What is adverse impact, and what can be done to prevent it?

The most successful companies today realize that diversity and inclusion in hiring are critical to achieving cultural, productivity, performance, and even financial outcomes. In fact, recent McKinsey research found that “companies in the top quartile for racial and ethnic diversity are 35 percent more likely to have financial returns above their respective national industry medians.”

But to truly achieve fair and diverse recruitment, HR professionals must look beyond the surface. And that’s where the concept of the adverse impact comes into play.

In an HR context, the term ‘adverse impact’ is used to describe the effects of employment policies, procedures, or practices that appear fair on the surface but, in practice, are disadvantaging or discriminating against a protected group.

This could be, for example, setting an experience range in your job advertisement. While it might feel like you are being targeted by saying you require ‘3-5 years’ experience,’ you are also actively discouraging applications from, and therefore incidentally discriminating against, older workers.

  • Measuring adverse impact: The 80% rule

    In 1978, government departments in the U.S. came together to define the four-fifths or 80% rule. This rule essentially asks companies to compare the dominant hiring group with others in the talent pool. If the selection rate for any of the comparison groups is lower than 80% of the dominant group, the rule posits that adverse impact has taken place. In the U.S., this can be used by Federal Enforcement Agencies as evidence.

    While many other countries do not have specific laws prohibiting adverse impact, most offer legal avenues to challenge discriminatory employment practices under human rights and discrimination laws.

  • A general framework for measurement

    1. Select which factor to test from any parameters that define a protected class, including age, gender, sex, ethnicity, disability status, religious beliefs, etc.
    2. Determine the selection rate of each group. For the purposes of this example, we will test against age. In this case, we will use over and under 40 as our parameters.
    3. Identify the group with the highest selection rate. This is your dominant group.
    4. Test selection rates against the dominant and protected groups to calculate impact rates. In our example, if we hired ten people under 40, we would need to have also hired at least eight people over 40 to reach the 80% or 4/5 If we have hired less than eight, we have evidence of adverse impact.

    Of course, any measurement framework has challenges, exceptions, and potential issues.

    Some issues you may encounter with the 80% rule include:

    1. Small data sets can prove a problem, and the results can be misleading
    2. Candidates may not want to disclose their status within these protected groups and, in many cases, are not legally required to, which can lead to false identification of adverse impact.
  • Proactively eliminating adverse impact

    Of course, it’s important to identify adverse impact, but it’s better to put preventative measures in place before a recruitment process begins. Some of the areas to consider are:

    1. Ensuring your stated candidate requirements are 100% aligned with the job description. You should be able to draw a clear line between any criteria and the tasks listed in the formal job description. Where ambiguity exists, leave it off your essential criteria.
    2. Understanding the power of words. Word choice in advertisements and job descriptions can be very powerful and can unintentionally reduce your candidate pool. For example, ableist language choices that may seem neutral to you can discourage those with disability.
    3. Using the same formula and questions in every interview. While off-the-cuff discussions can be a great way to discover more about potential candidates, they create an uneven playing field as not every candidate is given the same opportunity.
    4. Harnessing an AI-powered recruitment automation platform to remove human bias from application to interview and shortlisting. Our unconscious biases can contribute to adverse impact. To make sure hiring is accurate and fair, look for a platform that is underpinned by organizational psychology principles and a vendor who can articulate how their platform eliminates bias.

Join our mailing list and stay updated on recruitment automation news & trends.

Transform your recruitment process, focus on what matters.

A unified AI platform constructed for recruiters, employers, businesses and people