632.3 Adequacy of sample size and statistical evidence

The Complainant's use of the employer’s data regarding the ages of people who had been hired for a custodial maintenance position during a seven-year period was insufficient to establish age discrimination since the statistical sample was too small to be of any significance and the Complainant failed to provide evidence of the number and distribution of older persons in the applicant pool. Kurtz v. Sch. Dist. of St. Croix Falls (LIRC, 06/10/93).

A Complainant’s attempt to establish disparate impact based upon two promotional postings failed because the numbers involved were simply too small to allow a conclusion to be drawn with any degree of confidence in its statistical significance. Moncrief v. Gardner Baking (LIRC, 07/01/92).

Where the effect of a challenged practice on two blacks, or at most seven minorities, was compared to its effect on 82 whites, it was concluded that the sample size was too small to allow a meaningful conclusion that there was a disparate impact. Davis v. City of Milwaukee (LIRC, 09/05/86).

Although an employee presented evidence that, in one year, 9% of the workforce was black (23 out of 253 hourly employees) and 57% of those discharged were black (eight out of 14), the actual number of black workers discharged is too small to draw any statistically significant conclusions. Story v. Massey-Ferguson (LIRC, 11/06/81).

To establish a prima facie case of disparate impact, a female job applicant must show that the hiring procedure has a substantial adverse impact on women, i.e., that the hiring process selects significantly more males than females when compared to the number that apply. However, the employer's lack of female police officers does not establish a prima facie case under a disparate impact theory where the size of the police force is small, the frequency of vacant positions is minimal and the number of female applicants is low. Tall v. City Council of Shullsburg (LIRC, 05/13/80).