… a statistical test of selections against availability

avail calculates the statistical significance of disparities in an employment selection process when there is no applicant flow data, and protected class availability has been estimated from Census or other labor market data.

avail calculates probabilities using the Exact Binomial and the Normal Approximation to the Binomial tests. EEOSTAT now includes data from the Census EEO Tabulation (2014-2018 ACS data) to calculate labor market availability.

Overview of Process

To test whether a selection process shows evidence of discrimination, enter the Total Selected (the total number of people that were hired, fired, or promoted), the Protected Class Selected (the number of selections that went to members of the protected class under consideration), and the Class Availability (the proportion of the protected class in the labor market).

EEOSTAT now includes data from the Census EEO Tabulation 2014-2018 (5-year ACS data) to calculate labor market availability. Provide a title for the report, then click the Calculate button to have avail compute the results, reported in three boxes: Descriptive Statistics, Normal Approximation to the Binomial, and Exact Binomial. The results may be printed or saved.

In the example below …

A woman has alleged the ABC Co. refused to hire her because of her sex. Looking at the hiring data, the ABC Co. hired 245 drivers, 21 of whom were women. It has been estimated that 17 percent of the local labor market for drivers is female.

avail calculates that there is a shortfall of about 21 female hires (21.38), and this disparity is statistically significant. For the Exact Binomial two-tail test, the probability is 0.0001, or about one chance in 1000. As a rule, probabilities less than 0.05, or 5 in 100 are considered statistically significant.

— click image to expand