Partners in Wall Street law firms write equally nice things about the work of their male and female junior lawyers, but when they use hard numbers, they rate the men higher, according to a study in the current Social Psychological and Personality Science.
The use of positive language may be to soften the blow of low evaluations or they may be based on lower expectations of female performance based on stereotypes, write Monica Biernat, of the University of Kansas, M.J. Tocci of Fulcrum Advisors and Joan Williams of Hastings College of the Law of the University of California.
The researchers looked at the performance evaluations of junior attorneys working in a Wall Street law firm. The mostly male senior lawyers rated more than 230 junior attorneys35% womenusing both number ratings and writing about one single-spaced page of text.
The numbers are what matter for raises; partnership and promotions go only to those with the highest numbersthe written text simply "explains" the numbers. By the numbers, men significantly outscored the women; the authors estimated about 14% of men and 5% of women were on track for promotion by this standard.
The written evaluations tell a different story. Independent experts, who did not know the gender of the person being written about, rated the competence communicated in writing; men and women equally received generally positive evaluations. When they counted the number of "positive performance words" such as "excellent," "awesome," or "stellar,' women received significantly more of this positive feedback. The men with more positive words had higher numbers, but for women receiving positive words was completely uncorrelated with their numerical ratings.
Because of this inconsistency, the firm was either biased in favor of men in the numerical ratings, or misinforming the women in the written explanations. Because even the very best womenmentioned as partner materialhad lower numbers than comparable men, the authors suggest that there was a male favoritism when using numbers.
"Although the difference in numerical ratings may not seem large," said the authors, "stereotypes led to pro-male bias when it mattered. The firm's reliance numbers for partnership consideration made it three times more likely that men will be promoted to partner."
Explore further: Republicans trust science—except when it comes to health insurance and gay adoption
More information: The article "The Language of Performance Evaluations: Gender-Based Shifts in Content and Consistency of Judgment" in Social Psychological and Personality Science is available free for a limited time at spp.sagepub.com/content/early/… 415693.full.pdf+html