gender bias assignment

• SUGGESTED TOPICS
• The Magazine
• Newsletters
• Managing Yourself
• Managing Teams
• Work-life Balance
• The Big Idea
• Data & Visuals
• Reading Lists
• Case Selections
• HBR Learning
• Topic Feeds
• Account Settings
• Email Preferences

Research: How Bias Against Women Persists in Female-Dominated Workplaces

• Amber L. Stephenson,
• Leanne M. Dzubinski

A look inside the ongoing barriers women face in law, health care, faith-based nonprofits, and higher education.

New research examines gender bias within four industries with more female than male workers — law, higher education, faith-based nonprofits, and health care. Having balanced or even greater numbers of women in an organization is not, by itself, changing women’s experiences of bias. Bias is built into the system and continues to operate even when more women than men are present. Leaders can use these findings to create gender-equitable practices and environments which reduce bias. First, replace competition with cooperation. Second, measure success by goals, not by time spent in the office or online. Third, implement equitable reward structures, and provide remote and flexible work with autonomy. Finally, increase transparency in decision making.

It’s been thought that once industries achieve gender balance, bias will decrease and gender gaps will close. Sometimes called the “ add women and stir ” approach, people tend to think that having more women present is all that’s needed to promote change. But simply adding women into a workplace does not change the organizational structures and systems that benefit men more than women . Our new research (to be published in a forthcoming issue of Personnel Review ) shows gender bias is still prevalent in gender-balanced and female-dominated industries.

• Amy Diehl , PhD is chief information officer at Wilson College and a gender equity researcher and speaker. She is coauthor of Glass Walls: Shattering the Six Gender Bias Barriers Still Holding Women Back at Work (Rowman & Littlefield). Find her on LinkedIn at Amy-Diehl , Twitter @amydiehl , and visit her website at amy-diehl.com
• AS Amber L. Stephenson , PhD is an associate professor of management and director of healthcare management programs in the David D. Reh School of Business at Clarkson University. Her research focuses on the healthcare workforce, how professional identity influences attitudes and behaviors, and how women leaders experience gender bias.
• LD Leanne M. Dzubinski , PhD is acting dean of the Cook School of Intercultural Studies and associate professor of intercultural education at Biola University, and a prominent researcher on women in leadership. She is coauthor of Glass Walls: Shattering the Six Gender Bias Barriers Still Holding Women Back at Work (Rowman & Littlefield).

Partner Center

Smart. Open. Grounded. Inventive. Read our Ideas Made to Matter.

Which program is right for you?

Through intellectual rigor and experiential learning, this full-time, two-year MBA program develops leaders who make a difference in the world.

A rigorous, hands-on program that prepares adaptive problem solvers for premier finance careers.

A 12-month program focused on applying the tools of modern data science, optimization and machine learning to solve real-world business problems.

Earn your MBA and SM in engineering with this transformative two-year program.

Combine an international MBA with a deep dive into management science. A special opportunity for partner and affiliate schools only.

A doctoral program that produces outstanding scholars who are leading in their fields of research.

Bring a business perspective to your technical and quantitative expertise with a bachelor’s degree in management, business analytics, or finance.

A joint program for mid-career professionals that integrates engineering and systems thinking. Earn your master’s degree in engineering and management.

An interdisciplinary program that combines engineering, management, and design, leading to a master’s degree in engineering and management.

Executive Programs

A full-time MBA program for mid-career leaders eager to dedicate one year of discovery for a lifetime of impact.

This 20-month MBA program equips experienced executives to enhance their impact on their organizations and the world.

Non-degree programs for senior executives and high-potential managers.

A non-degree, customizable program for mid-career professionals.

How storytelling helps data-driven teams succeed

Financial services’ deliberate approach to AI

MIT report details new cybersecurity risks

Credit: Shutterstock / igor kisselev

Ideas Made to Matter

3 ways to combat gender bias in the workplace

Oct 11, 2019

Many stand to benefit when companies embrace diversity and inclusion. Women, people from different perspectives, and people of color challenge cognitive biases , prompting higher quality ideas and innovation, according to MIT Sloan senior lecturer and research scientist Renée Richardson Gosline. Women also score higher than men on 17 of the 19 most important leadership skills, according to a recent study .

While some companies focus on creating inclusive practices, women continue to battle bias as they navigate their careers. Doing so while becoming a strong leader isn’t easy, according to three business execs who shared their experiences at the recent MIT Sloan Global Women’s Symposium . What have they learned along the way? Learn to say no, get comfortable talking about uncomfortable topics, and help others coming up behind you.

Inequality in workplaces didn’t happen overnight, and won’t be diminished without effort, said Gosline, who moderated a panel discussion. “There needs to be deliberate and conscious action to make this change … this is work that each one of us has to continue to partake in,” she said.

Here are some ways to fight bias in the workplace.  

Make the unconscious conscious

Anita Carleton, EMBA ’18, a software engineering executive, said she once noticed a male colleague addressing comments, feedback, and responses to other men in the room, even if a woman had asked him a question or made a comment.

Carleton, who is now the interim director of the Software Solutions Division at the Carnegie Mellon University Software Engineering Institute, said she brought up the issue directly with her co-worker, who said he wasn’t aware of their behavior.

“When I brought up what I witnessed, he said ‘I didn’t notice I was doing that!’” she said. “It’s bringing the unconscious to the conscious.”

Making people aware of biased behavior is a good strategy, Gosline said. This often means getting comfortable having uncomfortable conversations, she added — improv classes and asking people for feedback are a couple ways to practice. It can take more than one conversation, Carleton said.

“Getting comfortable with being uncomfortable simply means that you're able to really take on these challenges and not let them derail you,” Gosline said.

Addressing issues head-on is also empowering, according to Vincenza Nigro, EMBA ’12, the vice president of global medical affairs at Hansa Biopharma.

“Permit yourself to directly address the issue and ask, ‘How do we all succeed?’” Nigro said. “How do we create the right environment for others to succeed in the organization? Because if I’m feeling this way, others sense it too, and that’s not good for anyone. So, let’s talk about it.”

Manage how others view you (and how you view yourself)

Rebecca Kirk Fair, MBA ’02, said she has employed deliberate strategies to empower herself and others, including taking a middle seat at the conference table. She does the same for junior colleagues.

“If I’m going with a younger woman to an all-male dominated client site, which is typically what happens, I put them in the middle. I make sure they have a speaking part,” said Fair, who is the managing principal at Analysis Group, a consulting company in Boston. “It’s working from the visual [point of view] and your set of expectations that you’re presenting to the world.”

It is important to have a clear vision of what you offer a company, Nigro said, and what your limits are.

Women scored higher than men on 17 of the 19 most important leadership skills, according to a recent survey.

“You have to know your skill sets and be comfortable with what you bring to the table and be the best version of yourself, not someone else’s version of you,” she said. “That’s difficult to do, but it comes with knowing yourself and knowing what you bring to the table and having confidence as a leader.”

That includes saying no. “Women often take on more in an effort to be seen by people on the executive team, but then end up overloaded,” Nigro said. “We take on more than our fair share. We try to do everything.”

Gosline said girls are often taught to be people-pleasers at an early age. Women tend to take on emotional labor for free, she said, like teaching other people about gender inclusivity. Research has also shown that women tend to be the ones to perform tasks like cleaning up after meetings. “Thinking about not just the way that people view us, but the way in which we view ourselves is really important,” she said.

Find allies, and be an ally

Friends and mentors are important for everyone, especially those trying to advance in industries where the rules of success are often unwritten. In some situations — like male colleagues talking exclusively to other men during meetings — it helps to find an ally, Fair said. In one case, she enlisted a junior male colleague to help redirect questions to her in a meeting with a client prone to addressing only men.

“It behooves all of us to figure out ways to empower the people around us, to include everyone in the discussion, and looking to your male mentors to share that authority that they have almost just for the state of their gender,” she said. “Making it discussable in your own organizations will help you deal with the broader context in the environment that we’re all interacting in. I think it’s on us to share it with women behind us, but importantly also to share it with the men around us at all levels.”

Leaders who have reached the top should take the time to help others behind them, the panelists said.  “I try very, very hard for the new leadership teams that I bring in to inspire, mentor, give them opportunities, give them challenging projects, let them succeed and fail, because you learn a lot from all of those,” Carleton said. “I think those are sort of all the major ingredients to authentic leadership.”

“Leaving behind any one of us makes all of us suffer,” Gosline said. “So hold the door open when you walk through it.”

Read next: 4 ways to be an ally for female entrepreneurs

Related Articles

Confronting Gender Bias at School

• Posted September 8, 2015
• By Bari Walsh

Following a new report that found disconcerting levels of gender bias among teen boys, teen girls, and even the mothers of teen boys and girls, HGSE’s Making Caring Common project has now released a toolkit for educators [PDF] who want to spur discussion about gender stereotypes and discrimination in their school communities.

The toolkit offers questions and activities to help middle- and high school students begin to notice and confront gender bias. These can be challenging conversations, says MCC co-director and HGSE senior lecturer Richard Weissbourd , the lead author of the report. “For some young people, this is a very personal topic that brings up questions of equality and privilege. Some of them may question whether gender biases even exist,” he says. “And the idea that biases can be implicit or unconscious is something many teenagers may never have thought of before.”

But by allowing young people to explore these ideas, educators can foster empathy and encourage (and empower) students to see themselves as problem-solvers who have an active role to play in creating a classroom culture that includes and supports every one of their peers.  

A Focus on Girls

The MCC’s report, called  Leaning Out: Teen Girls and Leadership Biases [PDF], suggested that despite the gains that women have made in professional and political life, teen girls face a powerful barrier to leadership: the biased perceptions held by their own peers. In a survey of nearly 20,000 students, the report found that 23 percent of girls and 40 percent of boys preferred male political leaders to female. Only 8 percent of girls and 4 percent of boys preferred female political leaders.

And in terms of students’ preferences for leadership on their school’s student council, students overall were least likely to support giving additional power to the student council when it was led by white girls, and most likely to support giving more power when the council was led by white boys. The report also found that mothers’ average level of support was higher for a student council led by boys than one led by girls.

Resources for Teachers

The Educator Toolkit offers a path to navigate all this by leading students through a series of 10 questions based on their reading of the Leaning Out report.

After discussion, educators can:

• If you could be a leader, what would you want to be a leader of? Why?
• What obstacles might you face and how might you overcome them?
• Have teens participate in a series of reflective writing exercises about what is it like to be a girl, or, for boys, what they think it must be like). Try the same activity asking students to reflect on what it is like to be a boy. Allow them to share their writing, if they feel comfortable.
• Challenge teens to think about how gender roles have continued to evolve over time. Invite them to interview a person of a different generation. How were women treated when they were growing up? Has society changed its expectations of women? What challenges do women still face today?

The toolkit also presents three case studies — stories about a teen girl, a teen boy, and a high school guidance counselor — that describe scenarios of bias and discrimination that are common in many high schools.

There are discussion questions about each case, along with tips for teachers on how to overcome potential obstacles to a constructive and engaging discussion with high schoolers.

Finally, the toolkit includes an easy-to-use guide to help parents and educators find high quality girls’ leadership programs. In addition to listing effective programs, it outlines the components of those programs, including:

• career exploration and opportunities to meet female leaders in various fields
• development of skills like public speaking, networking, and problem-solving
• collaborative project experiences
• mentorship and peer leadership opportunities
• opportunities for girls to work on  projects that are meaningful to them, with high expectations attached

Additional Resources

• Learn more from Usable Knowledge about the MCC’s gender bias report.
• Find more Making Caring Common resources and information about the report.
• Download [PDF] MCC’s toolkit for parents on reducing and preventing bias.

Get Usable Knowledge — Delivered Our free monthly newsletter sends you tips, tools, and ideas from research and practice leaders at the Harvard Graduate School of Education. Sign up now .

Usable Knowledge

Connecting education research to practice — with timely insights for educators, families, and communities

Related Articles

When the Glass Ceiling is a Mirror

Will Teen Girls Close the Gender Gap?

Girls as Leaders? What Do Teens Think?

Teacher gender bias is real and has lasting effects on students’ marks and study choices

Lecturer in Economics, The University of Queensland

Disclosure statement

Rigissa Megalokonomou undertook the research discussed in this article with Professor Victor Lavy of the University of Warwick and the Hebrew University of Jerusalem.

University of Queensland provides funding as a member of The Conversation AU.

View all partners

Two important patterns in education are true world-wide. First, females outperform males in most subjects, and boys do not outperform girls in high school maths and physics. Second, more females than males enrol in higher education. However, female enrolments in science, technology, mathematics and engineering (STEM) degrees are disproportionately low.

My research with Professor Victor Lavy has shown teacher gender bias at least partly explains these low enrolments. We measured this bias in an innovative way based on how teachers graded different sets of students. We tracked the effects over many years, showing this bias distorts students’ grades in school and their post-school study choices.

We also found an association with teaching quality: the most effective teachers have a gender-neutral attitude.

What did the study look at?

There is evidence that beliefs about a specific group can determine individuals’ behaviours toward members of that group. And these behaviours, whether conscious or unconscious, may affect outcomes for the individuals exposed to them. So we explored the question: if you have a pro-boy maths teacher, how does it affect students’ performance in the subject a year later and their likelihood of enrolling in a maths degree two years later?

To answer this question, we used administrative data from Greece that match students, teachers and classrooms. Our study sample included more than 400 teachers from 21 high schools over eight years. The data record the progress of students from grade 10 through to grade 12, and are linked with university admission.

Thus, we can see students’ trajectory, including results in tests in year 11, standardised high-stakes exams in grade 12, attendance, the quality of the tertiary institution they enrol in, as well as degree choices.

How was teacher bias measured?

To measure teacher gender bias we exploited the difference between two tests that every student takes in all subjects in grade 11. One test is external, graded by an external examiner, and student names and thus gender are concealed. For the other test, graded by a school teacher, student names and their gender are revealed.

These tests cover the same curriculum and examine the same skills. Both tests are high-stakes, because results count for university admission two years later.

We calculated gender differences in outcomes in the two tests for each class a teacher taught in the sample. This measure shows whether teachers do consistently give higher or lower grades when they know the genders of students (compared to the external assessors who do not know this). In this way, we could identify a teacher’s gender biases in grading.

We were able to track outcomes for teachers over the eight years to get a persistent measure of their bias in different classes with different sets of students. We found teacher gender biases exist and are persistent. A teacher who acts in one class in a pro-boy way is very likely to act in the same way in a different class even seven or eight years later.

Our findings indicate these biases are deeply rooted in teachers’ attitudes and behaviours. Only 15% of teachers were gender-neutral in their behaviour.

Many teachers favoured boys, and many teachers favoured girls, with these behaviours varying by subjects. For instance, there was more pro-boy grading behaviour by teachers in algebra rather than in history or ancient Greek.

Teacher biases affect students a lot

We then investigated the impacts of these biases on students’ maths grades in high school and on university admission. We found lasting effects. Male students who had a pro-boy maths teacher in grade 11 did better in maths in grade 12. The opposite happened to female students in their maths class – they did significantly worse the next year.

Studies from France and Israel found a similar pattern. However, these studies used a weaker definition for teacher gender biases and could not follow the same teacher over time.

Using detailed student attendance data, we also found students with teachers biased in favour of their gender are less likely to miss classes without a reason and less likely to be expelled from the class. This suggest students exposed to biased teachers might be less motivated to attend class or less engaged with learning.

After school, teacher biases continue to have a significant effect on students’ probability of enrolling in tertiary education, quality of university and study program. These effects are similar for males and females.

However, only for female students do teacher biases have a significant effect on the chosen field of study. Female students who had pro-boy teachers in maths or physics in grade 11 were less likely to enrol in university maths or physics courses two years later. Teacher gender biases seem to have little effect on male students’ degree choices.

This could be partially explained by a discouragement effect on girls that lowers their self-confidence and their beliefs in their abilities and prospects of success.

The impacts are long-term

Teacher gender biases seem to have longer-term implications for females, affecting their career prospects and earnings.

In Australia, only 35% of university degrees in STEM disciplines are awarded to women. Although 58% of students in higher education are females, the rates are much lower in STEM subjects: 40% in architecture and building, 17% in information technology and 16% in engineering and related technologies.

These STEM degrees are associated with high salaries. This means women are underrepresented in high-paying occupations. This trend is true for most OECD countries .

Gender-neutral teachers are more effective

Our final important finding is that the most effective teachers have gender-neutral attitudes. This suggests less effective teachers can harm their students twice: first by being ineffective and second by discriminating against one of the genders.

From a policy perspective, training that improves teacher quality will also likely reduce gender discrimination in schools.

• Gender bias
• Teacher bias
• Women in STEM
• maths teachers
• Gender studies

Assistant Editor - 1 year cadetship

Program Development Officer - Business Processes

Executive Dean, Faculty of Health

Lecturer/Senior Lecturer, Earth System Science (School of Science)

Sydney Horizon Educators (Identified)

• Data, AI, & Machine Learning
• Managing Technology
• Social Responsibility
• Workplace, Teams, & Culture
• AI & Machine Learning
• Diversity & Inclusion
• Big ideas Research Projects
• Artificial Intelligence and Business Strategy
• Responsible AI
• Future of the Workforce
• Future of Leadership
• All Research Projects
• AI in Action
• Most Popular
• The Truth Behind the Nursing Crisis
• Work/23: The Big Shift
• Coaching for the Future-Forward Leader
• Measuring Culture

The spring 2024 issue’s special report looks at how to take advantage of market opportunities in the digital space, and provides advice on building culture and friendships at work; maximizing the benefits of LLMs, corporate venture capital initiatives, and innovation contests; and scaling automation and digital health platform.

• Past Issues
• Upcoming Events
• Video Archive
• Me, Myself, and AI
• Three Big Points

Assignments Are Critical Tools to Achieve Workplace Gender Equity

Work assignments can be a powerful means of propelling employees’ growth but — unless managed deliberately — they can also undermine efforts to build a diverse workforce.

• Workplace, Teams, & Culture
• Organizational Behavior

Facing unprecedented levels of employee burnout and historic quit rates , how can companies lead with a model that attracts and retains talent? This period of transition, and the lessons learned from the pandemic, offer organizations a unique opportunity to improve and refine their diversity, equity, and inclusion (DEI) strategies. 1 It is imperative that leaders consider the landscape of work assignments at their companies as a foundation for greater workforce equity.

“Assignments” can comprise work tasks, activities, or projects. Scholars have long identified a gender gap in access to the kinds of assignments — large in scope, highly visible, and strategically important — that are seen as essential to career advancement. An estimated 70% of leadership development occurs through experiential learning , especially the kind offered by these challenging stretch assignments.

Get Updates on Transformative Leadership

Evidence-based resources that can help you lead your team more effectively, delivered to your inbox monthly.

Please enter a valid email address

Thank you for signing up

Privacy Policy

Yet women are largely overlooked for challenging work assignments. One factor is that women typically have fewer ties to influential decision makers who connect people to assignment opportunities . Biased performance evaluations also may play a role, with women seeing no gains in their performance scores for the very behaviors (such as “taking charge”) for which men are rewarded. 2 One study showed how promotability depends on having had challenging past projects — setting up a vicious cycle in which women never get ahead. 3 Women of color, tasked with the additional burden of “fitting in” at predominantly White organizations, may find channels to career-advancing work blocked entirely. 4

Historically, companies have not tracked assignment processes. In one 2010 report, when HR leaders were asked the percentage of “business-critical/important” assignments held by women, the top two responses were “1% to 10%” and “not measured.” Both career-advancing work and meaningful work are cornerstones of positive professional experiences. But leaders may know little about who has access to significant assignments, or they may be unaware of how a lack of access drives burnout, turnover, and dwindling diversity on the leadership bench. 5

These many unknowns about assignments drive an information gap that grows riskier as countless organizations head into new hybrid work arrangements. To quantify this risk, our team at the Stanford VMware Women’s Leadership Innovation Lab ran a study of assignments, using data that many companies collect and managers review at least yearly: employee engagement survey (EES) data. We examined pre-pandemic EES results for a midsized global technology company. 6 Only one question on the survey asked about employees’ perceptions of access to career-advancing assignments.

The company did not track assignments by gender, but our analysis showed a statistically significant gender difference. Relative to men, women were 15% less likely to report opportunities for career-advancing assignments. 7 This gender difference held even after we adjusted for employee and job characteristics. That is, women were less likely than men to perceive their assignment opportunities as having career value. This was the case even among women and men in the same department and role.

We then supplemented this analysis with a descriptive look at two related survey questions on our case company’s EES: one about making meaningful contributions, and another about receiving recognition for one’s work. Of women, 39% saw greater contribution opportunities than recognition opportunities, compared with 34% of men. While this particular gender gap may seem small, limited opportunities can accumulate over the course of a career and contribute to the persistent underrepresentation of women in leadership. Imagine how these results could inform today’s leaders in an economy recalibrating during an ongoing global pandemic.

An Equity-Minded Assignment Framework

Unseen assignment disparities can destabilize efforts to build a diverse workforce all the way up the ladder, so we propose an equity-minded assignment framework for leaders and managers to implement in the short, medium, and long terms, starting today . The purpose of this framework is to better identify and strengthen the role of work assignments in meeting DEI goals.

