Chi-Squared Test

The chi-squared test (also written as chi-square test or χ² test) is a statistical hypothesis test used in bias auditing to determine whether differences in selection rates across demographic groups are statistically significant — that is, unlikely to be due to chance alone. The test compares observed selection outcomes to the expected outcomes under the null hypothesis that the selection process is independent of group membership. When sample sizes are sufficiently large (typically n > 40 with expected cell counts of at least 5), the chi-squared test provides reliable results. A p-value below 0.05 indicates the observed disparity is statistically significant. OnHirely uses chi-squared tests alongside the four-fifths rule and Fisher's exact test to provide a comprehensive assessment of whether AI hiring tools produce discriminatory outcomes.