The Effect of an Intervention to Break the Gender Bias Habit for Faculty at One Institution: A Cluster Randomized, Controlled Trial

Following a 2.5-hour departmental-level workshop that presented gender bias as a changeable habit, faculty reported personal changes supporting gender equity, both shortly after the intervention and at 3-month follow-up.

• Awareness of one’s own gender bias, measured with a scale composed of nine survey items, increased significantly among faculty in intervention departments, compared to faculty in control departments, with effects persisting after 3 months.
   
• Internal motivation to promote gender equity increased significantly among faculty in intervention departments, compared to faculty in control departments, though effects did not persist after 3 months.
• External motivation to promote gender equity increased significantly after 3 months among faculty in intervention departments, compared to faculty in control departments—suggesting a change in departmental norms.

• Confidence in one’s own ability to promote gender equity increased significantly among faculty in intervention departments, compared to faculty in control departments, with effects persisting after 3 months.
   
• Expectations of beneficial outcomes from promoting gender equity increased significantly among faculty in intervention departments, compared to faculty in control departments, though effects did not persist after 3 months, and there was no significant difference in expectations of negative repercussions.
   
• Self-reported action to promote gender equity in one’s department increased significantly after 3 months among faculty in intervention departments, compared to faculty in control departments—but only when at least 25% of faculty attended the workshop, suggesting a need for a higher critical mass for the intervention to be effective.
   
• Implicit bias, as measured by an Implicit Association Test (IAT), did not different between faculty in intervention departments and faculty in control departments—showing that changes in behaviors and attitudes were possible in spite of unconscious bias.
   
• Faculty in intervention departments also indicated on a climate survey that they felt like they fit in better, were more valued among colleagues for their scholarship, and felt more comfortable raising personal and family responsibilities at work, compared to faculty in control departments.

In short, a habit-breaking approach to gender bias can change individual behaviors and attitudes, in turn shifting organizational climate and norms.

The study was conducted across 92 departments or divisions housing 2,290 faculty members at the University of Wisconsin—Madison. Departments were pair-matched by school or college, size, and one of four disciplinary categories: biological sciences, physical sciences, social sciences, or arts and humanities. In each pair, departments were randomized to either the intervention group, or a waiting-list control group that was offered the intervention following the study’s conclusion.

Departments in the intervention group were offered a 2.5-hour interactive workshop presenting evidence on stereotype-based gender bias and its negative impact on science, health, and economics, followed by three habit-changing modules: (1) a review of research on the origins of bias as a habit; (2) bias literacy, describing six forms of stereotype-based gender bias (expectancy bias, prescriptive gender norms, occupational role congruity, redefining credentials, stereotype priming, and stereotype threat); and (3) self-efficacy for overcoming gender bias, describing five evidence-based behavioral strategies to practice (stereotype replacement, positive counterstereotype imaging, perspective taking, individuation, and increasing opportunities for contact with counter stereotypic exemplars), as well as two counterproductive strategies (“gender blindness” and belief in one’s own objectivity). Participants engaged in discussion, made a written commitment to action, and received materials to take home. In intervention departments, faculty attendance averaged 31% and ranged from 0% (in 3 departments) to 90%.

Two days before, 3 days after, and 3 months after the workshop, faculty were invited to complete an online survey including an implicit association test (IAT) and questions about gender bias awareness, motivation and self-efficacy to change, positive outcome expectations, and action. Faculty also received a climate survey before and after the intervention.