Updated 2026-09-16
Evidence-based hiring applies selection methods backed by research on predictive validity and fairness — structured interviews, work samples, calibrated rubrics — rather than relying on unstructured conversations, pedigree heuristics, or interviewer intuition alone.
Evidence in hiring means documented links between assessment design and outcomes: meta-analytic validity coefficients for structured interviews, local validation studies showing score-to-performance correlation in your organisation, adverse-impact monitoring showing selection rates by group, and post-hire quality-of-hire tracking that feeds back into rubric design.
Anecdotes — 'our best engineer came from a non-target school' — inspire but do not generalise. Evidence-based teams treat hiring process changes like product experiments: hypothesis, measurement, iteration. When a new phone screen question set launches, they check whether downstream performance scores shift six months later.
This approach aligns with skills-based hiring philosophically but emphasises measurement discipline. Not every trendy assessment has evidence; gamified personality tests and unstructured 'culture fit' chats often rank poorly in validity research.
Job analysis
Define competencies with SME input — what distinguishes strong performers in this role here.
Choose methods with validity literature
Structured behavioural interviews, work samples, job knowledge tests where appropriate.
Standardise execution
Same questions, rubrics, scorecards — structured hiring across the requisition lifecycle.
Monitor outcomes
Track quality-of-hire, retention, adverse impact, and interviewer score variance.
Large firms run formal validation studies; smaller teams can still adopt published best practices — structured interviews, job analysis, scorecards — and track basic quality-of-hire metrics.
Predictive validity is one measurement concept within evidence-based hiring. The philosophy applies validity research plus fairness monitoring and continuous improvement.
Only if rubrics are job-linked, outcomes are tracked, and bias is monitored with the same rigour applied to human interview programmes.
Head back to Hiring glossary or start now.
· 3 free credits · pay per interview · nothing recurring