In the short term, embed assignment conversations in the “return to office” tools for managers. Many companies are deploying managerial tools to support employees and teams in their decision-making about hybrid work arrangements. These one-on-one meetings offer a promising context for managers to discuss assignments with their direct reports. These discussions are critical, as will be the consequences of not talking about assignments: Hybrid work arrangements, where some employees are in the office while others are working from home, run the risk of creating inequality in employees’ visibility to leaders and thus who might be seen as the right person for a particular assignment.

These conversations present a unique opportunity to explore assignments and have a forward-looking career discussion. Managers may ask, for example:

• What are you currently working on that you see as critical to your career development and advancement? What work do you find especially meaningful? How do these areas overlap?
• As we return to the office, how can we align your work with your career advancement goals and your sense of fulfillment? Which assignments do you need in order to get there, and to whom do you need to be visible?
• How will your hybrid work arrangement give you exposure to the right people and workstreams?

These questions will encourage managers and employees to think through not just the where of the hybrid workplace, but the what and with what career outcomes . Answering them can push employees to think beyond work-life factors in their ideal hybrid design — and can nudge everyone in the organization to view assignments as a core tool in employee development.

In the medium term, develop a broader view of the assignments landscape in the organization. In the wake of workforce disruption and heightened attention to racism, sexism, and inequality, leaders have been called on to accelerate their DEI efforts. To achieve real change, assignments need to be embedded in DEI strategy. The first step is to get a better handle on the baseline landscape of assignments by identifying the most important assignments for career advancement and meaning. Conducting focus groups with employees across all organizational functions can help inform the strategy by identifying common assignment-related themes and persistent problems to tackle.

Once the landscape is understood, leaders can create accountability mechanisms for more equitable assignment allocations and outcomes. Leaders need to ensure that top assignments are made available across organizational functions and that supports are in place for people to execute them successfully. For example, the former CEO of Jamba Juice, James White, changed how high-profile work was assigned by deliberately giving defined strategic projects to people who were rarely selected for them and providing them with dedicated time to meet project goals. Rethinking these channels diversified the internal pipeline of people ready to advance to leadership roles .

In performance evaluation and talent calibration meetings, leaders must explicitly account for assignments — those assigned to employees who are promoted and, just as importantly, to employees who are not. Managers should consider whether promotion gaps between women and men, for instance, would shift if assignments were changed. Internal audits and assignment dashboards, which visually clarify who on which teams is doing what, can inform data-driven managerial decision-making about assignments. The goal is not to decrease managerial autonomy but rather to empower managers with a broad view of the landscape, to increase assignment transparency and build opportunities for connection.

Finally, the range of assignments needs to be balanced fairly within units and across different roles. In mapping and building on this landscape, leaders must not overlook “low-promotability” work. Linda Babcock and colleagues have shown that women are more often asked to volunteer for lower-leverage assignments than are men, and they agree to do this work more often, too. 8 Expectations about women’s propensity to volunteer for tasks that everyone wants completed but no one wants to do themselves can route women away from career-advancing work and ultimately deepen gender stereotypes and inequity in the workplace. Leaders must engage managers, HR professionals, and staff members focused on DEI efforts in building a more equitable assignment space to support the advancement of all workforce groups.

Over the long term, make assignments a core part of your employee engagement surveys, and link the results to your talent strategy. The EES has long been a tool for organizations to take the temperature of their workforces by collecting engagement data and identifying employee needs. But work assignments are rarely measured on EESs, despite their significance for motivation, engagement, and equitable advancement. (In examining EESs at four large multinational companies in various sectors, we found that, of nearly 200 total questions, only five explicitly mentioned work assignments.) Including even a few questions about assignments will allow for new insights, and running gap analyses that integrate EES data can lead to even more significant change. (A gap analysis is a tool that allows an organization to diagnose gaps between an organizational goal and an actual outcome.)

Questions included on a survey define what information leaders can know about their workforces. “What is not measured is critically important to consider,” said Molly Anderson, CEO of Exponential Talent, a diversity and inclusion consulting firm. She also noted that “companies often draw the wrong conclusions … through an error of omission.” We suggest using one question as a starting point for study: “To what extent do you have sufficient opportunities to work on assignments that are important to your career development?” Gathering information about access to critical assignments and their connections to particular employees’ goals is a good jumping-off point that organizations can track in real time.

After you ask questions, it’s crucial to examine group differences in the responses as part of conducting a larger gap analysis into which EES data can factor directly. 9 Say, for example, that an organization sets a DEI goal to increase the representation of women and people of color in leadership roles. Collected EES data might show that these groups perceive access to leadership development assignments differently than White men do. With assignment-specific EES data, leaders can then act to meet their DEI goal, equipped with information to open dialogue, inspire interventions, and course-correct.

Assignments Looking Forward

Related articles.

The best approach to incorporating assignments in your talent strategy is multipronged. As organizations prepare for hybrid work arrangements, assignments should be discussed in managers’ one-on-ones with their direct reports; embedded in DEI goals, performance evaluations, and promotion conversations; explored in focus groups; and measured on EESs and in gap analyses. When any of these approaches reveals potential disparities in the experiences or perceptions of assignments between groups, leaders should focus on revamping their processes.

Leaders don’t have to tackle all of these approaches at once. Any increase in understanding the state of assignments in an organization, and in beginning to act on these insights, will in fact be a talent differentiator. After the pandemic-driven exodus of women — especially women of color — from the workforce, companies cannot afford to lose more of them to the additional burnout wrought by unfairly allocated assignments. By keeping steady tabs on their workforces when change is both inevitable and highly uncertain, forward-looking leaders can quickly identify and intervene in emergent negative trends and drive positive changes to empower their workforces equitably.

About the Authors

Erin Macke is a Ph.D. candidate in sociology at Stanford University and a graduate research assistant at Stanford’s VMware Women’s Leadership Innovation Lab. Gabriela Gall Rosa is a research data analyst at the VMware Women’s Leadership Innovation Lab. Shannon Gilmartin is a senior research scholar at the VMware Women’s Leadership Innovation Lab. Caroline Simard is managing director of the VMware Women’s Leadership Innovation Lab.

1. “ Hybrid Working Is Here to Stay Post-Pandemic: Stanford’s Nicholas Bloom ,” Bloomberg TV, Dec. 30, 2020, video, 6:34, www.bloomberg.com; and J.M. Barrero, N. Bloom, and S.J. Davis, “ Why Working From Home Will Stick ,” working paper 28731, National Bureau of Economic Research, Cambridge, Massachusetts, April 2021.

2. S.J. Correll, K.R. Weisshaar, A.T. Wynn, et al., “Inside the Black Box of Organizational Life: The Gendered Language of Performance Assessment,” American Sociological Review 85, no. 6 (December 2020): 1022-1050.

3. I.E. De Pater, A.E.M. van Vianen, M.N. Bechtoldt, et al., “Employees’ Challenging Job Experiences and Supervisors’ Evaluations of Promotability,” Personnel Psychology 62, no. 2 (May 2009): 297-325.

4. T.M. Melaku, “You Don’t Look Like a Lawyer: Black Women and Systemic Gendered Racism,” (Lanham, Maryland: Rowman & Littlefield Publishers, 2019).

5. P.T.Y. Preenan, I.E. De Pater, A.E. van Vianen, et al., “Managing Voluntary Turnover Through Challenging Assignments,” Group & Organization Management 36, no.3 (April 2011): 3088-344; C. Maslach and M. Leiter, “Early Predictors of Job Burnout and Engagement,” Journal of Applied Psychology 93, no. 3 (June 2008): 489-512; and J.M. Hoobler, G. Lemmon, and S.J. Wayne, “Women’s Managerial Aspirations: An Organizational Development Perspective,” Journal of Management 40, no. 3 (March 2014): 703-730.

6. This EES data was collected in 2015 from over 4,000 respondents at this company.

7. For this analysis, we calculated predicted probabilities (57% for women and 67% for men, p<0.0001) from a logistic regression in which the dependent measure, agreement with “having opportunities,” is dichotomized into levels of agreement: “great/very great” and “very little/some/moderate.” A series of ordinary least squares regressions on a nondichotomized dependent measure yielded similar results.

8. L. Babcock, M.P. Recalde, L. Vesterlund, et al., “Gender Differences in Accepting and Receiving Requests for Tasks With Low Promotability,” American Economic Review 107, no. 3 (March 2017): 714-747.

9. It is worth noting that we could not conduct our case study analyses by employees’ race and ethnicity because this information was not collected on the company’s EES, so our analyses cannot speak to both gender and race assignment inequities. While legal and privacy considerations in different geographies may constrain what can be measured, companies should strive to examine such data by race and ethnicity, geography, and other social dimensions based on their diversity strategies.

Acknowledgments

More like this, add a comment cancel reply.

You must sign in to post a comment. First time here? Sign up for a free account : Comment on articles and get access to many more articles.

• Share full article

Advertisement

This Is How Everyday Sexism Could Stop You From Getting That Promotion

By Jessica Nordell and Yaryna Serkez Oct. 14, 2021

By Jessica Nordell Graphics by Yaryna Serkez

Jessica Nordell is a science and culture journalist. Yaryna Serkez is a writer and a graphics editor for Opinion.

When the computer scientist and mathematician Lenore Blum announced her resignation from Carnegie Mellon University in 2018, the community was jolted. A distinguished professor, she’d helped found the Association for Women in Mathematics, and made seminal contributions to the field. But she said she found herself steadily marginalized from a center she’d help create — blocked from important decisions, dismissed and ignored. She explained at the time : “Subtle biases and microaggressions pile up, few of which on their own rise to the level of ‘let’s take action,’ but are insidious nonetheless.”

It’s an experience many women can relate to. But how much does everyday sexism at work matter? Most would agree that outright discrimination when it comes to hiring and advancement is a bad thing, but what about the small indignities that women experience day after day? The expectation that they be unfailingly helpful ; the golf rounds and networking opportunities they’re not invited to ; the siphoning off of credit for their work by others; unfair performance reviews that penalize them for the same behavior that’s applauded in men; the “ manterrupting ”?

When I was researching my book “The End of Bias: A Beginning” I wanted to understand the collective impact of these less visible forms of bias, but data were hard to come by. Bias doesn’t happen once or twice; it happens day after day, week after week. To explore the aggregate impact of routine gender bias over time, I teamed up with Kenny Joseph, a computer science professor at the University at Buffalo, and a graduate student there, Yuhao Du, to create a computer simulation of a workplace. We call our simulated workplace “NormCorp.” Here’s how it works.

NormCorp is a simple company. Employees do projects, either alone or in pairs. These succeed or fail, which affects a score we call “promotability.” Twice a year, employees go through performance reviews, and the top scorers at each level are promoted to the next level.

NormCorp employees are affected by the kinds of gender bias that are endemic in the workplace. Women’s successful solo projects are valued slightly less than men’s , and their successful joint projects with men accrue them less credit . They are also penalized slightly more when they fail . Occasional “stretch” projects have outsize rewards, but as in the real world, women’s potential is underrecognized compared with men’s, so they must have a greater record of past successes to be assigned these projects. A fraction of women point out the unfairness and are then penalized for the perception that they are “self-promoting.” And as the proportion of women decreases, those that are left face more stereotyping .

We simulated 10 years of promotion cycles happening at NormCorp based on these rules, and here is how women’s representation changed over time.

Simulation of Normcorp promotions over 10 years, with female performance undervalued by 3 percent

Simulation results over time

These biases have all been demonstrated across various professional fields. One working paper study of over 500,000 physician referrals showed that women surgeons receive fewer referrals after successful outcomes than male surgeons. Women economists are less likely to receive tenure the more they co-author papers with men. An analysis at a large company found that women’s, as well as minority men’s, performance was effectively “discounted” compared with that of white men.

And women are penalized for straying from “feminine” personality traits. An analysis of real-world workplace performance evaluations found that more than three-quarters of women’s critical evaluations contained negative comments about their personalities, compared with 2 percent of men’s. If a woman whose contributions are overlooked speaks up, she may be labeled a self-promoter, and consequently face further obstacles to success . She may also become less motivated and committed to the organization . The American Bar Association found that 70 percent of women lawyers of color considered leaving or had left the legal profession entirely, citing being undervalued at work and facing barriers to advancement.

Our model does not take into account women, such as Lenore Blum, who quit their jobs after experiencing an unmanageable amount of bias. But it visualizes how these penalties add up over time for women who stay, so that by the time you reach more senior levels of management, there are fewer women left to promote. These factors not only prevent women from reaching the top ranks in their company but for those who do, it also makes the career path longer and more demanding.

Small change, big difference

Even a tiny increase in the amount of gender bias could lead to dramatic underrepresentation of women in leadership roles over time..

Women’s performance is valued 3 percent less

Women’s performance is valued 5 percent less

Half as many women at level 7 and

only 2 percent of women at C-suite.

Half as many women at level 7 and only 2 percent of women at C-suite.

Women’s performance is valued 3% less

Women’s performance is valued 5% less

When we dig into the trajectory of individual people in our simulation, stories begin to emerge. With just 3 percent bias, one employee — let’s call her Jenelle — starts in an entry-level position, and makes it to the executive level, but it takes her 17 performance review cycles (eight and a half years) to get there, and she needs 208 successful projects to make it. “William” starts at the same level but he gets to executive level much faster — after only eight performance reviews and half Jenelle’s successes at the time she becomes an executive.

Our model shows how large organizational disparities can emerge from many small, even unintentional biases happening frequently over a long period of time. Laws are often designed to address large events that happen infrequently and can be easily attributed to a single actor—for example, overt sexual harassment by a manager — or “pattern and practice” problems, such as discriminatory policies. But women’s progress is hindered even without one egregious incident, or an official policy that is discriminatory.

Women’s path to success might be longer and more demanding

Career paths for employees that reached level 7 by the end of the simulation..

successful projects

“William”

started at the entry-level and reached level 7 in 4 years.

It took “Jenelle”

8.5 years to get

to the same level.

Entry level

1 year of promotions

started at the entry-

level and reached level 7 in 4 years.

8.5 years to get to the same level.

It took “Jenelle” 8.5 years to get to the same level.

Gender bias takes on different dimensions depending on other intersecting aspects of a person’s identity, such as race, religion, ethnicity, sexual orientation, disability and more. Another American Bar Association study found that white women and men of color face similar hurdles to being seen as competent, but women of color face more than either group.

Backlash, too, plays out differently for women of different racial groups, points out Erika Hall, an Emory University management professor. A survey of hundreds of women scientists she helped conduct found that Asian American women reported the highest amount of backlash for self-promotion and assertive behavior. An experimental study by the social psychologist Robert Livingston and colleagues, meanwhile, found that white women are more penalized for demonstrating dominant behavior than Black women. Our model does not account for the important variations in bias that women of different races experience.

So what’s to be done? Diversity trainings are common in companies, educational institutions and health care settings, but these may not have much effect when it comes to employees’ career advancement. The sociologists Frank Dobbin and Alexandra Kalev found that after mandatory diversity trainings, the likelihood that women and men of color became managers either stayed the same or decreased , possibly because of backlash. Some anti-bias trainings have been shown to change behavior, but any approach needs to be evaluated, as psychologist Betsy Levy Paluck has said, “on the level of rigorous testing of medical interventions.”

We also explored a paradox. Research shows that in many fields, a greater proportion of men correlates with more bias against women . At the same time, in fields or organizations where women make up the majority, men can still experience a “glass escalator,” being fast-tracked to senior leadership roles. School superintendents, who work in the women-dominated field of education but are more likely to be men, are one example. To make sense of this, we conceptualized bias at work as a combination of both organizational biases that can be influenced by organizational makeup and larger societal biases.

What we found was that if societal biases are strong compared with those in the organization, a powerful but brief intervention may have only a short-term impact. In our simulation, we tested this by introducing quotas — requiring that the majority of promotions go to women — in the context of low, moderate, or no societal bias. We made the quotas time-limited, as real world efforts to combat bias often take the form of short-term interventions.

Our quotas changed the number of women at upper levels of the corporate hierarchy in the short term, and in turn decreased the gender biases against women rising through the company ranks. But when societal biases were still a persistent force, disparities eventually returned, and the impact of the intervention was short-lived.

Quotas may not be enough

In the presence of societal biases, the effect of a short-term program of quotas disappears over time..

Societal bias has moderate effect

100% of executives

Quotas are introduced. 70% of all promotions go to women.

Majority of executives are men

YEARS OF PROMOTIONS

Societal bias has no effect

Equal representation

representation

What works? Having managers directly mentor and sponsor women improves their chance to rise. Insisting on fair, transparent and objective criteria for promotions and assignments is essential, so that decisions are not ambiguous and subjective, and goal posts aren’t shifting and unwritten. But the effect of standardizing criteria, too, can be limited, because decision-makers can always override these decisions and choose their favored candidates.

Ultimately, I found in my research for the book, the mindset of leaders plays an enormous role. Interventions make a difference, but only if leaders commit to them. One law firm I profiled achieved 50 percent women equity partners through a series of dramatic moves, from overhauling and standardizing promotion criteria, to active sponsorship of women, to a zero-tolerance policy for biased behavior. In this case, the chief executive understood that bias was blocking the company from capturing all the available talent. Leaders who believe that the elimination of bias is essential to the functioning of the organization are more likely to take the kind of active, aggressive, and long-term steps needed to root out bias wherever it may creep into decision making.

Gender Bias In Academia And Research

Introduction and definitions.

Gender bias in academia is an approach to describing and understanding gender inequalities in academia. In addition to research on gender bias, this starting point is also taken up for change. To understand the mostly English-language research literature and approaches, the basic terminology - gender, bias, and homosociability/homophily - is explained below.

The English term gender is usually translated to German as "social sex" (Wende, 2002). Since there is no corresponding term in German-speaking countries, gender is also used in German. Gender refers to the socially constructed sex of a person. As early as 1949, Simone de Beauvoir sparked the discussion about the construction of gender with her statement, "One is not born a woman, one is made one" in her work "The Other Sex" (Holland-Cunz, 2021). In 1955, psychologist John Money introduced the concept of gender roles or gender identity (Money & Ebehardt, 1955). This describes behaviours of individuals that are related to their gender identity. For Money, these behaviours are a product of, for example, education and are not related to a person's biological sex. In 1987, West and Zimmermann broadened the understanding of gender to separate sex and gender. Thus, sex is referred to as the biological sex and gender as the social sex.

Doing Gender refers to the active production and performance of gender: Gender is not a finished, final product, but is created repeatedly. For this purpose, people draw on acquired knowledge about gender-typical ways of acting and behaving and thus reproduce the respective gender. Against Money's, West's and Zimmermann's assumption of separating social and biological gender and their assumption that only social gender can be constructed, the US-American philosopher Judith Butler (1990) postulates that sex and gender are constructed. The lived gender is dependent on the biological gender and the physiological possibilities, which are connected with it. This makes it impossible for Butler to separate the two. Butler also rejects the dichotomy of gender (binarity). For her, the body represents a kind of canvas that individuals describe according to existing possibilities and through cultural practices (Holland-Cunz, 2021). Thus, non-binary gender identities (non-binary) are an integral part of queer theory.

CEWS considers the current discourse around gender and the associated conceptualizations and differentiations. The studies presented on the topic page reflect the current state of research and the fact that this research literature still predominantly distinguishes between females and males in a binary way.

Butler (1990). Gender Trouble. Feminism and the Subversion of Identity. Routledge.

Holland-Cunz (2021): Geschlecht (sex and gender) . In: Kirchhoff, Thomas (ed .): Online Encyclopedia Philosophy of Nature / Online Lexikon Naturphilosophie . doi: 10.11588/oepn.2021.2.85090

Money, Hampson & Hampson (1955). An Examination of Some Basic Sexual       Concepts: The Evidence of Human Hermaphroditism. In: Bulletin of the Johns Hopkins Hospital . Band 97, Nr. 4, 1. pp. 301–319.

Money & Eberhard (1972). Man and Woman, Boy and Girl: Gender Identity from Conception to Maturity . Johns Hopkins University Press.

Wende (2002). Gender/Geschlecht. In : Metzler Lexikon Gender Studies/Gechlechterforschung.Ansätze – Personen – Grundbegriffe . Edited by Knoll, Renate. Stuttgart: J. B. Metzler, pp. 141-142.

West & Zimmerman (1987). Doing gender. Gender & society, 1(2), pp. 125-151.

Bias refers to cognitive distortion effects that influence our perceptions, actions, and behaviour. These biases affect preconceptions, attitudes, and stereotypes in an implicit, unconscious way, which is why implicit bias is usually used synonymously with unconscious bias. Implicit bias is thus the result of mental associations we experience through direct and indirect messages. Implicit Bias can be both positively and negatively. Implicit Bias is similar to Explicit Bias, but unlike Explicit Bias, Implicit Bias does not necessarily coincide with our conscious beliefs and reflective attitudes. As an unconscious and involuntary bias, Implicit Bias, present in all people, influences decision-making behaviour and causes structural inequalities. Therefore, it is essential to understand implicit bias to reduce social inequality and discrimination. The best-known online tool for checking one's bias is the Harvard implicit association test (IAT) , which is not based on individual self-reports but uses a measurement procedure to capture social attitudes.

