ResumeWorld

Hiring operations4 min read

Pre-Employment Skills Assessment Tests: Types, Validity and Fair Use

Pre-employment skills tests add evidence when they match the job. See the main types, how to check validity, limits on length and how to use them fairly.

RWThe Resume World Team
4 min read
Pre-Employment Skills Assessment Tests: Types, Validity and Fair Use — Resume World

Pre-employment skills tests add evidence beyond the resume and the interview when they measure something the job really needs. The main types are knowledge tests, work samples, situational judgement tests and cognitive ability tests. Choose the one that most closely resembles the job, keep it short, tell candidates what to expect and how it will be scored and look at results by group to check for adverse impact. Ask any vendor for evidence that the test predicts performance in similar roles. A test is one input to a decision, and it works best alongside a structured interview and a scorecard. Used badly, it screens out good people and adds legal risk.

The goal is the same as with every other step: collect job-related evidence consistently.

Types of test

TypeWhat it measuresGood forWatch out for
Job knowledge testFacts and concepts needed in the roleRegulated or technical rolesMemorisation over skill
Work sampleDoing a task from the jobWriting, coding, analysis, designTime, scoring effort
Situational judgement testChoices in realistic scenariosCustomer, management, support rolesObvious "right" answers
Cognitive ability testGeneral reasoningComplex rolesAdverse impact risk; weaker if generic
Personality questionnairePreferences and styleLimited use in selectionWeak validity for many uses; can be faked
Language or typing testSpecific skillsRoles that need themApplying where not needed

For work samples that candidates do on their own time, see take-home assignments.

Check validity

A test is valid for hiring if its results relate to performance in the job. Ask:

  • What does it measure, and why does it matter for this role?
  • Is there evidence, such as studies in similar roles, that scores predict performance?
  • Was it developed and checked by qualified people?
  • How was it checked for differences between groups?
  • What are the reliability and norm data?

Treat claims with caution if the vendor cannot provide documentation. A test that "looks relevant" may still not predict performance.

Keep it proportionate

  • Short: 20 to 40 minutes for most screening tests, longer only at later stages.
  • Relevant: tied to the main requirements on your scorecard.
  • Clear: instructions, scoring and time limit stated.
  • Accessible: offer adjustments. See reasonable adjustments in hiring.

Do not use a test for the sake of it. If a resume and a short conversation answer the question, you do not need one.

Where it fits in the process

StageTest?
ApplicationA very short job-specific question set may be fine
After phone screenA skills test for roles where it matters
InterviewA live work sample or scenario
Final stageA longer work sample for finalists, paid

See hiring process steps. Using a test early on every applicant for a high-volume role can be efficient, but check pass rates by group.

Set pass marks carefully

Do not choose a cutoff at random. Look at how current high performers would score, review the distribution and set a pass mark that separates candidates who could do the job from those who could not. Check that it does not screen out a group disproportionately. If it does, test whether the skill is genuinely required and whether another method would work as well. See the four-fifths rule and AI hiring audit checklist.

Use scores as one input

Do not decide on a test alone. Combine it with the structured interview and the resume. If a strong candidate fails a test on a day with a poor connection, allow a retake. If a person passes but their interview shows gaps, trust the combined picture. Record how the test was used in the decision.

Candidate experience

Tell candidates why you are testing, how long it will take and what you will do with the results. Give feedback on request, and respect their time. See candidate experience. A test that feels arbitrary or demeaning harms your reputation.

Data and privacy

Test results are personal data. Tell candidates how they are stored and for how long, keep them only as needed and make sure your vendor is bound by appropriate terms. See GDPR and recruitment data and candidate data retention. If the test involves automated scoring that influences decisions, check your disclosure and audit duties. See candidate AI disclosure.

Built-in tests

Some screening platforms embed short tests in the application. Resume World supports multiple-choice assessments attached to an application, scored automatically and shown beside the resume score. This suits knowledge checks, such as product or procedure knowledge, but it does not replace work samples for skills that need doing, not recalling. See pre-employment assessments for details.

A short decision guide

  1. Name the skill the job cannot do without.
  2. Ask whether a test measures it directly.
  3. Choose the shortest valid test, or a work sample.
  4. Check it for fairness and adjustments.
  5. Set a pass mark by evidence.
  6. Combine with interviews and resume review.
Frequently Asked Questions

Common inquiries regarding this topic.

Short assessments given to candidates before hiring, to measure skills the job needs, such as knowledge tests, work samples, situational judgement tests and cognitive tests. They add evidence beyond the resume and interview when they are job-related and scored consistently.

RW

The Resume World Team

Verified

Product & hiring research, Resume World

About Engine

We build the screening engine behind Resume World. Everything here comes out of working on resume parsing, scoring and hiring workflows day to day — including the parts that turned out harder than expected.

Resume World Intelligence

See more than just keywords.

Resume World extracts verifiable evidence from every applicant against role criteria and delivers an explained, ranked shortlist. 100% free to start with zero card required.

Related Research

Keep reading in this cluster