This opinion piece was written by Shoshana Davidson, a senior advisor at the Behavioural Insights Team. If you’re interested in writing an opinion piece, take a look at Apolitical’s guide for contributors.

The UK government recently celebrated a milestone in relation to gender equality: 100% of eligible companies had reported their gender pay gap data following recent legislation that requires companies with over 250 employees to report their data on a government-owned reporting website.

But there was also reason for concern: the data shows that over 75% of UK companies pay their male employees more, on average, than their female employees.

Looking at their figures — and sharing them publicly — has spurred many employers to think about how they can start to improve their gender pay gap for future years.

Some employers are now thinking about gender equality for the first time. Others have tried various ways to improve gender equality over the years but have not seen this effort translate into tangible outcomes.

For instance, many firms have rolled out mandatory diversity training. While well-intentioned, there is little evidence that diversity training will actually change behaviour, and research in the US, in fact, found it had either no effect on the number of women in management positions, or even reduced it. This could be because people might resent being made to attend training and fail to take it seriously, or training might bring to mind unhelpful stereotypes which people then act upon. Unfortunately for employers, this means that investing in mandatory diversity training may not be the best use of time or money.

To help employers to focus on the most effective actions, the Behavioural Insights Team produced a set of evidence-based actions for employers, published by the Government Equalities Office last month. The guidance summarises actions employers can take to reduce their gender pay gaps and improve gender equality within the workplace.

Professor Iris Bohnet, the academic dean of Harvard Kennedy School and author of What Works: Gender Equality by Design commented: “Evidence-based design of hiring practices, promotion procedures and compensation schemes helps our organisations do the right and the smart thing, creating more inclusive and better workplaces. This guidance is an important step towards helping employers know what works.”

The actions are based on the best available evidence currently, and the guidance classifies actions into three categories — “effective”, “promising” or “mixed results”. We’ve included a description of each category, and an example or two, below.

Effective actions are those that have been tested in real-world settings and found to have a positive impact. They include the following:

  • Use structured interviews for recruitment and promotions. Evidence suggests that unstructured interviews do not accurately predict job performance, perhaps because they are more likely than structured interviews to allow bias to creep in and influence decisions. Employers should instead use structured interviews to select candidates. These involve asking all candidates the same questions in a predetermined order and format, and grading candidate responses using pre-defined standardised criteria.
  • Appoint diversity managers or diversity task forces. Diversity managers and task forces monitor talent management processes and diversity within an organisation. They can help reduce biased decisions in recruitment and promotion as decision-makers know that their decisions may be reviewed. Diversity managers should have a senior role within an organisation, and be able to ask for more information on why decisions were made.

Promising actions require further research to improve the evidence on their effectiveness and how best to implement them, including:

  • Offer mentoring and sponsorship. Some evidence from the US suggests that mentoring and sponsorship programs have a positive effect on the representation of black, Hispanic and Asian women in management positions. There is no current evidence on which option is more effective, or how best to run mentoring or sponsorship programs.

Mixed actions have been shown to have a positive impact sometimes and at other times a negative impact. Due to the mixed evidence, we cannot yet make a general recommendation that these are good ways to reduce gender inequality. Here’s one example:

  • Performance self-assessments. Evidence suggests that women underestimate their abilities or are more conservative when assessing their abilities than men. We do not yet know how these differences in self-assessment affect women’s progression at work.

We hope that our evidence-based actions will be a helpful resource for employers looking to improve their gender equality by implementing evidence-based actions. We will continue working with the Government Equalities Office to build the evidence on what works. — Shoshana Davidson

(Picture credit: Pexels)