The connection between implicit and explicit bias is demonstrated by Mavda , among others, through relevant studies. The background of this connection is everyday experiences and socialization moments of the de facto separation of different social groups as well as the media's abundance of stereotypical representations. In addition, personal experiences, values and attitudes can reduce or increase implicit bias. While explicit bias transforms into implicit bias mainly through prejudice-reducing education, implicit bias becomes explicit bias through normalising prejudice.

Gender bias refers to systematic biasing effects shaped by gender-related stereotyping and prejudice and influence perceptions and decisions. Gender bias operates not only in everyday situations, in communication and decision-making but also in science and research, for example, concerning research design and results and personnel policy decisions. It thus shapes science and research despite supposedly objective standards of performance and evaluation and despite the ethos of independent, gender-neutral research. For example, studies demonstrate gender bias in selection and promotion processes, funding and financing opportunities, and teaching evaluations. You may find summary accounts in the publications of LERU , ECU or STI Conference ). The article " Does Gender Bias Still Affect Women in Science? " also provides a good overview of the state of research.

The papers refer to the connection between gender bias, racial bias and other biases , and the fundamental challenge of analyzing bias intersectionally , which includes considering several dimensions of inequality in their interconnection with the dimension of gender. To date, few studies perform an intersectional analysis in which they examine gender bias and racial bias in their interconnectedness. The 2012 collection of essays " Presumed Incompetent. The Intersections of Race and Class for Women in Academia " points to an increased awareness in research on bias from an intersectional perspective and an increasing number of studies on Black women in academia.

Homosociality describes the preference for people similar to us and the orientation of the members of this social group to each other. Rosabeth Moss Kanters first described this phenomenon in 1977 in the study " Men and women of the corporation ". In selection procedures at universities and research institutions, homosociality leads to hiring applicants who appear similar to the selectors in appearance, behaviour, and specific social categories such as gender and origin. This reproduces a certain type of manager and employee. This phenomenon also referred to as the mini-me effect and homophily , increases with increasing hierarchical levels and management positions and is one explanation for the glass ceiling that continues to be effective in denying women access to top positions and management levels. However, empirical studies exploring mini-me effects in application and selection processes are still largely lacking. Studies such as " Homophily in higher edcuation " refer to the phenomenon of homophily but do not investigate its exact mode of action.

Contact persons

Advertisement

Addressing Gender Bias in STEM Graduate and Post-graduate Students Using Equity in STEM for All Genders Course

• Published: 07 July 2022
• Volume 31 , pages 638–648, ( 2022 )

Cite this article

• Stephanie N. Knezz   ORCID: orcid.org/0000-0001-9445-2953 1 ,
• Evava S. Pietri 2 &
• Donald L. Gillian-Daniel 3  

1289 Accesses

5 Altmetric

Explore all metrics

Implicit gender bias is frequently cited as a contributor to the gender disparity that persists in STEM fields, despite continued efforts toward equity. While many bias interventions are aimed at faculty, scientific trainees (graduate students and post-docs) are a powerful group with the potential to enact future change. A graduate level, synchronous online course entitled, Equity in STEM for all Genders , is presented as a gender bias intervention. Course participants include graduate students, post-doctoral fellows, academic staff, and faculty. The course pairs weekly discussions (synchronous and asynchronous) about gender and gender bias-related topics with experimentally validated video interventions, primary literature, and popular articles. Over three course iterations, we observed increased bias literacy and participant motivation to mitigate gender-related bias within their local STEM contexts. We provide suggestions for making this course more widely available to STEM future faculty.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Addressing gender in STEM classrooms: The impact of gender bias on women scientists' experiences in higher education careers in Germany

Closing the Gender Gap in STEM MOOCs Through Brief, Novel Interventions

“Gendered differences versus doing gender”: a systematic review on the role of gender in CSCL

Availability of data and materials.

The datasets generated and/or analyzed during the current study are available in the Open Science Framework repository, https://osf.io/jrukm/?view_only=532767423c7c4818a301c66cef6276b8 .

Abbreviations

Cisgender woman

Cisgender man

Transgender man

Randomized controlled trials

Video Interventions for Diversity in STEM

Center for the Integration of Research Teaching and Learning

Bilimoria, D., & Stewart, A. J. (2009). “Don’t Ask, Don’t Tell”: the academic climate for lesbian, gay, bisexual, and transgender faculty in science and engineering.(Report). NWSA Journal, 21 (2), 85. https://doi.org/10.1353/nwsa.0.0077

Article   Google Scholar  

Carnes, M., Devine, P. G., Isaac, C., Manwell, L. B., Ford, C. E., Byars-Winston, A., Fine, E., & Sheridan, J. (2012). Promoting institutional change through bias literacy. Journal of Diversity in Higher Education, 5 (2), 63.

Carnes, M., Devine, P. G., Manwell, L. B., Byars-Winston, A., Fine, E., Ford, C. E., Forscher, P., Isaac, C., Kaatz, A., & Magua, W. (2015). Effect of an intervention to break the gender bias habit for faculty at one institution: A cluster randomized, controlled trial. Academic Medicine: Journal of the Association of American Medical Colleges, 90 (2), 221.

Case, K. (2007). Raising male privilege awareness and reducing sexism: An evaluation of diversity courses. Psychology of Women Quarterly, 31 (4), 426–435.

Cech, E. A. (2015). LGBT Professionals' workplace experiences in STEM-related federal agencies. 2015 American Society for Engineering Education (ASEE) National Conference, Seattle, WA.

Cech, E., & Pham, M. (2017). Queer in STEM organizations: workplace disadvantages for LGBT employees in STEM related federal agencies. Social Sciences , 6 (1), 12.  https://doi.org/10.3390/socsci6010012

Christine, W., & Agnes, W. (1997). Nepotism and sexism in peer-review. Nature, 387 (6631), 341. https://doi.org/10.1038/387341a0

Devine, P. G., Forscher, P. S., Austin, A. J., & Cox, W. T. (2012). Long-term reduction in implicit race bias: A prejudice habit-breaking intervention. Journal of Experimental Social Psychology, 48 (6), 1267–1278.

Ebert-May, D., Derting, T. L., Henkel, T. P., Maher, J. M., Momsen, J. L., Arnold, B., & Passmore, H. A. (2015). Breaking the cycle: future faculty begin teaching with learner-centered strategies after professional development. CBE - Life Sciences Education . https://doi.org/10.1187/cbe.14-12-0222

Gil, M., Naomi, I., & Markus, B. (2021). How to promote diversity and inclusion in educational settings: Behavior change, climate surveys, and effective pro-diversity initiatives. Frontiers in Education (Lausanne) . https://doi.org/10.3389/feduc.2021.668250

Ginther, D. K., Schaffer, W. T., Schnell, J., Masimore, B., Liu, F., Haak, L. L., & Kington, R. (2011). Race, ethnicity, and NIH research awards. Science, 333 (6045), 1015–1019.

Good, C., Rattan, A., & Dweck, C. S. (2012). Why do women opt out? Sense of belonging and women’s representation in mathematics. Journal of Personality and Social Psychology, 102 (4), 700–717. https://doi.org/10.1037/a0026659

Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79 (5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701

Hennes, E. P., Pietri, E. S., Moss-Racusin, C. A., Mason, K. A., Dovidio, J. F., Brescoll, V. L., Bailey, A. H., & Handelsman, J. (2018). Increasing the perceived malleability of gender bias using a modified Video Intervention for Diversity in STEM (VIDS). Group Processes & Intergroup Relations, 21 (5), 788–809. https://doi.org/10.1177/1368430218755923

Hill, C. (2010). Why so few?: women in science, technology, engineering, and mathematics . https://search.library.wisc.edu/catalog/9910090014902121 . Accessed 15 Sept 2021.

Hokanson, S., Grannan, S., Greenler, R., Gillian-Daniel, D., Iii, H., & Goldberg, B. (2019). A study of synchronous, online professional development workshops for graduate students and postdocs reveals the value of reflection and community building. Innovative Higher Education . https://doi.org/10.1007/s10755-019-9470-6

Houser, C., & Lemmons, K. (2018). Implicit bias in letters of recommendation for an undergraduate research internship. Journal of Further and Higher Education, 42 (5), 585–595.

Ian, M. H., Elizabeth, R. B., Corinne, A.M.-R., & Jessi, L. S. (2015). Quality of evidence revealing subtle gender biases in science is in the eye of the beholder. Proceedings of the National Academy of Sciences - PNAS, 112 (43), 13201–13206. https://doi.org/10.1073/pnas.1510649112

Jackson, S. M., Hillard, A. L., & Schneider, T. R. (2014). Using implicit bias training to improve attitudes toward women in STEM. Social Psychology of Education, 17 (3), 419–438.

Knobloch-Westerwick, S., Glynn, C. J., & Huge, M. (2013). The Matilda effect in science communication: An experiment on gender bias in publication quality perceptions and collaboration interest. Science Communication, 35 (5), 603–625. https://doi.org/10.1177/1075547012472684

Larivière, V., Ni, C., Gingras, Y., Cronin, B., & Sugimoto, C. R. (2013). Bibliometrics: Global gender disparities in science. Nature (London), 504 (7479), 211–213. https://doi.org/10.1038/504211a

Legault, L., Gutsell, J. N., & Inzlicht, M. (2011). Ironic effects of antiprejudice messages: How motivational interventions can reduce (but also increase) prejudice. Psychological Science, 22 (12), 1472–1477.

London, B., Rosenthal, L., Levy, S. R., & Lobel, M. (2011). The influences of perceived identity compatibility and social support on women in nontraditional fields during the college transition. Basic and Applied Social Psychology, 33 (4), 304–321.

Madera, J. M., Hebl, M. R., & Martin, R. C. (2009). Gender and letters of recommendation for academia: Agentic and communal differences. Journal of Applied Psychology, 94 (6), 1591–1599. https://doi.org/10.1037/a0016539

McDaniels, M., Pfund, C., & Barnicle, K. (2016). Creating dynamic learning communities in synchronous online courses: one approach from the Center for the Integration of Research, Teaching and Learning (CIRTL) [research mentoring; synchronous online learning; learning community; interinstitutional partnership]. 2016 , 20 (1). https://doi.org/10.24059/olj.v20i1.518

Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., Graham, M. J., & Handelsman, J. (2012). Science faculty’s subtle gender biases favor male students. Proceedings of the National Academy of Sciences of the United States of America, 109 (41), 16474. https://doi.org/10.1073/pnas.1211286109

Moss-Racusin, C. A., Pietri, E. S., Hennes, E. P., Dovidio, J. F., Brescoll, V. L., Roussos, G., & Handelsman, J. (2018). Reducing STEM gender bias with VIDS (video interventions for diversity in STEM). Journal of Experimental Psychology: Applied, 24 (2), 236–260. https://doi.org/10.1037/xap0000144

Moss-Racusin, C. A., van der Toorn, J., Dovidio, J. F., Brescoll, V. L., Graham, M. J., & Handelsman, J. (2014). Scientific diversity interventions. Science, 343 (6171), 615–616. https://doi.org/10.1126/science.1245936

Patridge, E., Barthelemy, R., & Rankin, S. R. (2013). Factors impacting the academic climate for LGBQ STEM faculty. Journal of Women and Minorities in Science and Engineering . https://doi.org/10.1615/JWomenMinorScienEng.2014007429

Pietri, E., Hennes, E., Dovidio, J., Brescoll, V., Bailey, A., Moss-Racusin, C., & Handelsman, J. (2019). Addressing unintended consequences of gender diversity interventions on women’s sense of belonging in STEM. Sex Roles, 80 (9), 527–547. https://doi.org/10.1007/s11199-018-0952-2

Pietri, E. S., Moss-Racusin, C. A., Dovidio, J. F., Guha, D., Roussos, G., Brescoll, V. L., & Handelsman, J. (2017). Using video to increase gender bias literacy toward women in science. Psychology of Women Quarterly, 41 (2), 175–196. https://doi.org/10.1177/0361684316674721

Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology, 34 (2), 243–281. https://doi.org/10.1111/j.1559-1816.2004.tb02547.x

Purdie-Vaughns, V., Steele, C., Davies, P., Ditlmann, R., & Randall Crosby, J. (2008). Social identity contingencies: How diversity cues signal threat or safety for African Americans in mainstream institutions. Journal of Personality and Social Psychology, 94 (4), 615–630. https://doi.org/10.1037/0022-3514.94.4.615

Roberson, Q. M. (2013). The Oxford handbook of diversity and work . Oxford University Press.

Book   Google Scholar  

Rudman, L. A., & Phelan, J. E. (2008). Backlash effects for disconfirming gender stereotypes in organizations. Research in Organizational Behavior, 28 , 61–79. https://doi.org/10.1016/j.riob.2008.04.003

Sheltzer, J. M., & Smith, J. C. (2014). Elite male faculty in the life sciences employ fewer women. Proceedings of the National Academy of Sciences - PNAS, 111 (28), 10107–10112. https://doi.org/10.1073/pnas.1403334111

Sheridan, J. T., Fine, E., Pribbenow, C. M., Handelsman, J., & Carnes, M. (2010). Searching for excellence & diversity: Increasing the hiring of women faculty at one academic medical center. Academic Medicine: Journal of the Association of American Medical Colleges, 85 (6), 999.

Shields, S. A., Zawadzki, M. J., & Johnson, R. N. (2011). The impact of the workshop activity for gender equity simulation in the academy (WAGES-Academic) in demonstrating cumulative effects of gender bias. Journal of Diversity in Higher Education, 4 (2), 120–129. https://doi.org/10.1037/a0022953

Swim, J. K., Aikin, K. J., Hall, W. S., & Hunter, B. A. (1995). Sexism and racism: Old-fashioned and modern prejudices. Journal of Personality and Social Psychology, 68 (2), 199–214. https://doi.org/10.1037/0022-3514.68.2.199

Teresa, M. E., Lindsay, B., Jazmin Beltran, G., Weiss, L. T., & Nathan, L. V. (2018). Evidence for a mental health crisis in graduate education. Nature Biotechnology, 36 (3), 282. https://doi.org/10.1038/nbt.4089

University-of-Wisconsin-Madison. (2008). Enhancing department climate: a guide for department chairs [Brochure] . https://wiseli.wisc.edu/wp-content/uploads/sites/662/2018/10/ClimateBrochure.pdf . Accessed 15 Sept 2021.

van Zomeren, M., Saguy, T., & Schellhaas, F. (2013). Believing in “making a difference” to collective efforts: Participative efficacy beliefs as a unique predictor of collective action. Group Processes & Intergroup Relations, 16 (5), 618–634. https://doi.org/10.1177/1368430212467476

Walton, G. M., & Cohen, G. L. (2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92 (1), 82–96. https://doi.org/10.1037/0022-3514.92.1.82

Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132 (2), 249–268. https://doi.org/10.1037/0033-2909.132.2.249

Download references

Acknowledgements

The authors would like to acknowledge Dr. Kaury Kucera, one of the creators of the course and the instructor of the course’s first offering. Additionally, Dr. Leo Taylor, instructor of 2019 and 2020 offerings of the course, was instrumental in the course development from the first to second iteration. We hope to see continued progress to report on in the future.

Author information

Authors and affiliations.

Department of Chemistry, Northwestern University, Evanston, USA

Stephanie N. Knezz

Department of Psychology and Neuroscience, University of Colorado, Boulder, USA

Evava S. Pietri

Collaborative for Advancing Learning & Teaching, Office of the Provost, University of Wisconsin-Madison, Madison, USA

Donald L. Gillian-Daniel

You can also search for this author in PubMed   Google Scholar

Contributions

SK and DGD instructed the 2018–2019 cohort of this course. SK instructed the 2019–2020 and 2020–2021 course with another instructor. EP analyzed and interpreted the survey data from students in the course. The authors read and approved the final manuscript.

Corresponding author

Correspondence to Stephanie N. Knezz .

Ethics declarations

Ethics approval.

This research has been declared as exempt by the Institutional Review Board of Northwestern University.

Conflict of Interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Knezz, S.N., Pietri, E.S. & Gillian-Daniel, D.L. Addressing Gender Bias in STEM Graduate and Post-graduate Students Using Equity in STEM for All Genders Course. J Sci Educ Technol 31 , 638–648 (2022). https://doi.org/10.1007/s10956-022-09983-y

Download citation

Accepted : 28 June 2022

Published : 07 July 2022

Issue Date : October 2022

DOI : https://doi.org/10.1007/s10956-022-09983-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Professional development
• Graduate education
• STEM education
• Gender bias
• Diversity and inclusion
• Find a journal
• Publish with us
• Track your research

Measuring Bias: Kate Zernike Shares How Exceptional Women Are Not the Exception

• Share on Facebook (opens in new window)
• Share on X (opens in new window)
• Share on LinkedIn (opens in new window)
• Print this page
• Share by email

A tape measure may seem an unlikely tool to identify unconscious bias within an organization, yet in the hands of biologist Nancy Hopkins, it quantified gender bias and became a powerful symbol of the fight for gender equality. 

At Carnegie Mellon University, during the most recent Scientists & Strategists lecture hosted by the  Carnegie Mellon Institute for Strategy and Technology (opens in new window) (CMIST) on Tuesday, April 23, guest speaker Kate Zernike recounted the story of Hopkins’ quest to unveil gender bias while working at the Massachusetts Institute of Technology’s Center for Cancer Research, later renamed the Koch Institute for Integrative Cancer Research.

A renowned journalist for the New York Times   since 2002 ,  Zernike connected with Hopkins in the spring of 1999   while working as a reporter for   the Boston Globe, eventually breaking the news of MIT’s historic admission of discrimination against women .  In 2023, Zernike published a book based on the full story of Hopkins’ journey titled “The Exceptions: Nancy Hopkins and the Fight for Women in Science.” 

The Scientists & Strategists event began with a welcome by CMIST Director  Audrey Kurth Cronin (opens in new window) , CMU’s Trustees Professor of Security and Technology, who expressed great pride in the prominent roles of women at CMU in the fields of science, engineering and humanities before turning the stage over to Provost  James H Garrett Jr (opens in new window) .

Garrett acknowledged progress since the days of the Margaret Morrison Carnegie School for Women, but stressed that fostering belonging, equity and inclusion remains a top priority at CMU as “there is always work to be done.” Before introducing Zernike, Garrett highlighted the importance of the evening’s discussion, noting that with the right environment, diversity gives rise to better work and solutions. 

The daughter and granddaughter of physicists, Zernike was surrounded by science as a child.  She was also inspired by the story of how her mother, who was initially discouraged from pursuing law, earned her Juris Doctor at the age of 45. So in March 1999, when Zernike received a “vague tip” that something was going on with women in science at MIT, along with a suggestion to contact Nancy Hopkins, she knew she needed to find out more. 

What started as a curious phone call sparked a nationwide conversation about gender bias as well as a longstanding friendship between Hopkins and Zernike.

During her talk, Zernike guided the audience through the at times infuriating, at times entertaining story of how this brilliant molecular geneticist and cancer researcher uncovered pervasive inequities at MIT by utilizing a highly developed skill: collecting and analyzing data. 

The catalyst of this groundbreaking discovery was a simple request by Hopkins in the early 1990s for 200 square feet of additional laboratory space. Hopkins, who by then was a tenured faculty member with over a decade of experience at MIT’s Center for Cancer Research, needed the additional space to house the fish tanks for her research into gene expression in zebrafish. After her request was denied, she decided to compare the size of the laboratory spaces afforded to male faculty members at MIT, including those junior to her. 

As Zernike explained, Hopkins “did not set out to be an activist '' nor a feminist. A strong believer in meritocracy, Hopkins had gone out of her way to prove that what was happening to her was not discrimination, thinking that any disparities in treatment must be attributed to deficiencies in the quality of her work. However, the data she collected told a different story.

“There is power in women working together and people working together.” — Kate Zernike

Using a measuring tape now on display at MIT, Hopkins literally got on her hands and knees to collect the measurements. She found that most male faculty members of similar rank at MIT had between 3,000 to 6,000 square feet of laboratory space as compared to her 1,500 — even junior male faculty averaged 2,000 square feet.

After confiding in a friend about her discovery, Hopkins found validation when other women faculty members shared similar stories of bias. Nearly all (16 of 17) female faculty members surveyed at MIT signed Hopkin’s report based on the collected data, which presented an evident case of bias to the university leadership. Ultimately, the president of MIT at the time, Charles M. Vest, accepted the findings of Hopkin’s report and made a public declaration of support regarding claims of gender discrimination.

Hopkins told Zernike that it took 20 years for her to realize there was gender discrimination occurring — 15 years to see it was happening to others and another five to recognize it was happening to her. The bias was not overt, but rather evident through small actions, such as being excluded from decision making, having others take credit for her ideas or work, and the spatial bias in lab allocations. 

The fact that gender bias did not present itself in the way Hopkins had expected calls attention to an ongoing challenge. Zernike pointed out that unconscious bias can happen even at a university that views itself as progressive, such as MIT. Therefore, it is vital to continue the conversation, because “things slide backwards if we don’t keep pressing forward.” 

