Tools › Finance and Business › Human Resources › Adverse Impact Analysis › Example

A ratio under four-fifths that small numbers can produce by chance

ToolAdverse Impact Analysis
Sourceoriginal
LevelProfessional

Problem

An employer hired 12 of 30 applicants from Group A and 4 of 14 from Group B. Find each selection rate, the impact ratio under the four-fifths rule of thumb of 29 CFR 1607.4D, the difference in standard deviations, Fisher's exact test, and how many more Group B hires would reach the four-fifths line.

Inputs to Enter

  • Counts are people selected; reference group: the group with the highest rate; significance level 0.05
  • Group A: pool 30, count 12
  • Group B: pool 14, count 4
Open the Adverse Impact Analysis tool (loaded with this example)

Answers

  • Selection rates 12/30 = 40.0 % and 4/14 = 28.6 %; impact ratio 0.2857 / 0.4000 = 0.714, under four-fifths
  • Four-fifths line: 0.8 × 0.40 × 14 = 4.48 hires, so 5: one more than the 4 made (ratio 0.89)
  • Overall rate 16/44 = 0.3636; standard deviation of the difference 0.1557; Z = −0.734 standard deviations (normal p = 0.46)
  • Fisher's exact test: 5.1 expected, 4 observed; two-sided p = 0.52, one-sided p = 0.35

Notes

The tool's default inputs are this case; press Submit, or open /analysis-tools/business/hr/adverse-impact-analysis/?try=small-numbers. Illustrative numbers, no real employer; checked against an independent computation in exact rational arithmetic. Section 4D of the Guidelines says that greater differences may not constitute adverse impact where they are based on small numbers and are not statistically significant. Computed figures, not legal advice.

More Examples for This Tool