Tools › Finance and Business › Human Resources › Adverse Impact Analysis › Example
A layoff where the retention ratio and the layoff-rate ratio read differently
| Tool | Adverse Impact Analysis |
| Source | original |
| Level | Professional |
Problem
A unit of 300 employees lays off 20 of the 200 in Group A and 22 of the 100 in Group B. Show how the four-fifths ratio is taken when the count is an adverse outcome, and test the difference.
Inputs to Enter
- Counts are people laid off or rejected; reference group: the group with the highest rate; significance level 0.05
- Group A: pool 200, count 20
- Group B: pool 100, count 22
Answers
- Retained 180 of 200 = 90.0 % and 78 of 100 = 78.0 %; impact ratio on the retention rates 0.78 / 0.90 = 0.867, above four-fifths
- Layoff rates 10.0 % and 22.0 %: Group B's is 2.20 times Group A's
- Z = −2.824 standard deviations (normal p = 0.0047); Fisher's exact test two-sided p = 0.0075, one-sided 0.0048
Notes
Open this case in the tool: /analysis-tools/business/hr/adverse-impact-analysis/?try=layoff. Illustrative numbers; checked against an independent computation. The ratio is taken on the rates of the favorable outcome, as Question 20 of the agencies' Questions and Answers does with a rule that disqualifies 10 % and 4 % (selection rates 90 % and 96 %). The two tests are the same whichever outcome is counted. Computed figures, not legal advice.