Cognitive Ability Tests in Hiring: What They Predict and How to Use Them Fairly
Umid Rahimov7 min read
Cognitive ability tests (also called aptitude or reasoning tests) are among the most studied tools in hiring psychology. They are also among the most misused. Used well, they help you find people who will learn the job quickly. Used badly, they become an arbitrary filter that rejects good candidates.
What does a cognitive ability test measure?
It measures how well someone processes new information: understanding, reasoning and solving problems they haven't seen before. Typical sections:
- Numerical reasoning: working with numbers, percentages, tables and charts.
- Verbal reasoning: understanding written information and drawing correct conclusions.
- Logical / abstract reasoning: spotting patterns and rules in unfamiliar material.
- Critical thinking: evaluating arguments, assumptions and evidence.
It does not measure knowledge of a profession, personality or motivation.
What does the research say?
For decades, cognitive ability was considered the single best predictor of job performance, largely based on a 1998 meta-analysis by Schmidt and Hunter. A 2022 re-analysis by Sackett and colleagues corrected some of the statistical assumptions behind those numbers. It found that cognitive ability is still a useful predictor, but a less dominant one. Structured interviews, job knowledge tests and work samples performed as well or better.
The practical conclusion for employers:
Cognitive tests are valuable, especially for complex roles and candidates without much work history. They work best combined with job-specific tests and a structured interview, not on their own.
When cognitive tests are most useful
- Graduate and intern hiring: no work history, so ability to learn is the main signal.
- Complex, fast-changing roles: analysts, product, consulting, engineering.
- Roles with long training periods: where fast learners pay back the investment sooner.
- High-volume screening: to keep interview time for candidates likely to succeed.
They add less for routine roles where a practical skills test tells you directly whether someone can do the work.
How to use them fairly
- Make sure the role really needs it. Document why reasoning ability matters for this job.
- Combine it with other measures. Never decide on a cognitive score alone.
- Avoid unrealistically high cut-offs. A sensible minimum plus other tests works better than "top 10% only", which can disproportionately exclude some groups without improving quality.
- Test in the candidate's working language. Otherwise you measure language, not reasoning.
- Explain the format up front and offer practice examples, so test-taking experience doesn't decide the result.
- Offer reasonable adjustments, such as extra time where a disability requires it.
Speed vs. accuracy
Most cognitive tests are timed. That's part of what they measure, but don't over-weight speed for roles where careful work matters more than fast work. Look at the accuracy of completed questions, not only the total.
Cognitive test vs. skills test
| Cognitive ability test | Skills test | |
|---|---|---|
| Measures | Ability to learn and reason | What the person can do today |
| Best for | Juniors, complex roles | Experienced hires, tool-heavy roles |
| Example | Numerical reasoning, critical thinking | Excel, SQL, accounting, customer support scenarios |
Most strong assessments use one cognitive test plus one or two skills tests.
How TheyTest supports this
TheyTest's library includes reasoning tests such as Critical Thinking (in an express and a standard version), alongside role-specific skills tests. You set the weight of each test in the overall score. Questions are randomised per candidate, and every session can be reviewed for integrity.
FAQ
Is a cognitive ability test the same as an IQ test?
They are related. Both measure reasoning, but hiring tests are shorter, focused on work-relevant reasoning, and should be interpreted alongside other job-related evidence.
Can cognitive tests be biased?
Any test can produce different average results between groups. Reduce the risk by testing only when the role needs it, using reasonable cut-offs, combining with other measures and testing in the candidate's working language.
How long should a cognitive test take?
Usually 10–25 minutes. Longer tests add little accuracy for screening purposes and increase drop-off.
