Are IQ Tests Legal for Hiring?

The short answer is yes, with a condition attached, and the condition is doing an enormous amount of work. In the United States a cognitive ability test is lawful in hiring if it is job-related and consistent with business necessity.

That phrase is not a vague standard someone invented. It comes from a case decided in 1971, and understanding that case explains almost everything about how employment testing works today.

Griggs v Duke Power

Duke Power required a high school diploma and passing scores on standardised tests for employees seeking transfers into better-paid departments. Black employees challenged the requirements, which disproportionately excluded them.

In 1971 the Supreme Court ruled unanimously against the company, and the decision changed employment law in two ways that still govern testing.

First, intent stopped being necessary. Before Griggs, a claimant had to prove the employer meant to discriminate. After it, they needed only to show that a neutral practice produced discriminatory effects. That is the doctrine of disparate impact, and it is the reason employment testing is regulated at all.

Second, the Court said what a test has to do. The often-quoted line is that what Congress required is that any test used “must measure the person for the job and not the person in the abstract.”

That sentence is the whole problem with using an IQ test in hiring. A general cognitive ability test is, almost by definition, a measure of the person in the abstract. Making it lawful means demonstrating it predicts performance in this job.

What job-related actually requires

If a selection procedure produces adverse impact, the burden shifts to the employer to justify it. In practice that means validation evidence: documentation that the test predicts performance in the role it is being used for.

This is why the industry looks the way it does. It explains:

• Why commercial tests come with technical manuals. Validation evidence is the product as much as the questions are.
• Why tests are marketed by role. Validating for “sales roles” or “technical roles” is defensible. Validating for “intelligence” is not.
• Why situational judgement tests proliferated. A test built from scenarios in the actual job has job-relatedness on its face, in a way a matrix reasoning puzzle does not.
• Why nobody calls it an IQ test. The label invites exactly the challenge the framework was designed to permit.

The naming convention is not evasion so much as accuracy. What employers use genuinely is not a clinical IQ assessment. It is a narrower, role-tuned instrument built to survive a legal standard.

The four-fifths rule

How does anyone know whether adverse impact exists? The EEOC uses a rule of thumb known as the four-fifths or 80% rule.

Compare selection rates across groups. If a protected group is selected at less than 80% of the rate of the highest-selected group, that signals potential adverse impact, and the employer may be asked to produce validation evidence.

Two things about it are widely misunderstood. It is not a law, it is an administrative guideline used by the EEOC and by courts as a screening heuristic. And falling below it does not prove discrimination. It raises a question that the employer then answers with evidence.

This matters for cognitive testing specifically, because general ability tests have historically shown group differences in scores. That is the reason the legal framework bites harder here than on, say, a typing test.

What employers actually use

Cognitive testing in hiring is real but narrower than the search volume implies. Where it appears, it is usually:

• The Wonderlic, in commercial use for over seventy years across many industries and validated across a range of job categories. It is also the test at the centre of the police case where an applicant was rejected for scoring too high.
• Role-specific aptitude batteries, particularly in aviation, the military and technical fields
• Situational judgement tests, now very common in graduate recruitment
• Structured interviews, which on current evidence predict job performance better than cognitive tests do

That last point is worth sitting with. Sackett and colleagues (2022) revised the correlation between cognitive ability and job performance down to about 0.31, placing it below structured interviews at 0.42. The best-predicting tool in the modern evidence is also the one with the least legal exposure, which is a rare alignment between what works and what is safe.

The UK position

Different route, similar destination. There is no Griggs, but the Equality Act 2010 prohibits indirect discrimination: a provision, criterion or practice that applies to everyone but puts people sharing a protected characteristic at a particular disadvantage, unless it is a proportionate means of achieving a legitimate aim.

Applied to testing, that means cognitive tests are lawful provided they are:

• Applied consistently to all candidates
• Checked for adverse impact against protected characteristics
• Accompanied by reasonable adjustments for disabled applicants, which is a positive duty rather than an option

The reasonable adjustments duty is the part employers most often get wrong. Extra time, alternative formats or a different assessment method may all be required, and a timed test administered without considering adjustments is a common source of complaint.

Algorithmic hiring gets the same rules

Worth knowing given how much hiring now runs through automated screening.

The EEOC has issued guidance confirming that selection procedures using artificial intelligence must comply with Title VII in exactly the same way as any other test. An algorithm that produces adverse impact faces the same job-relatedness requirement as a paper aptitude test from 1971.

And the employer remains responsible, regardless of who built the tool. Buying a screening product does not transfer the legal exposure to the vendor, which is a detail a lot of organisations discover late.

If you are asked to sit one

• It is probably lawful. Employers using validated instruments consistently are on solid ground, and most are.
• Ask for adjustments if you need them. In the UK this is a legal duty on the employer, and in the US the ADA imposes a comparable requirement. You do not have to justify the request beyond establishing the need.
• It is not measuring your IQ. It is a role-tuned instrument, usually coachable, and practice materials for the common ones are widely available.
• It is one input. On current evidence cognitive tests predict performance less well than the structured interview you will probably also sit.

Frequently asked questions

The bottom line

Cognitive ability tests are legal in hiring, conditionally, and the condition traces back to a unanimous 1971 Supreme Court decision holding that a test must measure the person for the job rather than the person in the abstract. That single sentence is why employers validate tests by role, why situational judgement tests spread, and why almost nobody calls what they use an IQ test.

The UK arrives at a similar place through indirect discrimination under the Equality Act 2010, with a positive duty to make reasonable adjustments. And algorithmic screening tools sit under exactly the same framework, with the employer carrying the responsibility rather than the vendor.