This International Women’s Day, let’s celebrate taking an evidence-based approach to understanding gender inequality in the labour market. Apparently, “men apply for a job when they meet only 60% of the qualifications, but women apply only if they meet 100%”. Several media outlets have circulated this claim, including reputable sources such as the Harvard Business Review and Forbes, and it appears in books such as Sheryl Sandberg’s Lean In.
However, an investigative journalist found that this statistical-sounding claim is likely based on a speculative comment made by a senior executive at Hewlett Packard, rather than quantitative data.
Its success in capturing the imagination of so many is likely because it feels intuitively true. Evidence from LinkedIn suggests that women may be more ‘selective’ when applying to jobs. Men and women view similar numbers of jobs, but women are less likely to apply, particularly to roles more senior than their current position. This selectivity may be harmful if it lowers the chances that women progress into more senior and better paid roles. With a record number of unfilled job vacancies, employers cannot afford to put qualified women off.
We explored the range of reasons that women may apply to fewer of the jobs they view than men in a previous blog, many of which relate to caring responsibilities. However, the claim at the top of the blog focuses specifically on women having a different response to job requirements than men. In response to the questions we posed in that blog, we carried out our own research to get closer to the truth.
Testing the claim in an online experiment
We ran an online experiment with over 10,000 participants who were all currently looking or had recently looked for a job. We asked them to view a job advert and report how willing they were to apply for the role. The job advert included details such as salary, commute length, flexible working policy, and company information worded in a way to minimise gender differences and appeal to as wide a range of personal preferences as possible, but also to make it less obvious the experiment was focused on job requirements.
Crucially, and unlike other studies, we measured how qualified participants were for the role in terms of the five soft skills it asked for. Since we could not find an adequate existing measure, we developed our own battery of 70 items that asked about a range of relevant experiences from multiple skill and experience levels. With this, we were able to measure differences in willingness to apply among similarly qualified applicants.
Men apply when they’re 52% qualified. Women apply when they’re 56% qualified.
We found that, among similarly qualified men and women, men were more willing to apply to a given role. However, the difference is not nearly as dramatic as the claim suggests. We find that women apply when they meet 56% of the requirements, whereas men apply when they meet 52% of them. Looking behind the averages, we also find that this gender difference emerges solely among less qualified participants. Among more qualified participants, there is no gender difference in willingness to apply.
Much of this small difference in willingness to apply is explained by (or mediated by) gender differences in self-perceptions of meeting the overall requirements. When reading the same job ad, similarly qualified men thought they met the requirements to a slightly greater extent than women.
Wider research suggests that men are more likely to overestimate their capabilities in stereotypically masculine contexts, which may explain these differences. We deliberately chose a leadership role in our experiment, which is stereotyped as masculine, since women’s lower representation in these roles is an important driver of the gender pay gap.
Opportunity for a solution
We also found an intriguing result in the way in which participants appraised job requirements individually compared to as a whole. Men were more likely to say they met the overall job requirements. But when we asked participants to rate themselves against each requirement separately, we found that the sum of women’s self-perceptions of meeting individual requirements was greater than men’s. This suggests that people base their decision to apply more on their overall impression of requirements. However, it also provides an exciting opportunity for an intervention to close the gender gap in self-perceptions.
Implications for employers, job platforms and policymakers
Test assumptions and experiment. This work shows the value of digging deeper behind received wisdom and discovering important nuances. Building on our results, we recommend:
- Testing nudges to applicant behaviour. Job platforms and recruiters could test a nudge that encourages potential applicants to consider how well they meet individual requirements rather than relying on their overall impression.
- Testing nudges to hiring manager behaviour. While many employers are already aware that they need to reduce the number of unnecessary job requirements in job adverts, it can be difficult to get this to happen in practice. We recommend job platforms and recruiters test a nudge that asks managers only to include requirements that would rule a candidate out to avoid creating a laundry list of things they would like.
If you would like to work with us to test these ideas and close the gender gap, please reach out to firstname.lastname@example.org.
This work was funded by the Government Equalities Office as part of the Gender & Behavioural Insights Programme.