Unconscious Bias

The field of unconscious bias embodies over 30 years of research from various academic disciplines. On this page, the Council presents a small annotated collection of literature on unconscious bias based on gender.

Title: Beyond the ‘Chilly Climate’: Eliminating Bias Against Women and Fathers in Academe
Author(s): Williams, Joan C.; Alon, Tamina; Bornstein, Stephanie
Published: 2006
Journal name: Thought and Action: The NEA Higher Education Journal, Fall 2006, pp. 79-96
Document available online at: http://www.nea.org/assets/img/PubThoughtAndAction/TAA_06_09.pdf 

In this review of research article, the authors assert that academia is a difficult workplace for female academics (and male academics who wish to take active roles in their family life). Gender stereotypes driven by unconscious or implicit biases affect a wide range of employment issues—e.g., performance appraisal, attribution of accomplishments, promotion process, etc., in a field that places high-level, constant pressures on achievement and advancement of knowledge. Female academics report that their achievements go unrecognized and are not perceived as high quality, with their successes attributed to luck, whereas men’s are understood as skills-based. Female academics in sex stereotypic roles find acceptance among male colleagues as the non-threatening and nurturing “mother,” as the “princess” who aligns herself with a powerful, insider man, or as the “pet,” cheerful and deferential. Those female academics unwilling or unable to assume such roles become rejected. Motherhood or taking family responsibility has a heavy price, too, as female academics face the “maternal wall,” presumed to be less than serious scholars or are overtly discouraged from having children. Male academics going against gender stereotype to take active roles in their family responsibilities are also penalized. More academics are (successfully) pursuing legal remedies when faced with such “chilly climate” issues regarding family responsibilities and the academe’s rigid response to its workers’ needs. The authors argue that the academe should take these trends seriously and address work-family conflicts that are derived from gender stereotypes and biases.

“Opt Out” or Pushed Out?: How the Press Covers Work/Family Conflict
Author(s): Williams, Joan C.; Manvell, Jessica; Bornstein, Stephanie
Published: 2006
Journal name: Center for WorkLife Law, University of California, Hastings College of the Law
Document available online at: www.worklifelaw.org/pubs/OptOutPushedOut.pdf

The so-called “Opt Out Revolution,” a narrative of women choosing to leave the world of paid labor, is re-visited in this content analysis of 119 news articles. The authors found that the media coverage overly simplified the notion of “opting-out” and rendered an unrealistically rosy picture of women leaving paid work. They argue that many women are being pushed out of the labor market, rather than freely opting-out, because of issues such as inflexible work/life conflicts, stalled progress in the division of household work, new cultural norms of intensive parenting/mothering, and workplaces that reinforce traditional gender-stereotypic roles, particularly through assumptions about the impact of motherhood on a worker’s productivity level—what the authors called, “maternal wall stereotyping.”

Interventions that Affect Gender Bias in Hiring: A Systematic Review
Author(s): Isaac, Carol; Lee, Barbara; Carnes, Molly
Published: 2009
Journal name: Academic Medicine (2009), Vol. 84, No.10, pp. 1440-1446
Document available online at: http://facultyhiring.uoregon.edu/files/2011/05/Isaac-Lee-Carnes-2009-26cqtlj.pdf

The authors conducted a systematic review of 27 experimental studies of gender bias in employment settings and interventions to reduce such biases. Beyond documenting pervasive negative gender biases, key findings from the studies included:

  • Evaluations were particularly negative when women applied for stereotypically male-dominated positions, including consistent recommendations for lower salaries.
  • For those male-dominated positions, women who interviewed “agentically” received lower ratings than their equivalent men, unless they exhibited gender-stereotypic, “communal” qualities. 
  • Women who expressed anger or were perceived as self-promoting were negatively evaluated especially when seeking high-status positions, whereas displaying anger helped male candidates.
  • When applicants—male and female—wore a “masculine” scent, the chances of being hired increased and decreased when wearing a “feminine” scented perfume.
  • Men and women who looked more stereotypically masculine were more likely to be hired for leadership positions.
  • The gender of the evaluators was not a determining factor; men and women evaluated female candidates more negatively than male candidates.
  • Emphasis on job-relevant competencies and credentials helped reduce bias against women.
    The authors conclude the article with a discussion on future research and recommendations.

State of the Science Implicit Bias Review 2013
Author(s): Staats, Cheryl
Published: 2013
Journal name: The Kirwan Institute for the Study of Race and Ethnicity, Ohio State University
Document available online at: http://kirwaninstitute.osu.edu/docs/SOTS-Implicit_Bias.pdf

The Kirwan Institute for the Study of Race and Ethnicity produced this “State of the Science Review of Implicit Bias Learning,” a 30-year collection of academic research on implicit/unconscious bias and related topics organized thematically. The article focuses primarily on race and ethnicity-based biases and provides basic background information on key concepts (such as “implicit and unconscious,” “bias”) and different ways to measure/test biases. It also reviews literature on the behavioral impact of these biases particularly in the areas of education, criminal justice and health. Notable studies summarized in this report conclude that most Americans have pro-White and anti-Black biases—both Whites and Blacks, and in children as young as six-year olds. Scholars hypothesize that non-White, non-dominant groups’ pro-White biases can be attributed to people’s desire to maintain existing social hierarchies.