Zernike left the audience with an unsettling insight taken from the title of her book. While all the women in the book were exceptional, they also viewed themselves as the exception — the one-off. This self-perception led them to believe that the discrepancies they encountered were a result of their individual efforts rather than the symptom of a collective issue.

Yet Hopkins’ story is a hopeful one that shows the importance of open communication and collective action to address a systemic issue. In Zernike’s words, “there is power in women working together and people working together.” 

— Related Content —

A Decade of Learning in the Heart of the Nation

CMU Researchers, Robots Head To Nation’s Capital for Robotics Showcase

Audrey Kurth Cronin To Lead Carnegie Mellon Institute for Strategy and Technology

• The Piper: Campus & Community News (opens in new window)
• Official Events Calendar (opens in new window)

• NAEYC Login
• Member Profile
• Hello Community
• Accreditation Portal
• Online Learning
• Online Store

Popular Searches:   DAP ;  Coping with COVID-19 ;  E-books ;  Anti-Bias Education ;  Online Store

Gender Identity and Expression in the Early Childhood Classroom: Influences on Development Within Sociocultural Contexts (Voices)

You are here

Thoughts on the Article | Barbara Henderson,  Voices  coeditor

Gender is an element of identity that young children are working hard to understand. It is also a topic that early childhood teachers are not always sure how best to address.

It’s not surprising, then, that Jamie Solomon’s article is the third teacher research study  Voices of Practitioners  has published that focuses directly on gender, joining articles from Daitsman (2011) and Ortiz, Ferrell, Anderson, Cain, Fluty, Sturzenbecker, & Matlock (2014). Jamie Solomon’s teacher research demonstrates how pedagogy that takes a critical stance on gender stereotyping is a social justice issue because the performance of femininity still maps directly onto disparities in opportunity within our society.

Further, she suggests how the male/female gender binary remains a default perspective and suggests how a more inclusive view of the gender spectrum can enhance and inform our practice and worldview. Her work interprets instances that arose naturally in her teaching, and it displays how teacher research is simultaneously a study of our professional and our personal selves.

During the past 10 years of teaching in the early childhood field, I have observed young children as they develop ideas about gender identity. I soon came to understand gender expression as a larger social justice issue, realizing how external influences were already at work inside the preschool classroom, impacting children's interactions and choices for play and exploration.

This matter became a great priority in my professional life, leading me to look for ways to advocate for change. Some of this eagerness stemmed from my own frustrations about gender inequity and how, as a woman, I have felt limited, misunderstood, and pressured by societal constructs. These personal experiences inspired me to help further discussions about gender development within the early childhood field so that, one day, young children might grow up feeling less encumbered by unfair social expectations and rules.

Teaching preschooI for six years at a progressive school, I was able to engage in ongoing learning opportunities, including observation and reflection. The school's emergent curriculum approach required me to pay close attention to the children's play in order to build the curriculum and create environments based on their evolving interests.

Early one semester, while on a nature field trip, I noticed great enthusiasm coming from a small group that consisted mostly of girls. They attempted to "make a campfire" using sticks and logs. After observing several other similar play scenarios and listening to their discussions, I began building a curriculum based on the children's evolving interests. I started by offering opportunities to encourage this inquiry—for example, through drawing activities and providing tools to more closely explore the properties of wood. Several weeks later, I was gratified to see that among those most deeply engaged in our emerging curricular focus on wood, fire, and camping, the majority continued to be girls. The girls' behavior and interests involved characteristics historically categorized as masculine: joyfully getting dirty, doing hard physical work (in this case with hand tools), and being motivated by a perceived sense of danger acted out in their play—for example, pretending that a fire might erupt at any moment.

These exciting observations prompted me to investigate how a particular curriculum might encourage and support children to behave outside of society's gender constructs. My understanding of gender influences built over time; each year I noticed the power and presence of these influences in the classroom.

These questions guided my study:

• How can I offer a curriculum that provides children with more opportunities for acting outside of traditional gender roles?
• How can I encourage and support children who wish to behave outside of traditional gender roles?
• How can I foster increasingly flexible thinking about gender among 4- and 5-year-old children?

The following study highlights excerpts not only from our major emergent project on camping and firemaking, but also from examples drawn from all of my teaching experiences that spring semester.

Literature review

Young children are continually making sense of their world, assimilating novel information and modifying their theories along the way. Most influences in the lives of young children—both human and environmental—reinforce existing stereotypes (Ramsey 2004).

Without prominent caring adults helping them consider perspectives that challenge the status quo, children, left to their own devices, tend to develop notions that conform with stereotypes (Ramsey 2004). If children are regularly exposed to images, actions, people, and words that counter stereotypes—for example through books, photographs, stories, and role models—they are likely to modify and expand on their narrow theories (Brill & Pepper 2008).

Thus, educators of young children should offer their student different perspectives, including those that counter society's confined constructs, to allow children access to a range of roles, expressions, and identities (Valente 2011). Without such efforts, we stymie young children's development, keeping them from realizing the extent of their potential.

During this teacher research project, I found many examples of girls crossing traditional gender role boundaries but only a few examples involving boys. Some researchers believe this phenomenon, a common finding in gender studies, results from our male-dominated culture, in which being male or having male characteristics is associated with power, opportunity, and prestige (Daitsman 2011).

Many young boys demonstrate awareness of these desirable qualities and perhaps worry about losing such advantages if they were to cross gender lines. Accordingly, educators must take an active role in providing both boys and girls counternarratives, and helping children question the status quo. Forman and Fyfe (2012) show faith in our human capacity to evolve, describing our understandings of the world as malleable. They write, "We hold that knowledge is gradually constructed by becoming each other's student, by taking an inquiry stance toward each other's constructs, and by sincere attempts to assimilate or reconcile each other's initial perspective" (247).

My goal is that this research will prompt educators to work on softening the system of gender rules that surrounds and governs our children. As Brown and Jones (2001) explain, "Changes in attitudes will not be achieved until certain fundamental dichotomies, which currently regulate aspects of classroom life, have been shifted" (143).

This study took place at a progressive San Francisco Bay Area preschool offering a full­ day, year-round program. The school serves 2 1/2 - to 5 1/2-year­ olds. I conducted the study in my classroom of twenty-one 4- and 5-year-olds.

The children were from diverse backgrounds racially, culturally, and socioeconomically and represented a wide range of family compositions. While all 21 children in my class were observed during the research process, particular children and groups of children became more visible in the data for various reasons. Some children stood out to me as particularly conforming or nonconforming to traditional gender roles, as compared to their peers. Alternatively, I also focused on cases where I felt I had witnessed a child break from their typical role or gender expression. I was the lead teacher and worked alongside and collaborated with two coteachers.

During the spring semester when this study was conducted, the children spent most of the morning hours in unstructured play time with the choice of working indoors or outdoors. We also spent at least one hour of every morning engaged in more structured activities, including circle time. The afternoons also included choices for indoor and outdoor play. Weekly field trips had long been integral to the school's program, so my class left the campus each Wednesday to embark on a local adventure together.

Beginning this study in the Spring, I benefited from having established relationships with the children over the first five months of the school year. By the time I began this teacher research, I had met with their parents during fall conferences and spent countless hours observing the children, connecting with them, learning their idiosyncrasies, and building trust. In fact, I had already come to know many of these children the year prior when preschoolers from various classrooms intermingled while playing in our shared yard.

My data sources included  field notes  and reflective notes, video and photos, and weekly journaling. The field notes generally consisted of my observations, which were recorded during natural discussions and spontaneous events. After leaving the classroom I revisited the field notes to fill in contextual holes or other missing information. Fully detailed, my field notes offered vivid samples that I could use to effectively recall experiences for analysis. I believe in many cases I reproduced conversations accurately. At other times, I captured more of the flow of an event. Excerpts from my field notes, in the upcoming Findings section, reflect this range of detail.

My analysis uses a theoretical lens suggested by Rogoff (2003), which holds that human thinking and behavior should be understood within its particular sociocultural context, that is to say an environment greatly influences those who live and learn within it and vice versa. Thus, the data is viewed in consideration of situational factors such as structured versus unstructured play, children’s varied personalities, and larger societal influences like the media. My analysis also includes self-reflection, as I continually questioned my views on gender, knowing that my data had been gathered through my personal feminist lens.

The data collected—notes and images capturing young children’s expressions, behavior, and interactions—was examined for evidence of gendered thinking and possible influences that caused it. After first organizing my data chronologically, I proceeded to go through it, jotting down one to five words to describe each data sample. Moving slowly, I regularly returned to previous samples, making comparisons between records and reevaluating the descriptions I was making. As new words or “codes” came to mind, I again returned to previous data samples to determine whether this concept was visible throughout the data. Thus, the process continued, moving forward and backward to compare, reevaluate, confirm new patterns, and then review.

Next, I studied my list of codes and pulled those that seemed most encompassing to serve as overarching themes. The three themes that resulted, in relation to gender, were (1) influences of materials and teacher expectations; (2) children’s desire and search for power; and (3) expressions and behavior illustrating children’s state of mind and development. In the following section I explore these themes, illustrating each with supporting data excerpts and my analysis of them.

Influences of materials and teacher expectations

Many factors influence children’s learning experiences in the early childhood classroom. This first theme examines how the available materials—whether closed or open-ended—might guide the children’s work and interactions with one another.

I primarily focus on the props and tools that I, the teacher, provided the children, the intention behind the materials offered, and my expectations on how they might be used. Of course other compounding factors should be considered here as well. For example, how our school’s philosophy plays out in our classroom, the physical environment, and the emergent curriculum topics we teachers have chosen. Such factors combine to create a stage upon which the children and teachers act.

Data collected on two different days revealed contrasting behavior among the children. The first excerpt focuses on two girls exploring new materials inspired by our emergent unit on wood, camping, and fire. During this play they assume less conventional female roles.

In the second sample, the subjects of my observation include three boys whom I observed handling baby dolls—props available throughout the year in our classroom—in a manner congruent with stereotypical gender norms.

Also included in this excerpt is a girl who was seeking to interact with me while I watched the boys. The first data sample stood out to me during analysis and I have included it for the reader because it caused me to consider how some curricular materials might offer children opportunities for acting outside of traditional gender roles. In contrast, the second sample made me think more deeply about the types of materials that we typically offer children (e.g., baby dolls), how many of these play props have strong associations with only one gender, and how open-ended materials might be less limiting for a child’s self-expression and learning. (See “Field Notes, February 12, 2014.”)

When the children approached the camping activity table, I gave very little instruction. Instead I explained I had seen them working with wood recently, and I wanted to give them more time and tools for their investigation. Whenever I share such observations about children’s work and express curiosity, it seems to validate their interests and encourage their exploration. The group readily experimented. The activity was approachable, open-ended, and afforded a safe place to try out new ideas, actions, and roles.

The girls appeared empowered and stayed with their work for as long as possible. Their verbal expressions resembled those I had heard more often from boys in my classroom. For instance, Caitlyn and Stella deepened their voices noticeably as they loudly delighted in each discovery, saying, “OHHH” and “WHOA!” Apparently, this natural wood paired with carpentry tools served as entry vehicles into the vigorous roles that the girls assumed.

The logs were like those they had been gathering on our field trip when they tried to make fire, while the hand tools suggested new ways to transform the wood. Something about this scenario obviously captivated them, as the girls’ interest in working with wood and dramatic play related to campfires and camping continued over the next several months.

In organizing this activity, I had expected more boys to be drawn to the wood and hand tools. On reflection, I see these expectations were based on my own gender-biased assumptions. Instead, this activity attracted more girls, providing them the opportunity to further explore an interest outside of traditional female roles. Such traditional roles are reinforced when girls role-play motherhood, princesses, or female characters commonly found in popular movies and other media—activities far more common in my classroom than these girls’ work with wood.

On a separate occasion, much later in the school year, I found myself drawn to a group of three boys working in the dramatic play area—Robby, Peter, and Mason—during unstructured play time. I noticed that they had picked up the baby dolls, and I was intrigued, as I hadn’t seen them use the dolls before. They had also brought over a roll of tape.

Perching on a nearby stepstool, I grabbed my camera, a notepad and pen, and began recording. Meanwhile, I was slightly distracted by Ella standing next to me, as she simultaneously began sharing her future plans for motherhood. (See “Field Notes, April 11, 2014”—the following dialogues are presented side by side, as they took place.)

These data samples stood out to me because of the coincidence of these two concurrent stereotypical portrayals of gender roles. While observing the group, I had perceived Ella’s dialogue as disruptive, unrelated to what I was in the process of capturing. In the moment, I was not fully focused on her thoughts and did not consider them significant to the situation. When I later reflected, however, I realized that Ella had noticed I was observing this group of boys and their rough play with the dolls. Looking to connect with me, she offered her perspective on babies and caregiving.

Upon reflection, the boys’ behavior reminded me of teacher researcher Aaron Neimark’s description of his preschool boys playing what he called “basketball babies” (2012). Through his studies, Neimark (2012) noticed how young children often use objects in silly ways that diverge from the expected or intended use—for instance, pretending that basketballs were babies—and that this sense of creativity and comedy is an important component of peer culture.

While there seemed to be an element of humor as the boys played with the baby dolls during my observation, I further wondered about possible gender-related influences that may have caused them to interact with the props in this way. Though connecting the babies to plates and flying them around was a creative idea—a divergent one from how I had expected children to use dolls—I felt that their gender role expressions guided their actions more than simple imagination. The girls in my class didn’t play with the dolls often, but when they did, their play was typically nurturing and gentle. I wondered if the boys had a tacit understanding that playing with dolls in a school setting is only acceptable if it is clearly distinct from the typical female version of such play (Brown & Jones 2001).

I find myself caught between a feminist perspective and that of the progressive teacher I sought to be: one who embraces each child’s unique interpretation of an activity or idea (Brown & Jones 2001). The gender roles that children assume, as defined by our culture, affect their play, from determining their interests to deciding how to play and how to make use of props (Meier & Henderson 2007). The data samples in this section suggest that the type of materials offered to children may provoke them to assume roles that are more or less stereotypical and could thereby influence their social interactions and learning. For instance, because baby dolls are socially constructed as feminine toys, they are less accessible for young boys.

With an understood purpose for caregiving role-play, young girls can feel comfortable behaving in line with their stereotypical gender role while playing with dolls. Boys, on the other hand, are perhaps implicitly excluded from using these toys, lest they should act outside of their traditional gender role. If they do use such materials, I have observed that their play usually deviates from the expected purpose. As a result, I find such gendered toys to be limiting for both young girls and boys. In contrast, materials that are less gendered and more open-ended—for example, natural materials such as sticks, pinecones, shells—encourage more creativity, stimulate imagination and allow for endless interpretations. Accordingly, open-ended materials are more likely to further children’s cognitive, physical and artistic development (NAEYC, n.d.).

Children’s desire and search for power

This second theme explores the human desire for control and power. I noticed that the children sought and expressed power, for example, using it to exclude or include others, to influence a situation in their favor, or to feel strong. As with the first theme, the key data samples occurred on different days. I chose examples that involved one child across two similar events: first in a position of subordination and then in a place of power. The first event took place at school and the second on a field trip.

Both events occurred during structured playtime and both observations involved a group of three children—two had already established their play when a third approached and tried to join in. As teacher researcher Chris Taaffee (2012) found, such triangulated situations often prove challenging for the third child. The excerpts from the two field notes (See “Field Notes, February 24, 2014” and “Field Notes, April 9, 2014”) demonstrate complex desires for power and how children learn approaches for exercising control.

In the field notes from February 24, Violet used her knowledge of gender constructs and her understanding of her friend Cora’s somewhat conforming gender expression to control the situation. Violet did not offer Cora any role, like a sister or mom role, other than a monster. She knew that playing the monster is a less conventional option for a girl, and thus, a choice that Cora would probably not accept. Cora seemed to be penalized here for acting within her predictable gender role, which I found thoughtprovoking, as acting within one’s gender role is frequently considered desirable and conducive to acceptance. Yet in this case, Cora’s preference to express female gender conventionally gave Violet an easy way to exclude Cora.

More than a month later, on April 9, I was fascinated to see Cora try a similar tactic with Lillian. This time, however, the interaction played out quite differently. Lillian readily seized the opportunity to become the monster, and I was pleased and surprised that Cora and Eddie were completely open to her involvement. While Violet’s intentions in the first scenario seemed clear to me, I was uncertain about Cora’s motivation. I had observed that unlike Cora, Lillian assumed nonconforming roles on a regular basis. If Cora really didn’t want Lillian to join the pair, she would have had to make a different kind of proposal.

Both scenarios demonstrate the complexity of young children’s interpersonal relationships within the sociocultural contexts influencing their lives. I and many other teachers have observed countless interactions involving a small group of children trying to protect their harmonious play from outsiders who could potentially disrupt the often fragile unity of young friendships (Neimark 2012; Taaffe 2012). I have witnessed children employ various strategies to exclude others and now realize how frequently they use their understanding of gender and culture to successfully block others from the play and determine who is permitted membership to the group (Brown & Jones 2001).

Like Cora, some children can be understood as behaving from within a dynamic process that includes learning from peers and the media, experimenting with ideas, and making sense of gender roles and relationships.

Expressions and behavior illustrating child’s state of mind and development

I have noticed that around the age of 4, children can become resolute in their thinking and uncompromising on their theories about the world, as they try to organize experiences and concepts into neat, often dichotomous categories. The following data sample typifies the kind of shortsighted perspectives children might adopt. Left unchallenged, these early views may be reinforced and become more permanent convictions. (See “Field Notes, February 25, 2014.”)

Addie has two younger brothers, one of whom is a very active 3-year-old and, according to Addie, “causes a lot of problems.” I thus attributed Addie’s concern mostly to her experiences at home. Still, I wondered about her belief that boys don’t like her. Where did this conviction come from? Teddy quickly disavowed Addie’s notion, and I noted how eager he was not to be implicated in an unfair assumption made about his gender.

In an effort to counter such gender stereotyping, my coteachers and I began implementing activities to acquaint children across genders, such as coed lunch seating arrangements and partnered projects. We also began performing childauthored plays in which crossgender roles were common (Paley [1984] 2014).

Discussion and implications

I began this study wondering how I might offer young children more opportunities to act outside of traditional gender roles. In the end, I realized that the children were working through complex ideas about the world. Our curriculum on fire and camping had encouraged some girls to step outside of gender roles, but it didn’t have a widening effect on all children—no single approach would. My findings showed that we needed a broader approach to advance children’s ideas about identity.

Accordingly, I selected the following strategies to modify my practice and undertake future teacher research:

• nurture flexible thinking across all situations
• find opportunities for children to step outside their comfort zones in regard to activities, peer relationships, and personal challenges
• foster advocacy skills in oneself and others

If people have the capacity to consider unconventional ideas and bend their thinking, our interactions with one another might look very different and be healthier for individual identity development. Furthermore, I realized that exploring and understanding gender identity shouldn’t be concentrated on the experiences of a select few, such as the girls who were so interested in the camping project. Rather, my goal should be to expand everyone’s mind, thereby making more room for children to express themselves individually across the identity spectrum.

While this research provides insight into the processes of children’s identity development, my findings are based upon one study I conducted independently over a spring semester. My feminist lens and personal perspectives influence all areas of my study—from gathering data to analyzing for interpretations, and deriving conclusions.

However, such subjectivity is inherent in teacher research and considered an advantage of the methodology, as it offers an honest insider’s perspective of a practitioner in action (Meier & Henderson 2007).

According to Meier and Henderson (2007), “Since early childhood is the foundation for young children’s views and experiences with getting along with one another, and with understanding and taking a stance toward the world of relationships, a focus in teacher research on social justice will deepen our character/social curriculum” (178). I began this research project to take action on a social justice issue, but, over the four months of this study, most of my work focused on first making sense of what I was seeing. I ended up generating more questions than answers. Yet, it was this process of questioning that helped me to deduce some useful ideas for how best to continue identity work with young children.

I hope this study encourages other early childhood teachers to question gender issues that they might have otherwise accepted at face value. Looking critically at gender can allow teachers to have broader perceptions and interpretations of daily classroom events, thereby allowing children more space as they develop their gender identities.

My data shows the complexity of this topic, including compounding factors, influences, and considerations. It also demonstrates how pervasive socialized ideas about gender roles and expression are in our lives. While my findings need to be considered within the study’s limitations, I feel that I have successfully achieved a personal goal of sharing my feminist thinking with a larger audience within the field of early childhood education.

Accordingly, this study gives voice to an important issue, and its value lies in my efforts to question the world, ease rigid thinking, and counter oppressive constructs (Valente 2011). Hopefully my teacher research “charges and challenges us to renew our commitment to an active, inclusive feminist struggle” (hooks 1994, 74).

Field Notes: Gender Identity and Expression in the Early Childhood Classroom

Field Notes | February 12, 2014

While on a field trip, a co-ed group of children worked together gathering sticks to build a fire. Several of the girls led the effort, directing others to gather more grass, sticks, and small logs. Meanwhile, the group discussed their theories about stoking a fire. Several days later, I observed many of the same children using trowels to chip away at bark while trying to “make fire” in the school garden. Thus, I decided to offer the class different types of wood, child-safe saws, and sandpaper during small group time in the classroom and see who was interested. I stayed close by to ensure that the tools were used in a safe manner. Four children, Stella, Caitlyn, Anna, and Robby, joined the activity when I invited them over, and I was pleased to see the three girls in this group so enthusiastic to use the tools and experiment with the wood.

