Adverse impact calculator
The four-fifths rule, with the small-cohort guards that decide whether a ratio is a finding or a coincidence.
The four-fifths rule compares each group’s selection rate to the highest group’s rate. A ratio below 0.8 is evidence of adverse impact requiring investigation. It is a screening heuristic from the US Uniform Guidelines, not a legal verdict: falling below it does not establish discrimination, and clearing it does not establish fairness.
Enter your numbers
One row per group, at a single stage of your process. Everything is computed in your browser and nothing is sent anywhere.
| Group | Applicants | Selected | Selection rate | Impact ratio | Remove |
|---|---|---|---|---|---|
| 25.0% | 0.80 | ||||
| 15.0% | 0.48 — below the four-fifths threshold | ||||
| 31.1% | 1.00 |
Highest selection rate: 31.1%. Every impact ratio above is that group’s rate divided by this one.
Total in this comparison: 245
1 group below the four-fifths threshold. That is evidence of adverse impact requiring investigation, not a finding of discrimination. The next questions are which stage produced it and whether the requirement driving it is genuinely job-related.
How to read the result
The impact ratio is one group’s selection rate divided by the highest group’s selection rate. If your strongest group is selected at 25 per cent and another at 17 per cent, the ratio is 0.68, which is below the threshold and worth looking into.
What it is telling you is that a neutral-seeming practice produced an unequal outcome. Intent is not part of it. The useful next questions are which stage produced the disparity, whether the requirement driving it is genuinely necessary for the role, and whether there is a less exclusionary way to test the same thing.
Measure one stage at a time. A ratio computed across a whole funnel routinely hides a single stage producing all of the effect, and the stage is what you can actually change.
Questions about the four-fifths rule
What is the four-fifths rule?
Does failing the four-fifths rule mean we are discriminating?
Does passing it mean we are safe?
Why does the calculator warn about small groups?
Is my data sent anywhere?
Which stage should I measure?
Read further
Adverse impact, bias audits, and where disparities actually come from.
The biases that actually change hiring outcomes
Awareness training does not fix bias. A guide to the biases that measurably move hiring decisions, and the process change that neutralises each one.
Read