Implicit Bias: A Primer for Courts: Prepared for the National Campaign to Ensure the Racial and Ethnic Fairness of America’s State Courts
Author(s): Kang, Jerry
Published: 2009 
Document available online at: http://www.americanbar.org/content/dam/aba/migrated/sections/criminaljustice/PublicDocuments/unit_3_kang.authcheckdam.pdf 

Kang writes this primer and literature review on implicit/unconscious bias for court and legal professionals. The author frames implicit biases as our brain’s way of organizing a complex world (objects, people, concepts, etc.) by using mental schemas and asserts that these biases are pervasive and systematic in all societies, among all populations. The document covers how to define implicit bias and identify its underlying causes, related concepts (e.g., stereotypes, attitudes, and discrimination), common ways of measuring it, and what it means in the courtroom. In addition, he presents a body of literature that illustrates real-life implications of implicit bias, in realms from employment to legal decisions to voting behavior. Kang concludes with notes on the importance of increasing awareness of and engaging interventions to prevent these biases in the courtroom.

Implicit Bias in the Courtroom
Author(s): Kang, Jerry; Bennet, Mark; Carbado, Devon; Casey, Pam; Dasgupta, Nilanjana; Faigman, David; Godsil, Rachel; Greenwald, Anthony G.; Levinson, Justin; Mnookin, Jennifer
Published: 2012
Journal name: UCLA Law Review (2012). Vol. 59, pp. 1124-1186
Document available online at: http://faculty.washington.edu/agg/pdf/Kang&al.ImplicitBias.UCLALawRev.2012.pdf

The authors of this article—a team of social science and legal experts and a federal judge interested in implicit bias—offer answers to the question of what can be done to reduce implicit bias in the courtroom. In addition to presenting an extensive literature review on implicit bias (including related concepts such as explicit bias and structural forms of bias), the authors provide empirical evidence and measurement of implicit bias and legal analyses of implicit bias as it operates in two court cases, one criminal and the other civil. The article concludes with specific interventions proven to be effective in reducing implicit bias in the judicial system.

Can I get a Witness? Challenges of Using Expert Testimony on Cognitive Bias in Employment Discrimination Litigation
Author(s): Bielby, William T.
Published: 2003
Journal name: Employee Rights and Employment Policy Journal (2003), Vol. 7, No. 2, pp. 377-400
Document available online at: http://www.kentlaw.edu/ilw/erepj/abstracts/v7n2/bielbyabstract.html

In this review of research article, the author presents social psychology research on unconscious biases in employment settings, often involving gender stereotypes, and ways to present psychological research as a witness/expert in litigations. He asserts that people’s perceptions and behavior are filtered through gender stereotypes, despite personal beliefs and commitment to egalitarianism. This includes reliance on gender schemas that designate women as oriented towards family and men towards their careers. They attribute women’s success to external factors such as luck, help from others or special circumstances, whereas men’s success is attributed to internal factors such as skills and ability. Gender stereotypes and biases are particularly pertinent in traditionally male-dominated environments such as managerial and professional ranks of large corporations or hierarchical organizations. Research supports the idea that stereotypes and biases are more likely to be activated when decision-making is informal, non-transparent and subjective; then, evaluators’ internal schemas about women generally influence their decisions, as opposed to specific knowledge about particular women evaluees. The author concludes the article with advice for both attorneys and social scientists who are often asked to participate as experts in discrimination trials.

Competent Yet Out in the Cold: Shifting Criteria for Hiring Reflect Backlash Toward Agentic Women
Author(s): Phelan, Julie E.; Moss-Racusin, Corinne A.; Rudman, Laurie A.
Published: 2008
Name: Psychology of Women Quarterly (2008), Vol. 32, pp. 406-413
Document available online at: http://onlinelibrary.wiley.com/doi/10.1111/j.1471-6402.2008.00454.x/abstract 

Researchers explored how people evaluated competent and confident applicants for leadership positions with particular interests in differential treatment of “agentic” women. Research participants–428 in total–were recruited and randomly assigned to watch videotapes of applicants for computer lab manager positions and to assess them in terms of competence, social skills, and hireability. The applicants were male and female actors trained to exhibit either competence or communal interview styles. This study found backlash effects, much like similar studies in the recent past. Agentic women were deemed competent but lacking in social skills, compared to their agentic male counterparts. Male agentic applicants were rated more hireable than their female agentic counterparts. The gender of the evaluators did not make any difference. When it came to agentic women, the evaluators weighted social skills over competence in the hiring criteria; for other candidates, competence weighted more. The article includes reflections on the real-world implications of the findings and suggestions for future research.