Photos capture the children’s intensity and concentration and, thus, their interest in the activity. Stella and Caitlyn focused intently on the wood as they worked solidly for over 35 minutes and stopped only because they were asked to clean up for lunch. Before leaving the table, Stella exclaimed, “I’ve never done anything so serious!”

Field Notes | April 11, 2014

Robby (R), Peter (P) and Mason (M) gather around a small table in the dramatic play area, while I, teacher Jamie (J), watch. Mason watches with interest while Peter and Robby play with the two baby dolls, which they have brought over.

R: Rip the head off. P: No—you do it. J: Pause and think, you guys. [They all look up and over, now realizing that I’m watching.] P: We’re not actually strong enough. Shiiiiing! [P pokes a stick into the doll’s eye.] R: Watch this. [R bangs the plate on the baby and then proceeds to tape the baby to the plate. P follows his lead. The two boys fly the babies around the room, having connected them to the plastic plates, which seemed to serve as the dolls’ wings.]

Field notes | April 11, 2014

Ella (E) leans in close to me (J), ostensibly wanting to chat, as she so often does. She shares the following idea with me, while I try my hardest to focus on the group of boys. After a couple of minutes, I realize how similarly meaningful Ella’s monologue is to my study on gender.

E: I’m gonna be a mommy when I grow up. J: Oh yeah? [I raise my eyebrows, hoping that my response won’t provoke her too much, as I try to return my focus to the other children.] E: I’m gonna have one baby, because it’s hard to carry 120, 120, and 120 babies! J: [I smile]

Field Notes | February 24, 2014

Ella and Violet, 4 and 5 years old, respectively, are playing house. It’s clear that they want to maintain their harmonious two-person play, as Violet tells Cora, “I just want to play with Ella right now.”

Usually, I would have respected the wish of two children to play alone, but because Violet and Ella spend the majority of their time playing together, without the inclusion of others, I decided to push and see if they could find a way to include Cora. “Can you think of a way for Cora to play?” I ask.

Violet offers, “She can be the monster.”

Cora immediately rejects the offer; she wants to be the baby. But, according to Violet, there are no babies in this game and the only possible role is that of a monster. Cora resigns herself to finding a different playmate, and Violet and Ella continue their game, uninterrupted.

Field Notes | April 9, 2014

Cora and Eddie are playing together while on our field trip in a wooded park. They walk closely side-by-side, talking quietly, every so often looking behind. Lillian follows after them and no matter how many times they change course, she remains several feet behind them, yet not really making her intentions known. Finally Eddie bursts out, “You can’t play!” and Cora adds, “Stop following us!”

I move closer, intending to ask Cora and Eddie to tell Lillian their feelings in a kinder way. As soon as Cora sees that I’ve noticed the conflict, she quickly offers Lillian an alternative: “You can be the monster.”

Lillian smiles and begins contorting her face and body to assume the role. Cora adds, “And you can chase us!” Lillian shows them she’s ready by creeping forward just as Cora and Eddie take off in the opposite direction, screaming happily!

Field Notes | February 25, 2014

Four-year-old Addie (A) approaches me (J) and shows me a jewel she is carrying in a special container. Her classmate Teddy (T) is playing nearby. I ask about it, and Addie explains why she is keeping her jewel in the container:

A: . . . the boys might break it. [A looks down at jewel while talking] J: The boys might break it? What makes you think that? A: Because boys don’t like jewels. [A continues to look down; T looks up from work and toward A] J: Is there a reason why you think boys don’t like jewels? A: Because they don’t like me. [Looking down] T: I like you. [Said seriously and honestly]

Brill, S., & R. Pepper. 2008.  The Transgender Child: A Handbook for Families and Professionals . San Francisco, CA: Cleis Press.

Brown, T., & L. Jones. 2001.  Action Research and Postmodernism: Congruence and Critique . Conducting Educational Research series. Philadelphia, FA: Open University Press.

Daitsman, J. 2011. "Exploring Gender Identity in Early Childhood Through Story Dictation and Dramatization."  Voices of Practitioners  14.  www.naeyc.org/files/naeyc/file/Publications/VOP_Daitsman_Final.pdf .

Forman, G., & B. Fyfe. 2012. "Negotiated Learning Through Design, Documentation, and Discourse." Chap. 14 in  The Hundred Languages of Children: The Reggio Emilia Experience in Transformation , 3rd ed., eds. C. Edwards, L.Gandini, & G. Forman, 247-71. Santa Barbara, CA: Praeger.

hooks, b. 1994.  Teaching to Transgress: Education as the Practice of Freedom . New York: Routledge.

Liang, Y. 2015. "A Journey of Journals: Promoting Child-Centered Second­ Language Acquisition in Preschool."  Voices of Practitioners  10 (2): 45-58.  www.naeyc.org/files/naeyc/file/vop/VOP_Summer_2015_joumal.pdf .

Meier, D.R, &B. Henderson. 2007-  Learning From Young Children in the Classroom: The Art and Science of Teacher Research . New York: Teachers College Press.

NAEYC (National Association for the Education of Young Children). N.d. "What the Research Says: Gender­ Typed Toys."  www.naeyc.org/content/what-research-says-gender-typed-toys .

Neimark, A 2012. "'Do You Want to See Something Goofy?' Peer Culture in the Preschool Yard." In  Our Inquiry, Our Practice: Undertaking, Supporting, and Learning From Early Childhood Teacher Research(ers) , eds. G. Perry, B. Henderson, & D.R. Meier, 53-64. Washington, DC: National Association for the Education of Young Children.  www.naeyc.org/files/naeyc/file/vop/Voices_NeimarkGoofy.pdf .

Ortiz, A, D. Ferrell, J. Anderson, L. Cain, N. Fluty, S. Sturzenbecker, & T. Matlock. 2014. "Teacher Research on Boys' Literacy in One Elementary School."  Voices of Practitioners  9 (1):1-19.  www.naeyc.org/files/naeyc/file/images jvoices/9_ortiz%20v9-1 .

Paley, V.G. 2014.  Boys and Girls: Superheroes in the Doll Comer . Chicago: University of Chicago Press.

Ramsey, P.G. 2004.  Teaching and Learning in a Diverse World . 3rd ed. New York: Teachers College Press.

Rogoff, B. 2003.  The Cultural Nature of Human Development . New York: Oxford University Press.

Taaffe, C. 2012. "Two's Company, Three's a Crowd: Peer Interactions in a Preschool Social Triangle." Chap. 3 in  Our Inquiry, Our Practice: Undertaking, Supporting and Learning From Early Childhood Teacher Research(ers) , eds. G. Perry, B. Henderson, & D.R. Meier, 21-35. Washington, DC: National Association for the Education of Young Children.

Valente, J. M. 2011.  D/ Deaf and d/ Dumb: A Portrait of a Deaf Kid as a Young Superhero . Disability Studies in Education series. New York: Peter Lang.

Voices of Practitioners: Teacher Research in Early Childhood Education , the online journal of the National Association for the Education of Young Children, has been published since 2004.

Voices editor Gail Perry passed away in the summer of 2015, and her presence, expertise in teacher research, and deep knowledge of early childhood education are sorely missed. Starting with this issue, a Voices of Practitioners article will be published in each issue of Young Children as well as online.

Voices of Practitioners is a vehicle for dissemination of early childhood teachers’ systematic study of an aspect of their own classroom practice. Deeply involved in the daily lives of children and their families, teachers provide a critical insider perspective on life in their classrooms through communication of their investigations, the results, and their reflections.

Photographs: courtesy of the author

Jamie Solomon, MA, worked in the early childhood field for over 10 years, teaching at the preschool and college levels in San Francisco. Her teacher research projects have focused on gender development and emergent curriculum. Jamie has recently relocated to Southeastern Michigan.

Vol. 71, No. 3

Print this article

• Search Menu
• Advance Articles
• Editor's Choice
• Author Guidelines
• Submission Site
• Open Access
• About Journal of Public Administration Research and Theory
• About the Public Management Research Association
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Prejudice, monitoring, and process, the significance of majority opinion assignments, data, methods, and measures, conclusions, race and gender bias in three administrative contexts: impact on work assignments in state supreme courts.

• Article contents
• Figures & tables
• Supplementary Data

Robert K Christensen, John Szmer, Justin M Stritch, Race and Gender Bias in Three Administrative Contexts: Impact on Work Assignments in State Supreme Courts, Journal of Public Administration Research and Theory , Volume 22, Issue 4, October 2012, Pages 625–648, https://doi.org/10.1093/jopart/mus020

• Permissions Icon Permissions

Do certain types of administrative processes better inhibit race and gender prejudices that may surface in the public workplace? We compare the effects of three distinct administrative settings on race, gender, and other biases in the workload assignments of state supreme court justices—important public policy making settings that have been understudied in public administration. In particular, we model the extent to which majority opinion–writing assignment processes exhibit prejudice in states that use randomized assignments, rotated assignments, or fully discretionary assignments, respectively. Our findings confirm that administrative process matters. We use theories of status characteristics and administrative oversight to explain the relationship between administrative context and workload assignment patterns. Based on data from all 50 states, we discover that prejudice exists but that certain administrative processes serve better than others to suppress race and gender biases.

Our study explores whether certain types of administrative processes in the public workplace can inhibit managers from acting on personal race- and gender-based prejudices. Public administrators routinely face competing value priorities ( Kaufman 1956 ; Rosenbloom 1983 ), and these can include personal biases and self-interests ( Bendor and Moe 1985 ; Brewer 2003 ; Miller 2000 ). Scholars and practitioners therefore have an abiding interest in public servants' discretion and factors that influence the exercise thereof.

Research on street-level bureaucrats ( Lipsky 1980 ) suggests that public administrators exercise discretion in a variety of ways. 1 Public servants make choices that are other-serving, often assuming the role of citizen advocates ( Maynard-Moody and Leland 2000 ; Maynard-Moody and Musheno 2000 ), even if it means acting beyond the rules ( Keiser 1999 ; Maynard-Moody and Musheno 2003 ). Extending representative bureaucracy theory, researchers ( Bradbury and Kellough 2008 ; Meier and Bohte 2001 ; Selden 1997 ; Wilkins 2007 ) also demonstrate that public servants can use discretion to actively improve services and outcomes for citizens of their own gender, race, or ethnicity. At other times, public servants choose to be rule-adhering agents of the state ( Maynard-Moody and Musheno 2000 ) 2 or to use their discretion to advance partisan politics ( Keiser 1999 ), often ignoring client needs and preferences. In the extreme, “rogue agents” use discretion to punish, exclude, and disentitle citizen clients ( Lipsky 1980 ; Maynard-Moody and Musheno 2003 ). Rogue agents “give in to favoritism, stereotyping, and routinizing and use the rules to discourage and harass citizens” ( Wenger and Wilkins 2009 , 315). Indeed, several studies (e.g., Wenger and Wilkins 2009 ; Wilkins 2007 ) find evidence that links discretion to gender-biased citizen outcomes.

A variety of individual, organizational, institutional, and other contextual factors shape the extent and direction in which public servants might exercise their discretion (for a helpful review see Wenger and Wilkins 2009 , 315–6). Our particular interest is the extent to which administrative context and processes shape how personal race and gender biases manifest themselves in the public workplace. For example, Wenger and Wilkins (2009) recently explored the impact of automated administrative processes on agent discretion and gender bias. In their study, they found that automating the manner by which unemployment insurance claims were processed curtailed agents' ability to discriminate, as they had been, against female applicants.

This article extends such inquiry in several ways. First, we explore the impact of other types of administrative processes on prejudices in the public workplace. Our data allow us, for example, to explore other routines that include work assignments based on random-, rotation-, and discretion-based administrative processes. Second, we examine race- as well as gender-based biases. To our knowledge, scholars have not simultaneously explored these common biases as a function of varying administrative processes. Finally, we extend our analysis to important policy settings that have been understudied in public administration, state high courts. Indeed, courts continue to play an important role in checking the effects of race and gender discrimination in society from both a legal (e.g., enforcing civil rights statutes) and representative bureaucracy perspective. To the former, Hettinger, Lindquist, and Martinek note the existence of a link between race, gender, and individual judge behavior in observing that “race and gender … shape a judge’s policy goals and objectives” (2003, 223). For example, scholars suggest that racial minority and female judges may be more sympathetic to litigation involving civil rights violations than are nonminority judges ( Farhang and Wawro 2004 ).

Our article continues by offering a brief review of how race and gender biases might manifest themselves generally and in the public sector. Drawing on a theory of status characteristics as well as work on the monitoring and expression of prejudice, we hypothesize how three different administrative processes might influence race and gender biases in assigning opinion writing in state supreme courts. We then present our analyses and discuss implications for administrative practice and research.

Although some of this journal’s readership may remember when society openly supported—whether legally, economically, or customarily—gender and race discrimination, explicit biases in the post-Civil Rights era have declined ( Goldin 1985 ; Quillian 2006 ). However, scholars (e.g., Dovidio, Evans, and Tyler 1986 ; Greenwald and Banaji 1995 ; Quillian 2006 ) have since identified the persistence of implicit biases and stereotyping. Quillian explains that “even among persons who hold a sincere belief in race blindness, images and depictions of members of racial groups learned beginning in childhood are influential on their thinking. Similar forms of implicit attitudes are also at play in nonracial situations, including implicit gender biases …” (2006, 323).

Whether explicit or implicit, evidence of these biases has been documented across disciplines and in a variety of study subjects. The general public ( Huddy and Terkildsen 1993 ; Peffley, Hurwitz, and Sniderman 1997 ; Sanbonmatsu 2002 ; Sigelman et al. 1995 ), NBA referees ( Price and Wolfers 2007 ), police ( Smith and Alpert 2007 ), and public policy elites ( Haynie 2002 ; L. R. W. Mattei and F. Mattei 1998 ) employ gender and racial stereotypes.

Status characteristics scholarship ( Ridgeway et al. 2009 ; Wagner and Berger 1997 ) explains how race and gender stereotypes might implicitly shape decision-making. According to this theory, race and gender are status cues that may signal individuals to subconsciously assume that members of nonmajority classes are of different status than majority class members ( Biernat and Kobrynowicz 1997 ). These status stereotypes often result in diminished expectations of competence for minority-classed groups ( Fridkin and Kenney 2009 ; Haynie 2002 ; Sigelman et al. 1995 ). In the public arena, for example, Haynie (2002) finds that state legislators perceive African–American colleagues as less effective, and Lawless (2004) finds that individuals tend to devalue the competence of women candidates and office holders when salient issues are involved (e.g., national security). Furthermore, in studies of state and federal civil service, researchers confirm that women and racial minorities occupy disproportionately fewer leadership and decision-making positions than do white men ( Guy 1993 ; Kelly et al. 1991 ; Riccucci 2009 ).

Monitoring, Rules, and Prejudice

Scholars also find that prejudice can vary according to the degree of monitoring in a setting where attitudes about race or gender might be expressed. In general, this research demonstrates that public and semipublic settings introduce a social desirability dynamic that can alter the expression of bias. For example, assessment of women ( Streb et al. 2008 , assessing support for a female president) and blacks ( Krysan 1998 ) in “public” settings (e.g., face-to-face interviews) is far less prejudicial than in settings (e.g., self-administered survey) that allow a more private expression of attitudes. This research also suggests that the social desirability effect is strongest for more educated respondents ( Krysan 1998 ) but is otherwise invariant across demographic groups ( Streb et al. 2008 ).

Research on the role of rules and discretion is also informative here and explains the effects of monitoring from a perspective based less on social desirability and more on administrative oversight. Perhaps the best-known work in this area is the pioneering research by Davis (1969) suggesting the role of oversight as a key condition to balance administrative discretion and ensure effective adherence to rules. Authors in this area observe that rules are most effective at eliminating bias when administrative systems incorporate both constraining rules and checking rules ( Davis 1969 ; Epp 2009 ; Walker 1993 ). The former concerns those constraints that “prescribe norms, set standards, [and] define limits,” whereas the latter checks are intended to “enforce the former” ( Epp 2009 , 25) by, for example, systematically exposing to mutual and hierarchical oversight whether administrative agents are abiding by the constraining rules.

On the issues of race and gender discrimination and prejudice, we can think of fewer administrative contexts in the United States that would be more steeped in norms of equal protection than courts of law. Indeed, these norms are taught to judges as law students, codified in the universally controlling language of the Constitution, and reinforced regularly in the numerous appeals to adjudicate society’s equal protection violations. These norms are not self-enforcing, however, so checking rules, which we view as analogous to monitoring (discussed previously in this section), also play a critical role. Taken together, we are interested in how discretion, constraining rules and norms, monitoring/checking rules, and prejudice interact within and are shaped by differing administrative processes.

Linking Administrative Process, Monitoring, and Prejudice in State Supreme Courts

Public management scholars have noted the varying effects of structural context on organizational behavior (e.g., O’Toole and Meier 1999 , contrasting management in hierarchical and networked structures; Rubin and Kellough 2012 , contrasting traditional and alternative personnel structures). In the administrative workload of state supreme courts, we have the opportunity to test the varying effects of three different types of administrative contexts/processes on race and gender biases in the assignment of writing the majority opinion. More specifically, do eligible 3 female and/or black judges receive disproportionately fewer writing assignments depending on the structure of the administrative process?

We explain in greater detail, below, the significance of receiving the assignment to write the majority opinion but are reminded that “the task itself does not presume a particular structural arrangement” ( O’Toole and Meier 1999 , 510). Indeed, states have chosen markedly different ways to structure the assignment process. Twenty-two states use a procedural arrangement based on rotating the writing assignment, 13 states randomly assign the writing task, and 15 states use a procedural arrangement based on discretion.

We thus extend the work on status characteristics and bias in public administration ( Maynard-Moody and Musheno 2000 ; Wenger and Wilkins 2009 ) to three different administrative contexts. Building on the notion that organizational context can shape the expression of discretion, but noting that rules cannot eliminate such discretion ( Dias and Maynard-Moody 2007 ), we expect that even among policy elites, like state supreme court justices, race and gender attitudes may result in a disproportionate distribution of writing assignments among eligible justices. Status characteristics theory suggests that this may be the result of diminished expectations of competence 4 for black and female justices. We therefore anticipate that eligible blacks and women will receive fewer majority opinion–writing assignments.

Notwithstanding this general expectation, our previous review of the influence of social desirability, monitoring, and rules on race and gender attitudes suggests more nuanced effects. In other words, we expect race and gender biases to be strongest in those administrative contexts that are least “open” or subject to monitoring and checking rules, where social desirability effects are likely to be minimal.

H1: In states using rotation-based assignment procedures, race- and gender-bias will be statistically insignificant. Eligible female and black justices will receive as many writing assignments as eligible male and white justices.

H2: In states using random-based assignment procedures, race- and gender-bias may be statistically significant because of the difficulty of monitoring, resulting in eligible female and black justices receiving fewer writing assignments than eligible male and white justices.

H3: In states using discretion-based assignment procedures, race- and gender-bias will be detectable such that eligible female and black justices will receive slightly fewer writing assignments than eligible male and white justices.

In summary, whether bias is intentional, we hypothesize that it may be more manifest when not exposed to monitoring and social scrutiny. This is the proposed link we draw between procedural context and bias: Administrative processes that include monitoring and easy detection of deviation will be more effective in discouraging prejudice. Before introducing data and methods by which we test these hypotheses, we briefly review the importance of opinion assignments.

The assignment to write the majority’s opinion has long been recognized ( Brenner 1982 ; Murphy 1964 ; Ulmer 1970 ) as one of the major tools for shaping judicial, and consequently, public policy. Slotnick succinctly summarized the nature of this influence (1979, 60): “It is the majority decision where controlling constitutional principles are established and broader policy directives beyond the immediate case are often fashioned. Thus, the designation of the majority opinion writer has critical significance for the kinds of public policy that ultimately emerge.” Even judges have admitted that assignment is more than a matter of protocol ( Douglas 1972 ) but an instrument that can be used to advance a particular agenda or policy position ( Bonneau et al. 2007 ; Maltzman and Wahlbeck 1996 ). Moreover, opinion assignment should matter even if the other judges have the opportunity to author separate opinions or modify the assignee’s opinion. Although one might intuitively expect this type of bargaining to lead to compromises over the opinion content, thereby making the actual author less relevant, this is unlikely to happen and certainly does not minimize the importance of opinion assignment. As Lax and Cameron (2007) explicate, due to the costs of authoring persuasive alternative opinions, the assignee can still craft opinions that reflect their own preferences.

These observations are true for state supreme courts as well. Through policy articulated in majority opinions, state supreme courts have promoted “major initiatives involving, among others, school finance, the rights of defendants, and the right to privacy” ( Wise and Christensen 2005 , 582). For example, a narrow majority of the Connecticut Supreme Court issued Sheff v. O'Neill (1996, 238 Conn. 1), a landmark school desegregation case that continues to shape how schools are organized and funded in Connecticut.

In part, because a substantial amount of work has examined assignment processes at the US Supreme Court, which hears a relatively small number of cases, our chosen laboratory comprises individual state courts of last resort (hereafter state supreme courts). In 2007, state supreme courts collectively disposed of nearly 65,000 appellate cases and nearly 10,000 original cases ( National Center for State Courts 2009 ). The significance of these courts as the final arbiters of state policies is well established ( Brace and Hall 2001a ; Kagan et al. 1977 ; Tarr and Porter 1988 ; Williams and Units 1999 ). Recent scholarship ( Bonneau et al. 2007 ; Brace and Hall 2009 ; Devins 2010 ; Emmert 2009 ; Gibson 2008 ; Hall 2005 , 2009a , 2009b ; Hall and Bonneau 2006 ; Songer and Tabrizi 2009 ; Woodruff 2010 ) reaffirms not only the policy potency of state supreme courts but also the dynamics that shape the state supreme courts’ influence.

Given that state supreme court majority opinions contain important policy pronouncements, and the content of the opinion is a function of preferences of the author (e.g., Farhang and Wawro 2004 ), the assignment process is of particular importance. Surprisingly, beyond the description by Hall (1990) of the variations in the assignment procedures across states, no one has systematically examined the factors that influence this process.

In our effort to do so, we use data from State Supreme Court Data Project by Brace and Hall (2001b) . These data contain decisions from all 50 state supreme courts between 1995 and 1998. 6 Mentioned previously, 13 states use random assignment, 22 states use rotated assignments, and 15 states use discretionary assignments (see table 1 ). In addition, the database contains biographical information for all justices (over 400) who sat during this same time period. Although limited in scope of time, we see at least two advantages unique to these data. First, we are able to link judge and decision data to information about procedural context to explore the influence of states’ variation in assignment methods. Second, because we can link assignments to the race/gender of assignees, we are able to use observational data to detect the impact of race and gender attitudes that may be implicit or otherwise difficult to capture with survey data.

State Supreme Courts Assignment Types, 1995–98

After some initial analysis that focused on states using random assignments, we also collected a limited amount of qualitative data in these states by contacting each of the 13 to probe how their random assignments were administered. These data are discussed in greater detail below in our discussion of the findings.

Dependent Variable and Method

Equitable assignment.

Our data are arranged so that a justice’s participation in the case is the unit of analysis. We naturally only included those justices that voted with the majority in the cases since they are the only eligible authors of the opinion of the court. The dependent variable, case assignment, is coded 1 if a judge is assigned to write the majority opinion and 0 if the judge is not assigned. By the way of illustration, on a supreme court with seven justices voting in the majority, each case would have seven entries, with one of these cases marked “1” on the dependent variable. This explains why assignment decisions outnumber total cases in table 1 . Our preceding specification and inclusion of only eligible judges allow us to detect whether certain types of judges—based on our independent variables (see below) of sex, race, age, training, tenure, ideology, etc.—received disproportionately more or fewer writing assignments. We can therefore explore whether status cues shape the equitable distribution of writing assignments.

Because the dependent variable is dichotomous, we estimate our models with logistic regression techniques. We model each procedural context (random, rotation, and discretion) separately to assess the impact on race- or gender-motivated assignment in these settings. Given the structure of our data, autocorrelation is a particular concern. For example, with multiple observations (one for each judge in the majority) per case, if our model incorrectly predicts the assignment outcome for one judge, the error could correlate with errors in predicting other observations within the same case. Similarly, unobserved court-level factors could result in correlated error structures for observations within courts. To account for intraclass error correlation within cases, we estimated White–Huber robust standard errors clustered by case ( Wooldridge 2002 ). 7 Additionally, within each of the three models, 8 we included state-level fixed-effects dummy variables 9 to account for the intraclass error correlation within courts. 10 We also verified that our models were robust to the potential for correlated errors across study years. 11

Independent Variables

Judge characteristics.

To determine whether a judge’s race or gender influences opinion assignment, we use dichotomous variables to indicate whether a justice is (1 = black male judge ) or (1 = white female judge ). 12 Table 2 reports descriptive statistics for these and other variables discussed below.

Descriptive Statistics

Beyond race and gender, there are naturally other “status” heuristics that court staff may use to evaluate a judge’s competence to use an important policy tool–like authorship of a majority opinion. These other status characteristics include education, age, experience, and ideology.

We measure education dichotomously by indicating whether the justice attended a prestigious law school ( Elite law school = 1) as measured by Slotnick (1983) . Based on previous research ( Moore 1968 ; Slotnick 1983 ), we expect as a counter hypothesis that prestigiously trained justices will be more likely to receive the writing assignment. We measure experience continuously ( Tenure at decision = years) as the number of years between a justice’s appointment and the year of the decision. We also include a dichotomous measure for experience/status to indicate whether the justice in question is chief ( Chief judge = 1). Drawing on US Supreme Court research ( Brenner and Hagle 1996 ; Maltzman et al. 2000 ; Slotnick 1979 ), we expect that chief justices and more experienced judges will be more likely to receive the writing assignment. Although age (visible characteristic) and tenure (nonvisible characteristic) are not highly correlated, 13 we expect that older judges ( Age at decision = years old at decision) will be more likely to receive the assignment ( Slotnick 1979 ). 14 Finally, we expect that judges of moderate ideology will be more likely to receive the writing assignment than very conservative or very liberal judges ( Bonneau et al. 2007 ; Danelski 1968 ). Our measure, Ideological Extremism , is continuous. Higher values reflect ideological extremism (regardless of direction) and lower values indicate moderation. The justice’s ideology is measured using the party-adjusted surrogate judge ideology score developed and discussed in full by Brace, Langer, and Hall (2000) . The variable reflects both the ideology of the state and the partisanship of the individual justice.

Case Characteristics

We also entertain that possibility that case characteristics may condition the likelihood of assignment ( Epstein and Knight 1998 ; Maltzman and Wahlbeck 2004 ). It may be, for example, that status characteristics (whether race, gender, experience, or ideology; Unah and Hancock 2006 ) matter more when assigning an important case opinion. The primary indicator we use to measure case importance is case salience. A common measure for salience in lower courts ( Hettinger, Lindquist, and Martinek 2004 , 2006 ; Lindquist, Haire, and Songer 2007 ), including state supreme courts ( Graves and Teske 2002 ), is whether an amicus brief has been filed ( Salient case = 1). To test for possible conditional relationships, we included the salient case variable, as well as two multiplicative terms: black male judge × Salient case and White female judge × Salient case . 15

Using the preceding measures, we conducted two analyses to test the likelihood of unbiased majority opinion assignment in the three procedural contexts (rotation, random, and discretion). 16 The first analysis focuses on whether judge status characteristics have any detectable influence on the assignment process. Table 3 reports the results of this inquiry. The second analysis focuses on whether case characteristics condition assignment—specifically whether assignment bias manifests itself differently in salient cases. Table 4 reports the results of this inquiry.

Logistic Regression Model of Whether Judge Characteristics Impact Work Assignments

Note: State fixed-effects dummies not reported; errors clustered by case. Effect = predicted probability, where change in probability for continuous variables = ±SD/2, for dichotomous = 0→1.

* p < .05, one-tailed, in hypothesized direction.

+ p < .05, two-tailed, but opposite of hypothesized coefficient.

Logistic Regression Model of Whether Case Salience Moderates Work Assignments

Note: State fixed-effects dummies not reported; errors clustered by case.

Status Characteristics in Three Administrative Settings

The results in table 3 suggest that status characteristics matter differently, depending on the administrative context. In states using rotation to assign opinion writing, minority (i.e., black male and white female) judges were no more or less likely to receive the writing assignment than other eligible judges. For example, black male judges were no less likely to receive a writing assignment than white male judges (the excluded reference category). This nonfinding supports H1 about rotation-based assignment. Ease of detection and simplicity of the assignment rule appear to eliminate the possibility of race or gender prejudice in the assignment process. Beyond race and gender, contrary to expectations, age served as a negative status cue, although a substantively minimal one. With each additional 4.5 years of age (=SD/2, see Age in table 2 ), a judge is 0.4% less likely to receive the assignment. Also contrary to our expectations, chief judges were 1.4% less likely to receive an assignment in rotation-based states. As expected, ideologically extreme judges were also less likely to receive a rotation assignment—0.4% for every 6.5 points toward conservative or liberal extremes.

In states using random assignment, status characteristics appear to influence assignment. Black male judges were over 2% less likely, and white female judges were almost 1.5% less likely, to receive writing assignments than white male judges—the excluded category. This suggests the possibility that bias can surface when deviations from administrative protocols (randomized assignments) are difficult to detect and monitor. Using qualitative data, we speculate, in the discussion that follows, how these deviations might occur.

Beyond race, we find that age serves as a negative status cue; with each additional 4.5 years of age, a judge is 1% less likely to receive a writing assignment in random-based states. In these same states, ideologically extreme judges were also less likely to receive a rotation assignment—0.7% with every 6.5 points toward the spectrum’s extremes. Although elite training did not serve as a cue in rotation states, it did in random-based states. As predicted, judges trained at elite law schools were 2.4% more likely to receive the assignment than nonelite alumni.

In states where justices have discretion in making the writing assignment, every status characteristic in our model influences the decision. We specifically found some evidence to support H3 and some to contradict. In support of H3, eligible black male judges were almost 4% less likely to be assigned a case than eligible white male judges. On the other hand, white female judges received a disproportionate number of assignments, as expected, but contrary to expectations in H3, female judges were 2.4% more likely to be assigned a case. Although we can only speculate, our findings potentially provide evidence for other research suggesting that (1) race stereotypes are activated before gender stereotypes ( Ito and Urland 2003 ), which may explain their primacy in discretionary opinion assignment or (2) gender’s effects are shaped by the courts’-specific cultural context ( Turco 2010 ), which may be more sensitive to gender than race equity in discretionary assignment. Engaging these possibilities are tasks for future research. 17

Beyond race and gender, chief justices were more likely to be assigned (2.8%), as were older judges (1.3 for every 4.5 years) and judges that went to elite law schools (3.7%). Ideologically extreme judges were less likely to receive an assignment (1% for every one half of a standard deviation away from the mean), as were judges with advanced tenure (1.2% for every 3.3 years of tenure beyond the mean).

The Conditioning Effects of Case Salience

We were also interested to see whether case importance (i.e., salience) conditions the assignment process in the three different procedural settings. This allows us to extend our understanding of prejudice to not only routine decisions but also to significant decisions that are likely to have the most noticeable administrative and policy impact.

Table 4 demonstrates that for rotation-based states, case salience has no interactive effects. In other words, black males and white women were no more or less likely to receive an assignment in either salient or nonsalient cases. The same is not true, however, for assignments in “random” states and states using discretion. We found a significant interaction between case salience, race, and gender in the former and case salience and gender in the latter. In random states, black male judges and white female judges were less likely than white male judges to receive a salient opinion assignment. 18 In discretion states, although white women were slightly more likely to receive the average writing assignment ( table 3 ), white women were actually less likely to receive the assignment when the case was important/salient.

To better understand the substantive importance of these findings, we calculated the conditional effects of assignment—the differences in the probability the judge would be assigned the opinion—in salient versus nonsalient cases. 19 Because changes in predicted probabilities of rare events belie their substantive importance, we also calculated percentage changes in predicted probabilities. Table 5 reports these results and we discuss case salience’s effects in states using random assignment and then in discretion-based states.

Postestimation Results of Probability of Assignment When the Case Is Salient

Note: % Change: The percent difference between the probability of the assignment of minority (black male; white female) judge and a white male judge. This is derived as follows: 100 × (Probability of assignment to the minority judge − Probability of assignment to the white judge)/(Probability of assignment to the white male judge).

+ p < 0.05, two-tailed, but opposite of hypothesized coefficient.

Salience in States Using Random Assignment

Echoing previous findings that partially supported H2 black judges are less likely (6%) to receive the majority opinion–writing assignment in nonsalient cases. However, the use of status cues is especially prolific in important cases. In salient cases, black male judges are 43% less likely, than white male judges, to receive the majority opinion–writing assignment. We view this as fairly strong and substantively meaningful evidence that case characteristics can condition the assignment process, even in random administrative contexts. The direction of this moderation suggests that white male judges, in accordance with the theory of status characteristics, are favored to write the majority opinion in higher profile cases. Salience also conditions the use of gender cues in random states, where white women are 8% less likely to receive a nonsalient assignment.

We interpret these interactions between salience, race, and gender as further support for H2. Even in settings where rules dictate random assignment, racial and gender biases can occur. Furthermore, race-biased assignments surface most prolifically in salient cases when a premium may be placed on competence and where it seems race-based cues are used to identify that competence via status characteristics theory.

Salience in States Using Discretionary Assignment

Assignments in discretion-based settings also suggest that case salience matters to the process by which judges use status cues. In state courts using discretion-based assignments, black male judges were 15% less likely than white male judges to receive the writing assignment in the most common types of cases (i.e., nonsalient cases). In salient cases, there is also a difference in assignment probability in the hypothesized direction (blacks receiving fewer assignments), but the difference does not meet our chosen significance threshold.

Salience also conditions gender cues in discretion-based assignments. In the vast majority of cases (nonsalient), women were 15% more likely to receive the assignment than white males. However, in salient cases, women received fewer assignments as we hypothesized, but again, the difference does not meet our chosen significance threshold. As we discussed previously, one possible explanation of salience’s conditioning of gender cues is that the ability to openly monitor the assignor may encourage a socially desirable over assignment of opinion writing to female judges. However, what is remarkable is that this pattern disappears when the case is important.

Together, the conditioned race and gender findings suggest a deeper more nuanced support for H1–H3 than we had discovered before considering case salience. 20 Unlike the random- and discretion-based process, judges in states using rotation-based assignment were neither influenced by race nor by gender cues. This finding held true even when we considered case salience. Race and gender did matter, however, in states where monitoring/checking was difficult (random and discretion states) and where social desirability potentially shaped assignment (discretion states). In addition, the influence of race and gender in these states was both tempered and concentrated by case salience.

We conclude this section with a reminder about our analysis and interpretation. One could argue that the observed tendency of random and discretionary assignment processes to result in fewer assignments to black males and white females to be an artifact of the possible tendency for these judges to disproportionately dissent from the majority coalition, thus limiting their opportunity to receive the opinion assignment. Although there is mixed evidence that this may be the case in other courts ( Hettinger et al. 2004 ), we explicitly control for this possibility by limiting our estimation sample to only those judges who are already in the majority and thus eligible for assignment. 21

Taken together, these findings are important for at least three reasons. First, we provide evidence from a relatively understudied area (state supreme courts) that status characteristics influence important administrative decisions even among policy elites like state supreme court justices. We further note that these important policy makers are quite often the very ones trusted to decide, define, and interpret the Constitution’s protections of equitable treatment. Depending on the administrative setting—rotation, random, discretion—we found evidence of bias based not only on race and gender but also on other cues such as age, ideology, position, tenure, and elite training.

Second, the importance of the administrative decision, which we measured using case salience, influenced the extent of prejudicial decision-making. As theory led us to expect, higher stakes writing assignments negatively prejudiced black and female judges. In random-based states, black male judges received disproportionately fewer important writing assignments. In discretion-based states, white female judges received disproportionately more routine (nonsalient) writing assignments but did not receive disproportionately more important (salient) writing assignments.

Finally, and most importantly, we extend earlier evidence to suggest that the use of status cues varies as a function of administrative context. Just as an automated-based administrative context curbed gender bias in unemployment insurance claim processing ( Wenger and Wilkins 2009 ), we found that the rotation-based administrative context was most effective in promoting race and gender equality in opinion-writing assignments—even in salient cases.

Why might this be? Presumably, recognizing the policy importance of the decision to assign the majority opinion, the assigner potentially acts like any other street-level bureaucrat. When they are monitored effectively, their discretion is limited by the administrative context. For rotation states, we surmise that the simplicity of the rotation-based rule, combined with the ability to easily detect deviations, creates an administrative context that minimizes the likelihood of shirking and/or prejudice. Should the assigner not follow the rotation pattern, the prejudice is obvious, and sanctions could then be applied. This is consistent with our findings. In rotation states, black men and white women are just as likely to receive the majority opinion assignment as white males. Indeed, none of the measures of competency in rotation states have the posited effects in rotation states. Additionally, the only discretionary hypothesis that received any support—the tendency not to assign the opinion to ideological extremists—had a small substantive effect. Applying the theory by Davis (1969) of oversight and discretion, rotation-based administrative contexts appear to effectively combine strong constraint and checking rules: Race and gender prejudice is precluded by a simple rotation (constraint) rule that is enforceable by a relatively straightforward and strong checking/monitoring process (see table 6 ).

Administrative Assignment Contexts and Rules: Theory and Results

By negative bias, we mean that the minority in question is underassigned (e.g., black judges receive disproportionately fewer writing assignments than white judges); by positive bias we mean the minority in question is overassigned.

Random-based and rotation-based assignment states share a common strong constraint rule: gender- and race-neutral assignment, whether through rotated or randomized selection (see table 6 ). For this reason, we might intuitively expect the same race- and gender-neutral assignment patterns over time that we saw in the rotation-based states. However, in our theory development, we have raised the possibility that, unlike contexts using rotation methods, shirking and/or prejudicial behavior is less apparent in a random system. The absence of effective monitoring or a checking rule, combined with the policy importance of the assignment decision, particularly in salient cases, creates a context in which judges as street-level bureaucrats theoretically have more discretion adhering to randomized processes. Empirically, we found evidence that this occurs. Judges in random assignment contexts appear to utilize three competency heuristics in the manner we posited: the prestige of the judge’s law school alma mater, as well as race and gender characteristics.

The evidence, therefore, demonstrates that the presence of constraint rule structures (e.g., automated or randomized processes) alone may not be sufficient to curb race and gender biases. Our finding confirms those of studies in other areas ( Davis 1969 ; Epp 2009 ; Walker 1993 ). Checking rules or the ability to monitor and detect deviations is critical—a lesson we draw from both our quantitative analysis and our attempt to qualitatively explore how deviations might be introduced.

Prior to each session of the Court, the assignment of the cases is determined by lot. Seven slips of paper are prepared by the Clerk of the Court with one slip bearing the number “1” and the remainder being blank. The slips are placed in a hat and the member of the Court (or a proxy) drawing the marked paper is assigned the first case on the docket. The remaining cases are assigned to the justices in descending order of seniority until all cases are assigned.

Only one of our respondents, Washington, suggested the possibility of deviation from their practice of “double-blind assignment in the conference room in the presence of the chamber’s law clerks.” For certain types of cases (death penalty, bar and judicial discipline, and appeals as to the recall elected officials), Washington assigns by rotation. We reran our models with and without Washington and found no meaningful differences from our original models. For state courts of last resort, we can identify neither specific cheating mechanisms nor alternative explanations for race bias in random-based states beyond the theory of status characteristics and monitoring that we have developed here. This is certainly one of the limitations of this study.

Of course, from a Weberian perspective, there are potentially great theoretical benefits in assigning a complex and important task to the most competent person—regardless of the administrative context. The normative problem occurs, however, when the cues (like race and gender) used to determine competency are faulty and/or morally unacceptable. This raises a second limitation of our study. We leave to others the task of exploring whether decisions written by black male and white female judges are less “competent.” In our own sample of opinions issued in state courts of last resort, we cannot conceive of a way to empirically test this assertion using typical methods (e.g., whether the higher court overturns the decision). However, we do note some evidence suggesting that minority judges are more sympathetic to certain types of cases ( Farhang and Wawro 2004 ) and that a judges’ policy goals may be influenced by their race and gender ( Hettinger et al. 2003 ). Neither of these observations, however, suggest a guiding rule for the use of race or gender in majority opinion assignment.

This brings us to our final administrative context. In states using discretion-based assignment, processes assignors are allowed significant latitude. Indeed, in these states alone were all the status cues, for which we had data, significant and in the posited direction. In other words, without strong constraint and checking rules, assignors use a broad range of status cues to make their decisions on a case-by-case basis (see table 6 ). Again, although there may be practical benefits to this system, we found empirical evidence that the judges assigning the opinions were influenced by race and gender biases.

Although discretionary assignment revealed a pattern of systematically bypassing black judges, it also revealed, contrary to theory, systematically favoring female judges. However, when we accounted for case salience, the findings went in the direction hypothesized. In other words, female status was a stronger negative cue in important cases than it was a positive cue in nonsalient cases. Perhaps, this bidirectional behavior with respect to female judges can be explained by assignors’ desire to generally behave according to societal norms (even favoring female assignments)—at least when cases are nonsalient. Or perhaps, other stereotypes beyond competency are triggered by gender. For example, the opinion assigner may subconsciously believe women will tend to complete the tasks faster and are therefore more likely to assign the task to women in nonsalient cases.

Our study is not without limitations. Our findings with respect to gender and race biases generally follow the direction status characteristics theory suggests, but we cannot determine the extent to which they are generalizable to other court, agency, or cultural contexts. Furthermore, our findings are admittedly based on a unique, but older, data set. Although status characteristic theory continues to be relevant, we were unable to test how race and gender attitudes may have changed over the last decade on the state supreme courts.

To this end, we see as useful future research that would explore current attitudes relative to the intersectionality of status characteristics (e.g., Collins and Moyer 2008 ). More broadly, we hope that future work will continue to explore the impact that additional administrative contexts have on fostering decision-making that is unprejudiced by status cues.

Our findings also raise several policy and management implications. First, even among policy elites trained in the law, bias is evident across a range of status characteristics. Previous studies have separately confirmed the link between agent discretion and racial/ethnic outcomes (e.g., Hindera 1993 ; Meier, Stewart, and England 1989 ; Pitts 2007 ) and between agent discretion and gender outcomes (e.g., Keiser et al. 2002 ; Meier and Nicholson-Crotty 2006 ; Wenger and Wilkins 2009 ; Wilkins 2007 ). Here, we offer positive evidence that bias shapes decision-making in an integrated model of gender, race, and other human capital characteristics across three administrative different settings.

Second, administrative context and process shapes behavior within public organizations ( O’Toole and Meier 1999 ; Rubin and Kellough 2012 ). Our findings here suggest that context influences the use of status characteristics to make administrative decisions. We perceive the link between administrative context and prejudicial behavior to be, at least in part, a function of constraining and checking rules.

Third, the constraints of the administrative context alone may be insufficient to eliminate bias ( Dias and Maynard-Moody 2007 ). Effectively constraining prejudicial behavior appears to also require effective monitoring. Prejudice based on status characteristics can be pervasive enough to manifest itself despite constraint rules working to the contrary. For example, our findings call into question the effectiveness of randomized administrative controls as a means to curb race and gender biases where the effectiveness of such controls in discouraging discretion is important for managing a broad variety public policy programs and concerns. These potentially include—and should be tested in future research—airport security screenings, vehicle safety, financial accounting, border security, workforce safety (e.g., drug screening), census collection, and policy program experiments.

In addition to administrative context, we are reminded that promoting a positive organizational culture of inclusiveness and equity through other means is also important to discourage prejudices ( Cox and Blake 1991 ; Riccucci 2002 ) that may be shaped by, but poorly supported, status cues.

In short, public administrators should be cautious about relying too heavily on oversight mechanisms and procedures that are not easily monitored for deviation. Returning to Davis’s distinction between types of rules (1969), our findings suggest that constraining norms discouraging judges’ personal race and gender biases may not be enough without checking rules to ensure that compliance is open to monitoring. At least in the opinion-assignment context, rules so designed can virtually eliminate race and gender biases. Louis Brandeis’s memorable observation that sunshine is an effective disinfectant is certainly appropriate here but with an important addition. The disinfecting power of sunshine is more powerful when its energy is focused by effective rules that both constrain and monitor. 22

We are indebted to three anonymous reviewers for their careful and constructive criticism and for their meaningful suggestions and encouragement. Their feedback helped us to improve this manuscript in many ways. Any shortcomings are, of course, our own responsibility.

Baum Lawrence The Supreme Court 1985 Washington, DC : CQ Press

Google Scholar

Google Preview

——— The puzzle of judicial behavior 1997 Ann Arbor: University of Michigan Press

Bendor Jonathan Moe Terry M An adaptive model of bureaucratic politics American Political Science Review 1985 79 755 74

Biernat Monica Diane Kobrynowicz Gender-and race-based standards of competence: Lower minimum standards but higher ability standards for devalued groups Journal of Personality and Social Psychology 1997 72 544 57

Bonneau Chris W Hammond Thomas H Forrest Maltzman Wahlbeck Paul J Agenda control, the median justice, and the majority opinion on the US Supreme Court American Journal of Political Science 2007 51 890 905

Brace Paul Melinda Gann Hall “Haves” versus “have nots” in state supreme courts: Allocating docket space and wins in power asymmetric cases Law & Society Review 2001a 35 393 417

———. (Producer) State Supreme Court Data Project 2001b http://www.ruf.rice.edu/∼pbrace/statecourt/

——— Neo-institutionalism and dissent in state supreme courts Journal of Politics 2009 52 54 70

Brace Paul Laura Langer Hall Melinda G Measuring the preferences of state Supreme Court justices Journal of Politics 2000 62 387 413

Bradbury Mark D Edward Kellough J Representative bureaucracy: Exploring the potential for active representation in local government Journal of Public Administration Research and Theory 2008 18 697 714

Brambor Thomas Clark William R Matt Golder Understanding interaction models: Improving empirical analyses Political Analysis 1982 14 63 82

Brenner Saul Strategic choice and opinion assignment on the US Supreme Court: A reexamination Political Research Quarterly 1982 35 204

Brenner Saul Hagle Timothy M Opinion writing and acclimation effect Political Behavior 1996 18 235 61

Brewer Gene Jack Rabin Bureaucratic politics Encyclopedia of public administration and public policy 2003 New York Marcel Dekker 141 46

Collins Todd Moyer Laura Gender, race, and intersectionality on the federal appellate bench Political Research Quarterly 2008 61 219 27

Cox Taylor H Blake Stacy Managing cultural diversity: Implications for organizational competitiveness The Executive 1991 5 45 56

Danelski David J Jahnige TP Goldman S The influence of the chief justice in the decisional process of the Supreme Court The federal judicial system 1968 New York Holt, Rinehart, and Winston 147 60

Davis Kenneth Culp Discretionary justice: A preliminary inquiry 1969 Baton Rouge Louisiana State University

Devins Neal How state supreme courts take consequences into account: Toward a state-centered understanding of state constitutionalism Stanford Law Review 2010 62 1629 711

Dias Janice Johnson Maynard-Moody Steven For-profit welfare: Contracts, conflicts, and the performance paradox Journal of Public Administration Research and Theory 2007 17 189

Douglas William O Memorandum 1972 Available in Justice William O. Douglas's papers . Unpublished manuscript. Washington, DC

Dovidio John F Evans Nancy Tyler Richard B Racial stereotypes: The contents of their cognitive representations Journal of Experimental Social Psychology 1986 22 22 37

Emmert Craig F An integrated case-related model of judicial decision making: Explaining state supreme court decisions in judicial review cases Journal of Politics 2009 54 543 52

Epp Charles R Making rights real: activists, bureaucrats, and the creation of the legalistic state 2009 Chicago, IL: University of Chicago Press

Epstein Lee Knight Jack The choices justices make 1998 Washington, DC CQ Press

Farhang Sean Wawro Gregory Institutional dynamics on the US court of appeals: Minority representation under panel decision making Journal of Law, Economics, and Organization 2004 20 299

Fridkin Kim L Kenney Patrick J The role of gender stereotypes in US senate campaigns Politics and Gender 2009 5 301 24

Gibson James L Challenges to the impartiality of state supreme courts: Legitimacy theory and “New Style” judicial campaigns American Political Science Review 2008 102 59 75

Goldin Claudi Understanding the gender gap New Perspectives 1985 17 9 13

Goodwin Barbara Justice by lottery 1992 Chicago: University of Chicago Press

Graves Scott Teske Paul State supreme courts and judicial review of regulation Albany Law Review 2002 66 857

Greenwald Anthony G Banaji Mahzarin R Implicit social cognition: Attitudes, self-esteem, and stereotypes Psychological Review 1995 102 4 27

Gujarati Damodar Porter Dawn C Basic econometrics. 2009 New York, NY: McGraw-Hill Irwin.

Guy Mary E Three steps forward, two steps backward: The status of women's integration into public management Public Administration Review 1993 53 285 92

Hall Melinda Gann Opinion Assignment Procedures and Conference Practices in State Supreme Courts Judicature 1990 73 209 214

——— State supreme courts in American democracy: Probing the myths of judicial reform American Political Science Review 2005 95 315 30

——— Constituent influence in state supreme courts: Conceptual notes and a case study Journal of Politics 2009a 49 1117 24

——— Electoral politics and strategic voting in state supreme courts Journal of Politics 2009b 54 427 46

Hall Melinda Gann Bonneau Chris W Does quality matter? Challengers in state supreme court elections American Journal of Political Science 2006 50 20 33

Haynie Kerry L The color of their skin or the content of their behavior? Race and perceptions of African American legislators Legislative Studies Quarterly 2002 27 295 314

Hettinger Virginia A Lindquist Stefanie A Martinek Wendy L Separate opinion writing on the United States courts of appeals American Politics Research 2003 31 215 50

——— Comparing attitudinal and strategic accounts of dissenting behavior on the U.S. courts of appeals American Journal of Political Science 2004 48 126 37

——— Judging on a collegial court: Influences on federal appellate decision-making 2006 Charlottesville: University of Virginia Press

Hindera John J Representative bureaucracy: Imprimis evidence of active representation in the EEOC district offices Social Science Quarterly 1993 74 95 108

Huddy Leonie Terkildsen Nayda The consequences of gender stereotypes for women candidates at different levels and types of office Political Research Quarterly 1993 46 503

Ito Tiffany A Urland Geoffrey R Race and gender on the brain: Electrocortical measures of attention to the race and gender of multiply categorizable individuals Journal of Personality and Social Psychology 2003 85 616

Kagan Robert A Cartwright Bliss Friedman Lawrence M Wheeler Stanton The business of state supreme courts, 1870–1970 Stanford Law Review 1977 30 121 56

Kaufman Herbert Emerging conflicts in the doctrines of public-administration American Political Science Review 1956 50 1057 73

Keiser Lael R State bureaucratic discretion and the administration of social welfare programs: The case of social security disability Journal of Public Administration Research and Theory 1999 9 87

Keiser Lael R Wilkins Vicky M Meier Kenneth J Holland Catherine A Lipstick and logarithms: Gender, institutional context, and representative bureaucracy American Political Science Review 2002 96 553 64

Kelly Rita Mae Guy Mary E Bayes Jane Duerst-Lahti Georgia Duke Lois L Hale Mary M et al.  Public managers in the states: A comparison of career advancement by sex Public Administration Review 1991 51 402 12

Krysan Maria Privacy and the expression of white racial attitudes: A comparison across three contexts Public Opinion Quarterly 1998 62 506

Lawless Jennifer L Women, war, and winning elections: Gender stereotyping in the post September 11th era Political Research Quarterly 2004 53 479 90

Lax Jeffrey R Cameron Charles M Bargaining and opinion assignment on the US Supreme Court Journal of Law, Economics, and Organization 2007 23 276

Lindquist Stefanie A Haire Susan B Songer Donald R Supreme Court auditing of the US courts of appeals: An organizational perspective Journal of Public Administration Research and Theory 2007 17 607 24

Lipsky Michael Street-level bureaucracy: Dilemmas of the individual in public services 1980 New York, NY Russell Sage Foundation

Long J. Scott Freese Jeremy Regression models for categorical dependent variables using Stata 2001 College Station, TX Stata Press

Maltzman Forrest Spriggs James F Wahlbeck Paul J Crafting law on the Supreme Court: The collegial game 2000 New York Cambridge University Press

Maltzman Forrest Wahlbeck Paul J May it please the chief? Opinion assignments in the Rehnquist Court American Journal of Political Science 1996 40 421 43

——— A conditional model of opinion assignment on the Supreme Court Political Research Quarterly 2004 57 551 63

Mattei Laura R. Winskey Mattei Franco If men stayed home. The gender gap in recent congressional elections Political Research Quarterly 1998 51 411 36

Maynard-Moody Steven Leland Suzanne Brudney J O’Toole L Rainey H Stories from the front-lines of public management: Street-level workers as responsible actors Advancing public management: New developments in theory, methods, and practice 2000 Washington, DC: Georgetown University Press 109 23

Maynard-Moody Steven Musheno Michael State agent or citizen agent: Two narratives of discretion Journal of Public Administration Research and Theory 2000 10 329 58

——— Cops, teachers, counselors: Stories from the front lines of public service 2003 Ann Arbor: University of Michigan Press

Meier Kenneth J Stewart Joseph Jr England Robert E Race, class, and education: The politics of second-generation discrimination 1989 Madison: University of Wisconsin Press

Meier Kenneth J Bohte John Structure and discretion: Missing links in representative bureaucracy Journal of Public Administration Research and Theory 2001 11 455 70

Meier Kenneth J Nicholson-Crotty Jill Gender, representative bureaucracy, and law enforcement: The case of sexual assault Public Administration Review 2006 66 850 60

Miller Gary Above politics: Credible commitment and efficiency in the design of public agencies Journal of Public Administration Research and Theory 2000 10 289

Moore James C Jr Status and influence in small group interactions Sociometry 1968 31 47 63

Murphy Walter F Elements of judicial strategy 1964 Chicago University of Chicago Press , 204, 100–352

National Center for State Courts An analysis of 2007 state court caseloads 2009 Williamsburg, VA: National Center for State Courts

O’Toole Laurence J Jr Meier Kenneth J Modeling the impact of public management: The implications of structural context Journal of Public Administration Research and Theory 1999 9 505 26

Peffley Mark Hurwitz Jon Sniderman Paul M Racial stereotypes and whites' political views of blacks in the context of welfare and crime American Journal of Political Science 1997 41 30 60

Pitts David W Representative bureaucracy, ethnicity, and public schools Administration & Society 2007 39 497

Price Joseph Wolfers Justin Racial discrimination among NBA referees 2007 Cambridge, MA National Bureau of Economic Research

Quillian Lincoln New approaches to understanding racial prejudice and discrimination Annual Review of Sociology 2006 32 299 328

Riccucci Norma M Managing diversity in public sector workforces 2002 Boulder, CO Westview

——— The pursuit of social equity in the federal government: A road less traveled? Public Administration Review 2009 69 373 82

Ridgeway Cecilia L Backor Kristen Li Yan E Justine E Tinkler Kristan G Erickson How easily does a social difference become a status distinction? Gender matters American Sociological Review 2009 74 44

Rohde David W Spaeth Harold J Supreme Court decision making 1976 San Francisco, CA W. H. Freeman

Rosenberg Gerald N The hollow hope: Can courts bring about social change 1991 Chicago University of Chicago Press

Rosenbloom David H Public administrative theory and the separation of powers Public Administration Review 1983 43 219 27

Rubin Ellen V Kellough J. Edward Does civil service reform affect behavior? Linking alternative personnel systems, perceptions of procedural justice, and complaints Journal of Public Administration Research and Theory 2012 22 121 41

Samaha Adam M Randomization in adjudication William and Mary Law Review 2009 51 1 86

Sanbonmatsu Kira Gender stereotypes and vote choice American Journal of Political Science 2002 46 20 34

Segal Jeffrey A Spaeth Harold J The Supreme Court and the attitudinal model revisited 2002 New York: Cambridge University Press

Selden Sally Coleman The promise of representative bureaucracy: diversity and responsiveness in a government agency 1997 Armonk, NY M.E. Sharpe

Sigelman Carol K Sigelman Lee Walkosz Barbara J Nitz Michael Black candidates, white voters: Understanding racial bias in political perceptions American Journal of Political Science 1995 39 243 65

Slotnick Elliot E Who speaks for the Court? Majority opinion assignment from Taft to Burger American Journal of Political Science 1979 23 60 77

——— Federal trial and appellate judges: How do they differ? Western Political Quarterly 1983 36 570 8

Smith Michael R Alpert Geoffrey P Explaining police bias Criminal Justice and Behavior 2007 34 1262 83

Songer Donald R Tabrizi Susan J The religious right in court: The decision making of Christian evangelicals in state supreme courts Journal of Politics 2009 61 507 26

Streb Matthew J Burrell Barbara Frederick Brian Genovese Michael A Social desirability effects and support for a female American president Public Opinion Quarterly 2008 72 76

Tarr G. Alan Porter Mary Cornelia Aldis State supreme courts in state and nation 1988 New Haven, CT: Yale University Press

Turco Catherine J Cultural foundations of tokenism American Sociological Review 2010 75 894

Ulmer S. Sidney Use of power in the Supreme Court: The opinion assignments of Earl Warren, 1953–1960 Journal of Public Law 1970 19 49

Unah Isaac Hancock Ange-Marie US Supreme Court decision making, case salience, and the attitudinal model Law & Policy 2006 28 295 320

Wagner David G Berger Joseph Gender and interpersonal task behaviors: Status expectation accounts Sociological Perspectives 1997 40 1 32

Walker Samuel Taming the system: The control of discretion in criminal justice, 1950–1990 1993 New York: Oxford University Press

Wenger Jeffrey B Wilkins Vicky M At the discretion of rogue agents: How automation improves women's outcomes in unemployment insurance Journal of Public Administration Research and Theory 2009 19 313 33

Wilkins Vicky M Exploring the causal story: Gender, active representation, and bureaucratic priorities Journal of Public Administration Research and Theory 2007 17 77 94

Williams Robert F State constitutional law: Cases and materials 1999 Charlottesville, VA LEXIS Law

Wise Charles R Christensen Robert K A full and fair capacity: Federal courts managing state programs Administration Society 2005 37 576 610

Woodruff Michael J Special interests and institutional change in state Supreme Courts (October 25) 2010 http://ssrn.com/abstract=1416968 or http://dx.doi.org/10.2139/ssrn.1416968

Wooldridge Jeffrey M Econometric analysis of cross section and panel data 2002 Cambridge, MA MIT Press

For a more complete review of the varying perspectives, see Wenger and Wilkins (2009 , 314–6).

Maynard-Moody and Musheno (2000) are careful to note that these roles (state- vs. citizen-centered agents) are simultaneous, even if agents do not always recognize the duality of their roles.

In our models, we control for “eligibility” by stipulating that our subjects are members of the majority coalition in a particular case (i.e., only members of the majority coalition would be eligible to write the majority opinion).

Recall Haynie’s (2002) work demonstrating that white state legislators ascribe lower competence and diminished effectiveness to their black peers.

Typically, state courts of last resort decide cases in five, seven, or nine judge panels.

Although we would like to examine race and gender effects more recently, these data linking judge and case characteristics are unique and are only recently becoming widely used.

Following arguably the most sophisticated existing study of majority opinion assignment ( Maltzman and Wahlbeck 2004 ), as an alternative to clustered standard errors, we also estimated a random effects model with case cluster. The results of all hypotheses tests for the main and control independent variables were identical to those in the models we chose to present.

Alternatively, instead of using separate models for each of the assignment regimes, one could pool the data, analyzing all 50 states together. In brief, we did this as an additional robustness check and found that none of our conclusions with respect to our hypotheses changed. We also note the following for readers who might be particularly interested in the pooled analysis. First, we included dummy variables for two of the three regimes (e.g., rotation and discretion) and multiplied each dummy variable separately with each of the covariates. For example, the model with rotation and discretion dummy variables included Tenure, Tenure × Random, and Tenure × Discretion. The coefficient for the Tenure variable then reflected the effects of tenure on opinion in the reference category regime—in the rotation states. The multiplicative term Tenure × Random, conversely, estimated the effects of tenure in random assignment states compared to the effects of tenure in rotation states. Of course, interpretation of the pooled model becomes relatively more complex. First, to estimate the overall effects of the independent variables in each regime (as opposed to effects of the variable in the regime compared to the reference category), we had to include a model for each regime. Additionally, each pooled model included more than three times as many nonfixed effects covariates. In return for this dramatic loss of parsimony, we gained little. Again, when we estimated the pooled models, the results conform with the findings of the nonpooled models presented herein.

The dummies are used in each model (see tables 3 and 4 ) but, to display parsimonious models, are not reported.

There is, of course, a third possible cause of autocorrelation in our data: the judges. Most of the judges sit in the majority in multiple cases, and it is possible that errors explaining the assignment to a judge in one case could be correlated with errors in other observations for that judge. Given the total number of judges in the data (several hundred in each subset we analyzed), combined with the nature of the data, we could not account for all types of autocorrelation simultaneously. Indeed, mixed models of random and fixed effects did not converge. Similarly, fixed-effects models (for both states and judges) with clustered standard errors for the case did not converge. We did, however, estimate the models with state-level fixed-effects while clustering by judge. In the random and rotation models, the changes in the results were minor: the results of the hypotheses tests for the main independent variables did not change and the changes in the hypotheses tests for the controls were minor (most were still significant at the 0.05 level; at worst, the significant controls in the presented models were significant in the judge-clustered models at the 0.063 level or less). The changes in the discretion models were more severe. Although the results of the hypotheses tests regarding the black male judge variables were consistent with those that we presented, the white female judge results did change. Specifically, the findings that contradicted our theory—that women judges are more likely to receive the opinion (in the model with all cases and in the nonsalient cases)—disappears. The coefficients are no longer significant, even at the 0.10 level. The gender finding that was consistent with our theory—that women are less likely to receive the assignment in salient cases (based on the statistical significance of the interaction of white female judge and salience)—is also less robust but still statistically significant at the 0.10 level. Additionally, although all the controls were significant in the discretion models we presented, three of the five are not close to statistically significant in the models that cluster the standard errors by judge. Of course, we are not concerned with the statistical significance of the controls. Overall, although this robust check slightly diminishes our findings regarding one of our main hypotheses in one model (inferences regarding the interaction of salience and white female judge is now less certain), it also strengthens our findings regarding another hypotheses (i.e., in these models, there is no evidence that white females are more likely to receive the assignment under any conditions).

To account for this possibility, we also estimated the models using year fixed effects. The resulting estimates were substantively indistinguishable to those estimates in the more parsimonious models that we chose to present.

