Tools › Finance and Business › Human Resources › Adverse Impact Analysis › Example
A ratio above four-fifths with a difference of almost six standard deviations
| Tool | Adverse Impact Analysis |
| Source | original |
| Level | Professional |
Problem
A written test was passed by 2,400 of 4,000 candidates from Group A and 1,590 of 3,000 from Group B. Find the impact ratio and test whether the difference in pass rates is within what chance produces.
Inputs to Enter
- Counts are people selected; reference group: the group with the highest rate; significance level 0.05
- Group A: pool 4000, count 2400
- Group B: pool 3000, count 1590
Answers
- Pass rates 60.0 % and 53.0 %; impact ratio 0.53 / 0.60 = 0.883, above four-fifths
- Overall rate 3,990/7,000 = 0.57; Z = −5.854 standard deviations (normal p = 4.8 × 10⁻⁹)
- Fisher's exact test: 1,710 expected, 1,590 observed; two-sided p = 5.5 × 10⁻⁹
Notes
Open this case in the tool: /analysis-tools/business/hr/adverse-impact-analysis/?try=large-numbers. Illustrative numbers; checked against an independent computation. Section 4D of the Guidelines says that smaller differences in selection rate may nevertheless constitute adverse impact where they are significant in both statistical and practical terms; the tool measures the statistical part only. Computed figures, not legal advice.