Sample Size Required for Adverse Impact Analysis

Abstract

Practitioners who rely on adverse impact analysis for the evaluation and development of selection procedures need to carefully consider the adequacy of the samples used to compute adverse impact statistics. A method is presented for estimating the minimum sample size needed to test for adverse impact with adequate statistical power. The required sample size is often extremely large, particularly for small effects when the selection rate and the proportion of minority applicants are small. However, smaller samples will be adequate when the magnitude of the group difference is large.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 25 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.