Print Friendly, PDF & Email


How can we close the gender gap in high-level positions in organizations? Prior studies have shown that one possible reason for this gap is that, because of gendered expectations, women are less likely to enter competitions than men. This research article shows that changing competitions from an “opt-in” default, where people who wish to compete must self-nominate, to an “opt-out” default, where everyone competes by default but can opt-out if they choose, eliminates gender differences in the decision to compete without affecting performance or the wellbeing of applicants. These results suggest that organizations can use an opt-out default to reduce gender gaps in promotions or other competitive processes— such as pitch competitions and innovation contests—that are often plagued by an underrepresentation of women.

Research and Findings

Due to prevailing gender norms, women are less likely to promote themselves, exaggerate their accomplishments, and show (over)confidence when compared to men because of the penalties they experience for doing so. Thus, they are less likely to nominate themselves for competitive selection processes, such as promotions, awards, and admissions. Through three experiments, this study showed that changing the decision to compete from an opt-in process involving self-nomination to an opt-out process where competition is the default can reduce or even eliminate these gender differences.

The first two experiments we conducted in research labs and involved 482 and 639 undergraduate students, respectively. Students were told to add five two-digit numbers for compensation. There was a non-competitive compensation scheme ($0.50 per correct answer) and a competitive compensation scheme ($2 per correct answer, but only if the student had the highest score compared to three randomly chosen competitors). Some students were randomly assigned to an opt-in default: everyone received non-competitive compensation but could choose to get competitive compensation instead. Others were randomly assigned to an opt-out default: everyone received competitive compensation but could choose to get non-competitive compensation. In other words, for those assigned to opt-in, the default was non-competition, but participants could self-nominate to compete. For those assigned to opt-out, the default was competition, but they could still choose not to compete.

Both experiments showed that the opt-out default eliminated the gender gap in competition. For those who had to opt-in to competition, far fewer women than men chose to compete (47% of women vs. 72% of men in the first experiment and 53% of women vs. 72% of men in the second experiment). For those who had to opt-out from competition, women and men chose competition at similar rates (75% of women vs. 76% of men in the first experiment and 74% of women vs. 78% of men in the second experiment). One might worry that eliminating the competition gap might subject women to more stressful or anxiety-provoking activities that they wouldn’t have chosen otherwise; however, a post-activity survey showed that there was also no significant difference in anxiety levels between the opt-in and opt-out groups.

The third experiment tested these results in a real-life setting through a large-scale experiment on Upwork, an online market for clients to find freelancers. The authors operated as a client and hired 477 freelancers to complete a data entry job. The freelancers did not know they were part of an experimental study and completed the task as part of their day-to-day jobs. Workers first did a test project with a base compensation of $5 and a commission of $0.25 per correct data entry. After this, freelancers were randomly assigned to either opt-in or opt-out of competing for a more advanced task. The advanced task paid more: $7.50 base compensation with $1.00 bonus commission. Freelancers were told that if they competed for the advanced task, they would only get to complete it—and get compensated—if they were in the top 25% of performers. Otherwise, they would not complete any further tasks.

Results of this experiment showed that women again were significantly less likely than men to compete for the advanced task when they had to opt-in to competition (57% of women vs. 73% of men chose to compete). For those who were assigned to opt-out, there was no significant difference by gender in decision to compete (67% of women vs. 72% of men chose to compete, but this was not statistically significantly different). Because there is a concern that being in a competition might negatively affect women’s task performance, the researchers verified that there were also no negative consequences in performance for those in the opt-out group.


Organizations can eliminate gender gaps in competitions by using opt-out framing—Women tend to be less likely than men to nominate themselves for competitions. This study shows that when the default is for participants to opt-out from rather than opt-in to a competition, this gender gap disappears. Women’s supposed aversion to competition may depend on how the option to compete is presented. Thus, organizations may be able to use an opt-out process to close gender gaps in various competitive processes such as for promotions.

Opt-out framing in competitions does not decrease performance or wellbeing—Results suggest that those in the opt-out group did not show worse performance or higher anxiety compared to those who had to opt-in to compete. Opt-out framing of competition signals that competing is a norm for everyone, which may reduce women’s perceptions that competing is a counter-normative way of promoting oneself and/or demonstrating (over)confidence.


Research brief prepared by:

Carmina Ravanera

He, Joyce C., et al. “Opt-out Choice Framing Attenuates Gender Differences in the Decision to Compete in the Laboratory and in the Field.” Proceedings of the National Academy of Sciences, vol. 118, no. 42, 2021,


Opt-out choice framing attenuates gender differences in the decision to compete in the laboratory and in the field


Joyce C. He, Sonia K. Kang and Nicola Lacetera


Proceedings of the National Academy of Sciences of the United States of America






Research brief prepared by

Carmina Ravanera