for faculty and research positions at medical schools. In the new study, enough applicants and positions will be included so that the researchers can use the actual decisions of search committees to determine the influence of letters' communal and agentic terms in the hiring decisions.
Words in the communal category included adjectives such as affectionate, helpful, kind, sympathetic, nurturing, tactful and agreeable, and behaviors such as helping others, taking direction well and maintaining relationships. Agentic adjectives included words such as confident, aggressive, ambitious, dominant, forceful, independent, daring, outspoken and intellectual, and behaviors such as speaking assertively, influencing others and initiating tasks.
"Communal characteristics mediate the relationship between gender and hiring decisions in academia, which suggests that gender norm stereotypes can influence hireability ratings of applicants," Martin said.
The "pipeline shortage of women" in academia is a well-known and researched phenomenon, but this study is the first of its kind to examine the recommendation letter's role in contributing to the disparity and evaluate it using inferential statistics and objective measures. It's also the first study to show that gender differences in letters actually affect judgments of hireability.
"This research not only has important implications for women in academia but also for women in management and leadership roles," said Hebl, professor of psychology and management at Rice. "A large body of research suggests that communality is not perceived to be congruent with leadership and managerial jobs."
The research team also noted that letter writers included more doubt raisers when recommending women, using phrases such as "She might make an excellent leader" versus what they used for male candidates, "He is already an established leader."
"Subtle gender discrimination continues to be rampant," Hebl said. "And it's important to acknowledge this because you cannot remediate discrimination until you are first aware of it. Our and other research shows that even small differences -- and in our study, the seemingly innocuous choice of words -- can act to create disparity over time and experiences."
Martin, Hebl and Madera's study, "Gender and Letters of Recommendation for Academia: Agentic and Communal Differences," was published last year in the American Psychological Association's Journal of Applied Psychology. They are currently beginning data collection on their next study on recommendation letters for medical faculty positions.
On the Net: