Psychometrics · 10 min read

Competency Tests in Hiring: What Employers Need to Know

What does a competency test measure, which formats exist and how reliable and predictive is each one? With example questions, pitfalls, placement in the process and FAQ.

Door Ingmar van Maurik · Founder & CEO, Making Moves


Short answer

A competency test measures whether someone shows the behaviour a role requires, such as collaboration, customer focus or planning and organising. "Competency test" is an umbrella term: it can be a questionnaire, a situational judgement test, a competency-based interview or an assessment centre exercise. How reliable a competency test is depends entirely on that format. Behaviour-based formats (structured interview, work sample, SJT) predict job performance better than a questionnaire in which candidates rate themselves.

This article is part of our complete guide to assessments in hiring.

What is a competency test?

A competency is a combination of knowledge, skills and behaviour needed to do a task well. Examples are customer focus, collaboration, persuasiveness, planning and organising, and dealing with pressure. A competency test tries to establish how well a candidate masters those competencies.

That is a different question from a cognitive ability test. A cognitive ability test measures how well someone reasons. A competency test measures how someone behaves in situations that occur in the role.

The formats at a glance

FormatHow it worksStrengthWatch out

|---|---|---|---|

Competency questionnaireCandidate rates themselves on statementsFast, cheap, scalableSensitive to socially desirable answers [Situational judgement test](/en/articles/situational-judgement-test-hiring)Candidate picks the best response to work situationsBehaviour-based, scales wellNeeds good scenarios per role Competency-based interviewQuestions about concrete past behaviour (STAR)Strongly predictive when structuredRequires trained interviewers and fixed scoring Assessment centre exerciseRole play, in-tray, presentation or caseShows real behaviourExpensive and time-consuming, assessors must calibrate Work sampleA real task from the roleDirectly relevantMust be developed per role

How reliable is a competency test?

Reliability means: does the test give roughly the same result when repeated and across different assessors? This varies a lot by format.

  • Questionnaires often reach an internal consistency of 0.70 to 0.85 per scale. That is reasonable, but the score mainly tells you how someone sees themselves, or wants to be seen.
  • Interviews and exercises are scored by people. Reliability then depends on agreement between assessors. With fixed questions, behavioural anchors and trained assessors you usually reach 0.70 to 0.85. Without structure, agreement is low.
  • SJTs often measure several things at once and therefore have lower internal consistency (0.50 to 0.70), while still predicting reasonably well.
  • Every score contains measurement error. Never let small differences between candidates decide, and work with score bands. Read more in what makes an assessment valid and reliable.

    How well does a competency test predict performance?

    The large meta-analysis by Sackett, Zhang, Berry and Lievens (2022) shows that the way you measure matters a lot:

    MethodAverage validity (r)

    |---|---|

    Structured interview0.42 Work sample0.33 Situational judgement test0.26 Personality: conscientiousness0.19 Unstructured interview0.19

    So competencies are best measured through behaviour: what someone has done before (structured interview), what someone shows in a realistic task (work sample) or which choices someone makes in a work situation (SJT). A self-report questionnaire is mainly useful as input for the conversation, not as a selection criterion on its own.

    Example questions per format

    Questionnaire (customer focus): "I contact a customer myself when I suspect something has gone wrong." (strongly disagree to strongly agree)

    Competency-based interview (collaboration): "Tell me about a time you disagreed with a colleague about how to approach a project. What was the situation, what exactly did you do and what was the result?"

    SJT (planning and organising): "You have two deadlines today and your manager asks you to do an urgent job. What is the best response? A) You say no. B) You discuss which task gets priority. C) You do everything and work late. D) You ask a colleague to do the urgent job."

    Which competencies should you measure?

    The biggest mistake with competency tests is trying to measure too much at once. Stick to four to six competencies that demonstrably make a difference in the role.

    1. Start with a job analysis. What do the best employees do differently from average ones? Ask managers and look at concrete situations.

    2. Describe behavioural indicators. Not "is customer focused", but "summarises the customer's question before proposing a solution".

    3. Choose the best format per competency. Persuasiveness is better measured in a role play than in a questionnaire.

    4. Measure each competency twice where possible, for example in an SJT and in the interview. That increases reliability.

    Using it fairly and carefully

  • Explain what you measure. Candidates perform better and see the process as fairer when they know which competencies are assessed.
  • Use fixed scoring. The same questions, the same behavioural anchors and preferably two assessors.
  • Beware of cultural assumptions. What counts as "assertive" or "proactive" differs by background. Describe the behaviour concretely.
  • Let a human decide. A competency score is input for the conversation. Under Article 22 GDPR, a rejection may not be based solely on automated processing when it significantly affects the candidate.
  • Monitor outcomes per group, so you can see whether a component disadvantages certain candidates.
  • Where in the process?

  • Early: a short SJT or competency questionnaire to prioritise when volumes are high.
  • Middle: a competency-based interview with fixed questions and scoring.
  • Final stage: a work sample or assessment centre exercise for the last candidates, especially for leadership or commercial roles.
  • Which combination works per role is covered in the guide to assessments. How to set up the whole screening process is explained in the best candidate screening process.

    Off-the-shelf or custom?

    Off-the-shelf competency tests use generic competency models and generic situations. That is quick, but the situations often look little like your daily work and the norm group does not always fit your role. A custom competency test uses scenarios and behavioural indicators from your organisation and compares candidates with your own successful employees. See how we develop custom assessments.

    Frequently asked questions

    What is a competency test?

    A test that measures whether someone shows the behaviour a role requires, such as collaboration, customer focus or planning. It can be a questionnaire, an SJT, an interview or an exercise.

    How reliable is a competency test?

    It depends on the format. Structured interviews, work samples and SJTs with fixed scoring are reasonably to highly reliable and predict job performance best. Self-report questionnaires are sensitive to socially desirable answers.

    What is the difference between a competency test and a cognitive ability test?

    A cognitive ability test measures reasoning: how well someone thinks with words, numbers and patterns. A competency test measures behaviour in work situations. They complement each other.

    What is a competency-based interview?

    An interview in which you ask about concrete past behaviour, often using the STAR method: situation, task, action and result. With fixed questions and scoring it is one of the most predictive selection methods.

    How long does a competency test take?

    A questionnaire or SJT usually takes twenty to forty minutes. A competency-based interview about an hour. An assessment centre can take half a day to a full day.

    Can candidates fake a competency test?

    A questionnaire, partly: candidates can give the desired answer. With an SJT, a structured interview with follow-up questions or a work sample, that is much harder.

    Summary

  • A competency test measures behaviour in work situations, not reasoning ability.
  • "Competency test" is an umbrella term: reliability depends on the format.
  • Behaviour-based formats (structured interview, work sample, SJT) predict better than self-report.
  • Measure four to six competencies with concrete behavioural indicators and fixed scoring.
  • Use the result as input for the conversation and let a human decide.
  • Want a competency test with scenarios from your organisation? See how we develop custom assessments or get in touch.


    Book intake call · View our AI Hiring System