Implicit Gender Bias in the Legal Profession: An Empirical Study
Author(s): Levinson, Justin D.; Young, Danielle
Published: 2010
Journal name: Duke Journal of Gender Law and Policy (2010), Vol. 18, pp. 1-44
Document available online at: http://scholarship.law.duke.edu/cgi/viewcontent.cgi?article=1182&context=djglp

Women and men graduate from law schools at similar rates, yet the representation of women in high-level legal positions—in law firms, academia, and the judiciary—has not progress as well. The authors suggest that decision-makers’ implicit gender bias could explain this gender leadership gap and explored their hypothesis through this experimental study. Law students participants (50 in total) were tested on whether they held implicit gender-biases towards women in the legal field and the extent to which these biases were linked to discriminatory decision-making, as measured through hypothetical hiring decisions or hypothetical budget cuts for student groups (i.e., women’s organization). Researchers found pervasive implicit gender bias among female and male law students; participants associated judges with men more often than with women and paralegals with women more often than with men; and they linked home and family more to women, whereas the workplace was more linked to men. However, researchers also found that despite these biases, when faced with employment or budget-related decisions, participants made gender-neutral decisions. In the end, the authors express cautious optimism for future legal professionals’ ability to make non-biased decision.

Science Faculty’s Subtle Gender Biases Favor Male Students
Author(s): Moss-Racusin, Corinne A.; Dovidio, John F.; Brescoll, Victoria L.; Graham, Mark J.; Handelsman, Jo
Published: 2012
Journal name: Proceedings of the National Academy of Sciences (2012), Vol. 109, No. 41, pp.16474-16479
Document available online at: http://www.pnas.org/content/109/41/16474.full.pdf+html 

In a randomized, double-blind study, science faculty members of research universities were recruited as research participants and assigned to two sets of application materials for a hypothetical laboratory manager position. Unbeknownst to the research subjects, the application materials were identical, except for gender. Researcher found that both male and female participants in the study deemed male applicants more competent and hireable than their female competitors. In the conclusion, the authors highlight how pervasive implicit gender bias is among scientists and note the long-term problem posed by the growing under-representation of women in the scientific community.  

Can the Language of Tenure Criteria Influence Women’s Academic Advancement?
Author(s): Marchant, Angela; Bhattacharya, Abhik; Carnes, Molly
Published: 2007 
Journal name: Journal of Women’s Health (2007), Vol. 16, No. 7, pp. 998-1003
Document available online at: http://www.ncbi.nlm.nih.gov/pubmed/17903076

Despite higher rates of enrollment of women in medical schools, women often do not occupy high-status, elite roles in academic medicine, such as deans. The authors speculate that this phenomenon could exist in part because of the mere presence of the word “leader” in job descriptions. They contend that the word can trigger unconscious, gender-stereotypic notions among evaluators and decision-makers—that is, leaders are men, not women. The authors compared schools that have announcements with and without the word “leader” in their tenure criteria description. They found that when the word “leader” was included, the school had lower rates of tenured female faculty, and vice versa. The authors speculate that this may be due to the gender-linked association that the word “leader” has for evaluators. The authors conclude the article with recommendations for institutions and directions for further research. 

Can an Agentic Black Woman Get Ahead? The Impact of Race and Interpersonal Dominance on Perceptions of Female Leaders
Author(s): Livingston, Robert W.; Rosette, Ashleigh S.; Washington, Ella F.
Published: 2012
Journal name: Psychological Science (2012), Vol. 23, No. 4, pp. 354-358
Document available online at: http://www.ncbi.nlm.nih.gov/pubmed/22421203

Research has shown that White female leaders are subjected to the so-called “agency penalty”. Are Black female leaders subjected to the same treatment? That is the core question behind this research. It considers two hypotheses: 1) Black female leaders would face harsher treatment due to their identity as women and as Black—the “double-jeopardy” hypothesis; 2) due to their subjectivity--existing in a conceptually “invisible” space, not being White and male—as Black female, they would face less agency penalty compared to White female and Black male leaders, as they are buffered by their “invisibility”. In this experimental study, 84 non-Black participants were recruited for an online study. Each participant was assigned to one of eight conditions, and these conditions were derived from different combinations of leader’s race, gender, and behavior: 2 (race: White vs. Black) * 2 (gender: male vs. female) * 2 (behavior: dominant vs. communal). Research participants rated the leader in terms how well the leader handled situations with his/her subordinate, his/her likability, expected salary, etc. Key findings were: 1) dominant leaders were perceived more negatively; 2) dominant White female leaders were perceived more negatively than dominant Black female leaders; 3) no difference was observed when female leaders were communal; and 4) dominant Black male leaders were perceived more negatively than dominant White male leaders. The authors state, however, that this research does not imply that Black women leaders have the same standing as white male leaders do—as indicated by the underrepresentation of Black women leaders. Researchers conclude the article by recommending future research in this area.