When Performance Trumps Gender Bias: Joint Versus Separate Evaluation

Decrease gender bias in hiring and promotion decisions by evaluating candidates in groups, rather than one by one.


While women make up more than half of management and professional jobs within the U.S. workforce, they hold only 5 percent of CEO positions and fewer than 17 percent of board seats (Catalyst, June 2014). One contributing factor could be implicit, or unconscious, bias in hiring and promotion decisions, where the evaluator unconsciously ascribes group stereotypes to the individual interviewing. When hiring and promotion decisions are made for upper management positions, they are often done so one at a time, or as a separate evaluation. Evidence suggests that implicit gender biases are automatically activated when evaluators learn a candidate’s gender, which can lead to unintentional discrimination rather than impartial judgment. Instead, evaluating candidates in groups, or in joint evaluations, can mitigate these unintentional biases, enabling the evaluator or recruiter to compare a candidate’s performance, anticipated future productivity, and economic value. This study examines how joint evaluation could decrease gender discrimination in performance evaluations, job assignments, and hiring and promotion processes.


Joint evaluation decreased biases and increased the likelihood that employers will assess individuals based on their performance and potential, rather than gender stereotypes.

  • In separate evaluation, employers chose men over equally qualified women for male-stereotypical assignments and even preferred lower performing men to higher performing women for these tasks. Employers preferred women to equally qualified men for female-stereotypical assignments.
  • In joint evaluation, the gender gap disappeared. Employers were as likely to choose women as men and preferred higher performing employees to lower performing employees in both tasks.
  • Employing joint evaluation processes encourages more information-based, comparative, and data-driven promotion, job assignment and hiring decisions. This increases the likelihood of choosing the more qualified candidate by decreasing the employer’s reliance on stereotypes.

Employers may make different decisions in joint evaluations because they have directly comparable candidates. This nudge enables them to take a more reasoned approach, relying on performance data and discourages implicit, stereotype-based judgments. Furthermore, joint evaluations provide more information on candidates’ abilities, potentially providing more counter-stereotypical data points and encouraging evaluators to update their beliefs about the stereotyped group’s (e.g., women) abilities.


Study participants played either an employer or employee role. “Employees” were given a stereotypically male task (math) or a stereotypically female task (verbal) and received payment based on their performance. They performed the task twice. “Employers” were informed of the employees’ first-round performance as well as the average performance level of all the employees, and selected an employee for hire in the second round in the two following scenarios:

  • Separate evaluation scenario: Employers were presented with one low- or one average-scoring male or female employee and chose between hiring that employee and being assigned a different employee at random.
  • Joint evaluation scenario: Employers were given one female and one male employee who were either low or average performers and chose between hiring one of these two employees or a randomly assigned employee.

In all scenarios, the lower-performing employee scored slightly below the average – this selection was intentional based on beliefs that gender-based discrimination would be more likely to affect “close-call” hiring and promotion decisions. There was no significant gender difference in performance on either task in both the first and second rounds, and the first-round performance was equally predictive of future performance for both genders. In both scenarios, employers were paid based on the second-round performance of either their chosen or the randomly assigned employee.

Related GAP Studies