structured-interviewing

# Structured Interviewing ## Why It Matters The evidence on selection methods is unambiguous (Schmidt & Hunter, 1998; updated Schmidt, Oh, & Shaffer, 2016). Validity coefficients for predicting job performance: - Unstructured interview: r ≈ .20–.38 - Structured interview: r ≈ .46–.58 - Work sample / job audition: r ≈ .44–.54 - General mental ability: r ≈ .51–.65 (with disparate impact concerns) Across thousands of hires, the difference between structured and unstructured methods is enormous in mis-hire cost, ramp time, and disparate impact. ## Six Components of a Structured System ### 1. Job Analysis - **Outcomes**: what must this person produce in 12 months? - **Competencies**: 4–7 capabilities required - **Counter-indicators**: what disqualifies (especially for senior roles) - **Stage fit**: what differs vs. the same role at another stage ### 2. Loop Design - 5–7 stages typical for senior IC; 4–6 for mid-level - Each interviewer assigned 1–2 specific competencies - Non-redundant coverage - Total candidate time bounded (typically 4–8 hours including any take-home) ### 3. Scorecards - Filled out *before* debrief - Specific evidence cited - 4-point scale with anchored descriptions - One independent rating per interviewer ### 4. Question Bank - Behavioral questions ("Tell me about a time when...") - Probe with STAR (Situation, Task, Action, Result; emphasize *Action*) - Anchored ratings: what does 5/5 sound like? 1/5? - Calibrated periodically ### 5. Work Samples - Representative of actual job, not contrived - Time-bounded (60–120 min onsite, capped take-homes) - Compensated when meaningfully extensive (3+ hours) - Same prompt for every candidate - Multi-evaluator review ### 6. Calibrated Debrief - Scorecards in before debrief - Each interviewer presents evidence, not just verdict - Decision criteria stated up-front - Hiring manager owns the call; bar raisers / senior leaders may have veto ## Common Behavioral Question Anchors For "leadership at senior IC": - 5: Has built and reorganized teams; concrete frameworks for performance management; specific examples of difficult feedback delivered well - 3: Mentored ICs; mixed examples; clear views on hiring - 1: No mentorship experience; weak feedback philosophy; couldn't give concrete examples For "judgment under ambiguity": - 5: Walked through a recent ambiguous decision with explicit consideration of tradeoffs and evidence; admitted what they didn't know; followed up post-decision - 3: Articulated the decision but glossed over tradeoffs; some hindsight reasoning - 1: "I just went with my gut"; unable to articulate decision process ## Avoid - Brain teasers (no validity for most roles; high bias) - Same questions every interviewer (redundant; coverage gaps) - "Conversational" interviews - Verdict-first scorecards - "Culture fit" without operationalization - Long unpaid take-homes (>3 hours) — deters underrepresented candidates ## Hiring for Culture-Add, Not Culture-Fit "Culture fit" correlates with homogeneity and weak performance prediction. Replace with two distinct concepts: - **Values alignment**: behavioral evidence of operating consistent with stated company values - **Culture-add**: complementary perspective, lived experience, or strength absent from the existing team ## Funnel Audit Quarterly, by stage and demographic: - Application → recruiter screen → onsite → offer → accept - Quality-of-hire: 90-day rating, retention 12/24 months - Demographic conversion at each stage Drop-offs at a single stage indicate process problem at that stage. Demographic drop-offs are bias signal — investigate cause. ## Key References - Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology. *Psychological Bulletin*, 124(2), 262–274. - Schmidt, F. L., Oh, I.-S., & Shaffer, J. A. (2016). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 100 years of research findings. - Bock, L. (2015). *Work Rules!* Twelve. - Bohnet, I. (2016). *What Works: Gender Equality by Design*. Harvard University Press. - Smart, G., & Street, R. (2008). *Who: The A Method for Hiring*. Ballantine. ## Pitfalls - Building structure without training interviewers - Rubrics on paper that nobody uses in debrief - "Bar raisers" without accountability - Skipping work samples for senior roles where they're most valuable - Audit data not segmented by demographic