There were an insufficient number of black female judges to explore assignment and the intersectionality of race and gender. Similarly, due to the small number of Latinos and Asian Americans overall (1% each) as well as the small number of states with Latinos (6%) and Asian Americans (2%), we were also unable to explore these groups. Indeed, in some assignment subsamples, we had no Latinos and/or no Asian Americans. To enhance parsimony and facilitate interpretation, we excluded all observations unless the judge was white or an African American male.

Given both the inclusion of state-level fixed effects, multiplicative terms (see Brambor, Clark, and Golder 2005 ), variables measuring related concepts (e.g., age and tenure), high multicollinearity is always a concern. However, the variance inflation factors (VIFs) for all the variables are relatively low. The highest VIF in any model, for one of the state dummy variables, is 3.18, well below troublesome levels (see Gujarati and Porter 2009 ). Moreover, in half the models, the highest VIF was under 1.5.

It is possible that age and/or tenure are nonlinearly related to opinion assignment. For example, due to various biases, younger and older judges might be perceived as less competent compared to middle-aged jurists. To account for this potential parabolic relationship, we estimated the models including squared terms for the age and tenure variables. In the rotation assignment models, we observed no evidence of a parabolic relationship of any kind. In the other two analyses of the other two assignment processes, we observed some evidence of parabolic relationships. In other words, in some of the models, one or more of the coefficients (for the stand alone term and the squared term) were statistically significant. However, the direction of the coefficients varied by assignment process. Most importantly, the inclusion of either squared term had no effect on the hypotheses for the main independent variables (or the other control variables). For that reason, along with the mixed results and the desire for more parsimonious models, we chose not to include the squared age and tenure terms.

Of course, it is possible that case salience conditions the effects of our other independent variables. If that is true, one might consider splitting the samples into salient and nonsalient cases. As a robust check, we split the samples and then estimated the models in all three assignment regimes. Using this method, we found no evidence of an interaction—at least not the one posited. Instead, the results in the nonsalient models mirrored the results in our table 3 , whereas the results in the salient case samples were quite different. None of the controls were significant, and the main independent variable coefficients in the rotation and discretion models were not significant. The only evidence of an interaction: we did observe the expected relationships for black males and white females (though the coefficient was only significant at the 0.055 level) in so-called random assignment courts in salient cases but did not observe a statistically significant relationship in nonsalient cases. Overall, although we still find support for the salience interaction in one context (random assignment), the lack of observed interaction (at least the one posited) in the other contexts, or among any of the controls in any contexts, suggests that there is no need to split the samples further. Moreover, the counterintuitive findings suggest that the results are a function of the different sample sizes (as noted in table 2 , only 8.4% of the cases are salient). Finally, we tested whether there were indeed interactions between salience and the control variables by estimating models including multiplicative terms (multiplying the control by the salience variable) for each control. The results suggest there is little reason for including these multiplicative terms. First, the models do not change the results of our hypotheses tests for our main independent variables (or controls). Also, across the three models, only 2 of 15 controls (chief justice and tenure in the discretion model) exhibited signs of a conditional relationship. Weighing the limited utility gained from including the additional multiplicative terms for our control variables with the obvious loss in parsimony, we chose not to present those models.

With the exception of salience, all our independent variables have a predicted unidirectional impact on likelihood of assignment (e.g., tenure should increase likelihood of assignment). We therefore use one-tailed significant tests at the p < 0.05 level of impact, unless otherwise noted in the individual tables.

We also recognize that although status characteristics explain inequitable assignment in some administrative contexts, the discretionary use of status characteristics may lead to other outcomes unrelated to equity (e.g., decreased chance of a decision being overturned in a federal appellate court). Although we are unable to explore this in the current data, we recognize the value of a more thorough exploration of the outcomes related to the discretionary use of status characteristics.

White women were no more or less likely than white men to receive an assignment in a salient case.

To calculate the conditional probabilities, we used Stata’s prvalue command, developed by Long and Freese (2001) as part of the spost postestimation suite of commands. So doing, we were able to focus on salience’s conditioning effects on race and gender while controlling for all the variables in our model. We also note alternative specifications of salience’s impact as discussed in footnote 13.

It is possible that a handful of states are driving the observed results in the random and discretion states. We account for this in part by including the state-level fixed effects. We also tried to test for this (in models that are not presented) by multiplying the main independent variables (black male, white female, and the salience multiplicative terms) by each state dummy. Interestingly, the only observed effect appeared to be in the rotation cases, where some of the State × White female judge and State × Black male judge multiplicative term coefficients were statistically significant. In the random and discretionary assignment states, this was not the case. The only significant state dummy multiplicative term coefficients appeared to be those with salience but without gender and race.

One might argue that there are still selection effects since we are only examining those judges that decide to vote in the majority. To account for this, we also estimated the models using a Heckman's selection probit model. The results of the Heckman model were, if anything more robust, suggesting that our findings are not driven by possible selection effects. Given that the interpretation of the Heckman models is more complex and that the presentation and interpretation would consume significant journal space, and that the results suggest that there are no selection effects, we chose not to present the Heckman models.

We are indebted to an anonymous reviewer for the deft articulation of this particular insight.

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1477-9803
• Print ISSN 1053-1858
• Copyright © 2024 Public Management Research Association
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Gender Bias

Gender bias refers to the unequal treatment of individuals based on their gender. This can take many forms, including stereotyping, discrimination, and unequal access to opportunities and resources. Bias manifests itself in various ways and causes us to form prejudices against others. It assists us in categorizing things in order to make sense of the world around us. Biases, while common, are not harmless. Gender bias, for example, is a very common issue in the workplace. It is the preference for one gender over another. Companies and recruiters generally prefer men to women. But we’re not saying it was a conscious decision.

Gender bias can be present in many areas of society, including education, employment, politics, and the media. For example, women may be paid less than men for the same job, or they may be passed over for promotions because of their gender. In the media, women may be portrayed as overly emotional or weak, while men are portrayed as strong and capable.

Gender bias is a pervasive problem that can have significant negative effects on individuals and society as a whole. It can limit opportunities for women and girls, perpetuate harmful stereotypes, and undermine efforts to achieve gender equality. Addressing gender bias requires a concerted effort from individuals, organizations, and governments to promote awareness, education, and action to create a more equitable and just society.

Gender bias can be conscious or unconscious and can occur in various domains such as education, employment, healthcare, politics, and media. For example, in the workplace, gender bias may lead to men receiving preferential treatment in hiring, promotion, and salary negotiations. Similarly, in healthcare, women may be treated differently than men due to stereotypical beliefs about their pain tolerance or emotional capacity.

Addressing gender bias requires awareness and understanding of its presence and its negative impact on individuals and society. It involves actively challenging gender stereotypes and promoting equal opportunities and treatment for people of all genders. This can be achieved through policies, education, and training programs that promote gender equality and diversity.

Gender bias is a real issue because men are statistically more likely than women to receive preferential treatment in the hiring process. To be a successful business, you must have both men and women on your team. If you don’t do this, your company will never be diverse and inclusive. Because gender bias occurs as early as the hiring process, you should begin by addressing it there.

Gender bias is detrimental to both men and women as it limits individuals’ potential and contributes to social inequality. Therefore, it is crucial to promote gender equality and challenge gender stereotypes to create a fair and just society where everyone has equal opportunities and rights.

Social Commerce Concept

Lessons from the holocene on sustainability, evolution, and human adaptation, discuss on national martyr’s memorial, socio cultural evolution, critical financial statements, mobile web development knowledge, presentation on mixed economy, sample dishonored check notice, about yellow fever, environmental justice – a social movement, latest post, needle valve, hydroforming, solving the puzzle of early medieval money, human activities have a significant impact on the deep underground fluid flow, a microbial plastic factory for high-quality green plastic, power optimizer – a dc to dc converter technology.

• MyAucklandUni
• Student Services Online
• Class search
• Student email
• Change my password
• MyCDES+ (job board)
• Course outlines
• Learning essentials
• Libraries and Learning Services
• Forms, policies and guidelines
• New students
• Enrol in courses
• Campus card
• Postgraduate students
• Summer school
• AskAuckland
• Student Hubs
• Student IT Hub
• Student Health and Counselling
• Harassment, bullying, sexual assault and other violence
• Complaints and incidents
• Career Development and Employability Services (CDES)
• Ratonga Hauātanga Tauira | Student Disability Services (SDS)
• Rainbow support
• Covid-19 information for our community
• Emergency information
• Report concerns, incidents and hazards
• Health and safety topics
• Staff email
• Staff intranet
• ResearchHub
• PeopleSoft HR
• Forms register
• Careers at the University
• Education Office
• Early childhood centres
• University Calendar
• Opportunities
• Update your details
• Make a donation
• Publications
• Photo galleries
• Video and audio
• Career services
• Virtual Book Club
• Library services
• Alumni benefits
• Office contact details
• Alumni and friends on social media
• No events scheduled for today You have no more events scheduled for today
• Next event:
• Show {0} earlier events Show {0} earlier event
• Event_Time Event_Name Event_Description
• My Library Account
• Change Password
• Edit Profile
• My GPA Grade Point Average About your GPA GPA not available Why can't I see my GPA?
• My Progress
• Points Required Completed points My Progress Progress not available All done!
• Student hubs
• Health and counselling
• All support
• Health, safety and well-being

Breadcrumbs List.

• News and opinion

Does gender-segregated schooling make a difference?

30 April 2024

Education and society , Faculty of Science

Opinion: Students at girls-only and boys-only secondary schools do better at maths and science than their peers at co-ed schools, but why, asks Tanya Evans

The debate over single gender versus co-educational schooling has long been controversial. I went to a co-ed school and was inspired by a remarkable woman who was my maths teacher, and because of her deep knowledge and passion for the subject, I knew that maths was definitely an option for girls – she was my role model.

It may well be that because of her, I went to uni and majored in maths – a path that led to a scholarship to complete a PhD in pure mathematics at Rice University in Houston, Texas. I loved my co-ed school – which is just as well as there was no alternative. Everyone went to co-ed schools in my hometown. Would it have made any difference if I’d gone to a single-gender school?

It came as a big surprise to me when I moved to New Zealand to realise that over 14 percent of secondary students attend single-gender schools. Moreover, unlike the USA and Australia, where single-gender schooling is predominantly offered by private and/or Catholic schools, most New Zealand gender-segregated schools are state schools.

In many parts of New Zealand, students don’t have a choice – they are zoned only for single-gender schools. For a maths education researcher like me, New Zealand presented an exceptional opportunity to investigate the impact of gender segregation on maths education.

A few years ago, I was contacted by Distinguished Professor Geoff Chase from the University of Canterbury. He and his colleagues noted a strange statistic: of all girls enrolled in engineering degrees at the University of Canterbury between 2005 and 2017, 56 percent had attended girls-only schools. This is almost five times higher than expected.

Another report by the Ministry of Women found that girls from girls-only schools in Auckland ranked more male-dominated professions in their top 10 occupational choices than girls from co-educational schools in the same city. But the research into the impact of single-gender versus co-ed schools and whether it affects students’ interest and achievement in STEM subjects (Science, Engineering, Technology, and Mathematics) is complicated, and often contradictory. A comprehensive study of schools in Seoul in South Korea, where assignment to single-gender or co-ed high schools is random, found significantly positive effects of all-boys schools consistently across different STEM outcomes, but not for girls.

My summer research scholar, Alice Smith, and I decided to investigate the role of school type on the maths and science achievement of Year 9 girls and boys in New Zealand.

What could explain these findings? We don’t know. We can speculate that single-gender schools may excel in counteracting prevailing gender stereotypes that have long plagued maths and science. 

We examined data from more than 5,900 Year 9 students, sourced from the Trends in International Mathematics and Science Study (TIMSS) 2019 dataset. The findings of our study reveal a compelling narrative: students at girls-only and boys-only secondary schools did better at maths and science than their peers at co-ed schools.

However, the extent of this advantage and its significance depended on gender and the socio-economic background of the students. Boys at single-gender schools perform significantly better than in co-ed schools with small effect sizes across all socioeconomic bands, which is in line with international evidence. However, the results for girls were surprising.

The outperformance was most pronounced by students in low socio-economic environments. Girls at low-decile girls-only schools performed significantly better than their counterparts in co-ed schools, and the effect size was large. They also surpassed boys in both boys-only and co-ed schools with medium effect size.

In the high-decile band, however, the advantage of girls-only schools was insignificant.

What could explain these findings? We don’t know. We can speculate that single-gender schools may excel in counteracting prevailing gender stereotypes that have long plagued maths and science. We can speculate that by purposefully positioning themselves as bastions of gender equality, these schools create environments where aspirations aren’t circumscribed by gender norms. St Cuthbert’s College for Girls in Auckland is an example of a school that actively promotes the involvement of women in STEM fields through deliberate messaging and engagements with prominent female STEM leaders who serve as role models.

However, we must approach these findings with a critical lens. We need to be mindful of the potential confounding factors such as selection bias. It is plausible that academically inclined students gravitate towards single-gender schools, skewing the results. On the other hand, students in many New Zealand regions have no choice because of zoning since public single-gender schools are the only option.

Other study limitations are the lack of longitudinal data – we do not know what happens during all five years of secondary schooling and our study involved a small sample size of low-decile schools; only two low-decile girls-only schools were in the sample (versus 26 co-ed schools), raising the possibility that idiosyncratic factors had an effect and the results can’t be generalised.

But we clearly need to understand what those two low-decile girls-only schools did so well. Could the absence of boys in maths and science classes diminish the manifestation of societal stratification and remove the stereotyping effect – a well-documented phenomenon known as stereotype threat? What is more than apparent is that while we still have single-gender schools, the impact of that on students’ education and their career choices needs to be better understood.

Dr Tanya Evans is a senior lecturer in the Department of Mathematics

This article reflects the opinion of the author and not necessarily the views of Waipapa Taumata Rau University of Auckland.

This article was first published on Newsroom,  Does gender-segregated schooling make a difference? 30 Aoril 2024

Media contact

Margo White I Research communications editor Mob 021 926 408 Email margo.white@auckland.ac.nz

• Global Presence

• A Premier Business School in Asia with Global Influence
• Message from the Dean
• Vision & Mission
• Partnership & Global Network
• Beta Gamma Sigma HKU Chapter
• Rankings & Accreditations
• Corporate Collaboration
• Faculty Management
• International Advisory Council
• Our Stories
• Staff Portal

• Diversified Learning Experience Empowers our Students
• Undergraduate
• Executive Education

• International Faculty Body that Offers Quality Education
• Post-Doctoral Fellows
• Demonstrators/ Teaching Assistants
• Research Postgraduate Students
• Administration

• Lead the Frontier of Knowledge Creation
• Latest Research Publications
• HKEJ Column
• In the Media
• Global Scholars
• Academic Areas
• Faculty Members
• Research Grants
• Edward K Y Chen Distinguished Lecture Series
• Research Centres & Institutes
• Hong Kong Macroeconomics Forecast
• HKU Knowledge Exchange
• HKU Scholars Hub
• Shenzhen Research Institutes

• Inspire the Society and Community with Rigorous Insights
• School News
• Press Release

• Fostering Strategic Partnerships and Talent Development
• Recruit our Talents
• Student Career Development
• Faculty Recruitments
• HKU Career Site

Gender Bias in Job Assignment: Evidence from Retail Frontline Managers

Dr. Susan Feng Lu Associate Professor Gerald Lyles Rising Star Professor of Management The Krannert School of Management, Purdue University

Anecdotal evidence suggests that gender disparity in job assignment might be a main driver behind the gender pay gap. However, existing gender studies have focused on individual workers or C-suite executives, while systematic empirical studies on low-level managers are rare. In this study, we empirically investigate gender disparity in the job assignment of frontline managers, specifically the effect of gender on store-manager assignment, using personnel, sales, and operational data from a large sportswear retail chain. Our estimation strategy is a two-step fixed effects framework: 1) Separately identify manager fixed effects and store fixed effects by exploring manager switching across stores; 2) Estimate the effect of gender on store-manager assignment while controlling for the estimated manager fixed effects from the first step. Applying this estimation framework to the retail chain’s data, we uncover strong evidence for a gender bias in store-manager assignment—male mangers are more likely than their female counterparts to be assigned to stores with high sales potential, i.e., core stores, stores with a large number of sales clerks, or stores with a high sales target. To support our finding on gender bias in store-manager assignment, we test three alternative hypotheses. First, we find that male managers achieve similar sales when replacing their female counterparts in the same stores, and male managers are not better at managing the turnover of sales clerks, neither of which supports an alternative hypothesis of differential managerial abilities; Second, we find that male sales clerks on average achieved similar or lower sales individually than their female counterparts, thereby not supporting a career selection bias hypothesis that males seeking the sales clerk positions at the retail chain tend to have high sales skills. Third, we conduct a maximum likelihood estimation of a discrete-time Markov chain that models the processes of store-manager assignment and manager turnover, and do not find evidence to support the hypothesis that inequitable store-manager assignment is driven by different gender preferences to leave stores with low sales potential. Translating our findings into an actionable insight, we develop a gender inequity index (GII) to help organizations identify potential gender bias in job assignment. Applying this index to a simulation study using the retail chain’s store manager compensation data, we demonstrate the implications of gender-biased job assignment on the gender pay gap.

EEOC says workplace bias laws cover bathrooms, pronouns and abortion

The U.S. Equal Employment Opportunity Commission on Monday said employers refusing to use transgender workers’ preferred pronouns and barring them from using bathrooms that match their gender identity amounts to unlawful workplace harassment under federal anti-discrimination law.

The EEOC  updated its enforcement guidance on workplace harassment for the first time in 25 years, including to reflect a landmark U.S. Supreme Court ruling that anti-bias laws cover LGBTQ workers, after an earlier attempt stalled during the Trump administration.

The commission in the new guidance also addressed the rise of remote work and said that discriminating against employees based on their decisions to have abortions or use contraception is a form of sex discrimination.

The guidance is not legally binding, but lays out a blueprint for how the EEOC will enforce anti-bias laws and can be cited in court to back up legal arguments.

Some Republicans and conservative and religious groups had criticized the expansive guidance after the commission  unveiled a draft version  in September. They said it conflicts with state laws on abortion and LGBTQ issues and fails to acknowledge that religious employers are exempt from anti-discrimination laws in many cases.

EEOC Chair Charlotte Burrows in an interview said the guidance reflects decades of court rulings that have expanded workers’ rights to be free from workplace harassment, and that updating it was necessary to ensure that employers are aware of their legal obligations.

“We felt we really needed to lay out the contours of the law and where it stands,” said Burrows, an appointee of Democratic President Joe Biden.

The Democrat-led commission approved the guidance in a 3-2 vote on Friday, with Republican Commissioners Andrea Lucas and Keith Sonderling dissenting, a spokesman for the agency said.

Lucas, an appointee of Republican former President Donald Trump, said in a statement that the guidance would push many employers to eliminate single-sex bathrooms and facilities for showering, dressing, and sleeping, exposing women to increased risk of harassment and assault.

“Biological sex is real, and it matters. Sex is binary (male and female) and is immutable,” she said.

More than one-third of the hundreds of thousands of worker complaints the EEOC received between 2016 and 2023 included allegations of harassment, the commission said.

The agency has recently filed high-profile harassment lawsuits  accusing Walmart Inc  of failing to stop severe sexual harassment of female workers by the manager of a West Virginia store, and claiming Tesla Inc has for years  ignored complaints of rampant harassment  of Black workers at its flagship Fremont, California, assembly plant. Both companies have denied wrongdoing.

The commission had not updated its legal guidance on harassment based on sex, race, and other protected traits since 1999, a year after the U.S. Supreme Court  issued a major ruling, involving workplace sexual harassment.

The agency released proposed changes in January 2017, just as Republican former President Donald Trump was taking office and months before the #MeToo movement intensified scrutiny of workplace sexual harassment, but they were never finalized under Trump.

The lengthy new guidance covers an array of topics related to workplace harassment and reflects significant changes in the law over the last quarter-century, such as  a 2020 U.S. Supreme Court ruling  in Bostock v. Clayton County that said workplace discrimination based on sexual orientation or gender identity is a form of unlawful sex bias.

The commission said that while Bostock involved a gay worker’s termination and not harassment claims, the Supreme Court’s reasoning in the decision extends to cases alleging harassment of LGBTQ workers.

The commission in the guidance also detailed how harassment can occur even between remote workers, saying that conduct “within a virtual work environment” such as racist or sexist comments made during a video meeting can contribute to an unlawful hostile work environment.

And conduct that occurs away from work, including workers’ posts on their personal social media pages, can constitute illegal harassment if it affects an employee’s working conditions, the EEOC said.

EEOC says workplace bias laws cover bathrooms, pronouns, abortion

• Medium Text

• Company Tesla Inc Follow
• Company Walmart Inc Follow

Sign up here.

Reporting by Daniel Wiessner in Albany, New York

Our Standards: The Thomson Reuters Trust Principles. New Tab , opens new tab

Thomson Reuters

Dan Wiessner (@danwiessner) reports on labor and employment and immigration law, including litigation and policy making. He can be reached at [email protected].

Read Next / Editor's Picks

Industry Insight Chevron

Mike Scarcella, David Thomas

Karen Sloan

Henry Engler

Diana Novak Jones