Personnel selection Study Guide

📖 Core Concepts Personnel Selection – systematic process to hire or promote the individual most likely to succeed and add value. Job Analysis – mandatory first step; produces a job‑related list of KSAOs (knowledge, skills, abilities, and other characteristics). Bona Fide Occupational Qualification (BFOQ) – a narrowly defined job requirement that would be discriminatory if not truly essential (e.g., gender‑specific prison guards). Validity – evidence that a selection system predicts the intended job outcome. Types: Content – items reflect actual job tasks. Construct – test measures the theoretical trait it claims to assess. Criterion‑related – test scores correlate with a performance criterion (e.g., job performance). Predictor Validity (Predictive Validity) – how well a test differentiates successful from unsuccessful workers at a chosen cutoff. Selection Ratio (SR) – proportion of applicants who will be hired: $$ SR = \frac{n}{N} $$ where n = number of openings, N = number of applicants. Base Rate – expected proportion of all applicants who would perform satisfactorily if hired. Decision‑error Types – true/false positives & true/false negatives. 📌 Must Remember Structured interview validity ≈ 0.63; unstructured ≈ 0.20. Structured board interviews (consensus ratings) ≈ 0.64 validity. Situational interview questions ≈ 0.50 validity (highest among interview content types). Job‑related interview questions ≈ 0.39 validity; psychological interview questions ≈ 0.29 validity. Cognitive ability tests are the most valid general predictors for most occupations and for training performance. Cognitive tests often create adverse impact (lower scores for Hispanic & African‑American groups; higher for Asian‑American groups). Griggs v. Duke Power → business‑necessity justification required when adverse impact is present. Adding a personality test to a cognitive test does not reliably reduce adverse impact, though it raises incremental validity. Selection ratio ≥ 1 → selection device offers little discrimination value. Raising the cutoff lowers false positives but raises false negatives (and vice‑versa). 🔄 Key Processes Job Analysis → define KSAOs. Develop Selection Criteria (content, construct, criterion alignment). Choose Selection Devices (applications, interviews, cognitive/personality tests, work samples, etc.). Collect Evidence of Validity (content, construct, criterion‑related). Set Cutoff Score → decide acceptable balance of true/false positives/negatives. Apply Banding (if needed) → group similar scores, use secondary info to reduce adverse impact. Validate System → demonstrate statistical relationship between test scores and job performance. 🔍 Key Comparisons Structured vs. Unstructured Interviews Structured: standardized questions & scoring → higher validity (≈ 0.63). Unstructured: free‑form, variable → lower validity (≈ 0.20). Situational vs. Job‑Related vs. Psychological Interview Content Situational ≈ 0.50 > Job‑Related ≈ 0.39 > Psychological ≈ 0.29. Cognitive Ability Tests vs. Personality Tests Cognitive: highest overall predictive validity, strong generalizability. Personality: lower historical validity, privacy concerns, limited impact on adverse impact. True Positive vs. False Positive True Positive = passes test and performs well. False Positive = passes test but performs poorly. ⚠️ Common Misunderstandings Higher validity = no adverse impact. Validity and adverse impact are independent; a highly valid test can still be illegal if it disproportionately harms a protected group. Content validity guarantees job performance prediction. It only assures that items reflect job tasks; predictive power still requires criterion‑related evidence. Selection ratio of 1 means “no selection needed.” It means the pool is as large as the openings; the choice of device still matters for fairness and legal compliance. 🧠 Mental Models / Intuition Funnel Model: Job analysis → criteria → tools → funnel of applicants → cutoff → hires. Visualize each step narrowing the pool. Validity as Correlation: Think of validity coefficients (e.g., 0.63) as the strength of a correlation between test scores and later performance—higher numbers mean tighter “predictive thread.” Error Trade‑off Slider: Raising the cutoff slides a slider toward fewer false positives but more false negatives; lowering does the opposite. 🚩 Exceptions & Edge Cases BFOQ – permitted discrimination when the characteristic is truly essential to the job (e.g., gender‑specific prison guard). Adverse Impact – cognitive ability tests may be disallowed unless a business‑necessity justification is solidly documented (Griggs). Selection Ratio ≥ 1 – when openings equal/exceed applicants, any selection device adds little discriminative value; focus shifts to fairness and legal defensibility. 📍 When to Use Which Structured Interview – high‑stakes hiring, when consistency and legal defensibility are priorities. Situational Interview Questions – when you need the strongest interview predictive power. Cognitive Ability Test – jobs requiring high general mental ability or intensive training; also when you need a highly generalizable predictor. Personality Test – roles where specific traits (e.g., dependability) are critical and privacy concerns are manageable; useful as a supplemental predictor. Banding – when cognitive tests produce adverse impact and you want to mitigate it without discarding the test. 👀 Patterns to Recognize Validity ≈ 0.6–0.7 → structured, consensus‑rated board interviews. Validity ≈ 0.2–0.3 → unstructured or purely psychological interview items. Adverse impact flag → any cognitive test yielding large subgroup score gaps; expect a Griggs‑type business‑necessity question. Selection ratio near 0 → selection device will have a large impact on who gets hired; scrutinize validity and fairness. 🗂️ Exam Traps Confusing validity types – picking “content validity” when the question asks about correlation with performance (that's criterion‑related). Assuming “higher validity = no adverse impact.” Test‑wise, a high‑validity cognitive test can still be illegal if it shows disparate impact. Mix‑up of numbers: Structured interview validity ≈ 0.63 not 0.44 (the latter is a different study’s average). Selection ratio misinterpretation: Treating SR = 1 as “all applicants are hired.” In reality, SR = 1 means as many openings as applicants, but you still must select appropriately. Banding width error: Wider bands are more likely when reliability is low; the inverse relationship is easy to reverse‑read. --- Use this guide for a rapid, confidence‑building review right before your exam. Good luck!