hr-candidate-assessment

name	hr-candidate-assessment
description	Help recruiters and hiring managers design and run structured candidate assessments beyond interviews — skills tests, work samples, assessment centers, and scoring rubrics. Use when asked to "design a candidate assessment", "build a skills test for [role]", "create an assessment center", "build a scoring rubric", or "choose the right assessment method for this role".
metadata	{"author":"Tuan Duc Tran","version":"1.0.0"}

Candidate assessment

Design and run structured assessments — skills tests, work samples, simulations, and assessment centers — that predict on-the-job performance more reliably than unstructured interviews alone.

Supported tasks

• Selecting the right assessment method for a given role and level
• Designing skills tests and work-sample exercises tied to real job tasks
• Building assessment centers combining multiple exercises for senior roles
• Creating scoring rubrics that reduce rater subjectivity
• Calibrating assessors before a hiring round to improve consistency
• Balancing assessment rigor against candidate time and drop-off risk
• Evaluating third-party assessment tools and vendors for fit and validity
• Designing assessments that reduce adverse impact and bias
• Combining assessment scores with interview data into a hiring decision
• Communicating assessment results and feedback to candidates
• Piloting a new assessment before rolling it out broadly
• Auditing existing assessments for predictive validity over time

Key prompts

Designing assessments

1. "Design a work-sample exercise for [role] that reflects a real task they'd do in the first 90 days."
2. "What assessment method — skills test, case study, simulation, or work sample — best fits [role] at [level]?"
3. "Build an assessment center agenda for a senior [role] combining a presentation, case exercise, and structured interview."
4. "Design a scoring rubric for [exercise] with clear behavioral anchors at each rating level."

Calibration and fairness

1. "Write an assessor calibration guide to reduce scoring inconsistency across interviewers for [assessment]."
2. "Review this assessment exercise for potential adverse impact or unnecessary barriers unrelated to job performance."
3. "How do we balance assessment thoroughness against candidate drop-off from a lengthy process?"
4. "How do we handle a case where two calibrated assessors give meaningfully different scores for the same candidate?"

Evaluating tools and results

1. "What criteria should we use to evaluate a third-party assessment vendor for [role type]?"
2. "How do we combine assessment scores with interview feedback into a single hiring recommendation?"
3. "Design a process for auditing whether our [role] assessment actually predicts on-the-job performance."
4. "What should we do if our assessment vendor cannot show validity evidence linking scores to job performance?"

Candidate-facing communication

1. "Draft candidate-facing instructions for [assessment exercise] that set clear expectations without giving away the evaluation criteria."
2. "Write feedback for a candidate who didn't pass the assessment stage, focused on constructive, specific points."
3. "How much detail about our assessment methodology should we share with candidates before they take it?"
4. "Draft a response to a candidate who requests a re-assessment after receiving a low score."

Tips

• Anchor every assessment exercise to a real job task — assessments disconnected from actual work predict poorly and frustrate candidates.
• Use structured rubrics with behavioral anchors, not just numeric scales, to materially reduce rater bias and inconsistency.
• Calibrate assessors together before scoring live candidates, especially for subjective exercises like presentations or case studies.
• Keep total assessment time proportional to role seniority and candidate market competitiveness — over-testing loses strong candidates in competitive markets.
• Periodically check whether assessment scores actually correlate with post-hire performance; an assessment that isn't predictive is just candidate friction.