Updated 2026-09-16
An evidence-cited score is an interview rating where each competency mark is linked to specific candidate evidence — a quoted answer span, observed behaviour, or work sample reference — rather than a standalone number. It makes scoring auditable, supports calibration discussions, and reduces halo-driven rubber-stamping on scorecards.
Scorecards with numbers but no notes fail under scrutiny — hiring managers cannot recall why a '3' was given, and candidates receive generic rejections. Evidence-cited scoring forces interviewers to tie ratings to observable content at decision time.
In AI-led interviews, citation means highlighting transcript segments that triggered each competency inference — enabling human reviewers to override model mistakes before adverse decisions.
Evidence links support structured hiring defensibility when selection procedures face job-relatedness questions — though this page is not legal advice.
Skills inference produces competency claims; evidence-cited scores document the basis for those claims. Together they form evidence-based hiring — decisions grounded in documented behaviour, not intuition alone.
Scoring calibration improves when teams debate evidence, not just numbers — 'you scored 4 on teamwork but cited no collaboration example' surfaces training gaps.
Many allow notes but do not require them. Structured interview programs often enforce notes via process, not software alone.
Transcript highlighting is feasible; accuracy varies. Human review remains important for high-stakes decisions.
Brief notes add seconds per question but save hours in debrief disputes and improve consistency over time.
Head back to Hiring glossary or start now.
· 3 free credits · pay per interview · nothing recurring