Direct Answer
Adverse impact occurs when a hiring practice — a test, an interview, a reference check — results in significantly different selection rates for different groups of people, such as by gender, race, or ethnicity. It does not require intent to discriminate. A perfectly well-meaning hiring process can still produce adverse impact.
Why It Matters
Think about it this way: if a company uses a test to screen applicants, and 60% of one group passes but only 30% of another group does, something about that test is filtering people out at very different rates. That disparity — regardless of why it happens — is adverse impact.
This matters because fair hiring is not just an ethical goal. In many jurisdictions, including the United States and the European Union, employers can face legal challenges if their selection tools produce adverse impact and cannot be shown to be job-related and valid.
The Science Behind It
The most widely used measure of adverse impact is the four-fifths rule, established by the U.S. Equal Employment Opportunity Commission (1978). It works like this: if the selection rate for a minority group is less than 80% (four-fifths) of the selection rate for the majority group, there is evidence of adverse impact (Risavy & Hausdorf, 2011).
For example, if 50% of male applicants are hired and only 35% of female applicants, the adverse impact ratio is 35/50 = 0.70 — below the 0.80 threshold, indicating potential adverse impact.
Researchers also use Cohen’s d to quantify the magnitude of group differences on selection tool scores. A d of 1.00, for instance, represents a one standard deviation difference between groups — a level commonly observed between White and Black applicants on cognitive ability tests (Landers et al., 2023). Smaller values of d indicate smaller group differences and therefore lower risk of adverse impact.
Importantly, adverse impact varies by selection method. Cognitive ability tests tend to produce the largest group differences, while personality assessments and structured reference checks tend to produce much smaller ones. Fisher et al. (2022) found that structured employment references showed a gender difference of only d = 0.11 — well within the range considered negligible.
How This Connects to Better Hiring
Understanding adverse impact helps you evaluate whether your hiring tools are fair in practice — not just in intention. The goal is not to avoid assessment altogether, but to choose methods that are both valid and fair. Evidence suggests that structured approaches — whether in interviews, references, or assessments — tend to reduce adverse impact while maintaining or improving predictive accuracy (Fisher et al., 2022; Levashina et al., 2014).