Verbal reasoning tests: what they measure and when to use them

Verbal reasoning tests are a staple of graduate, professional and managerial selection. They are quick to administer and well supported by research, but they are only useful when the role genuinely depends on reading and reasoning with written information. Here is what they measure and when to use them.

What a verbal reasoning test measures

A verbal reasoning test asks candidates to read a passage of written information and judge what follows from it. The skill being tested is not vocabulary or general knowledge but disciplined reading: separating what a text actually says from what it implies, what it leaves open, and what a reader might assume.

Carroll's map of human abilities places language comprehension under crystallised ability, built up through education and experience, and reasoning about what logically follows under fluid ability; verbal reasoning tests typically draw on both.^[1] Different tests weight these differently. A verbal comprehension test leans towards accurate understanding of written material, while a verbal reasoning test or a more demanding verbal analysis test places greater weight on inference and evaluation of arguments.

How the True/False/Cannot tell format works

The most familiar format presents a passage followed by statements. For each, the candidate decides whether it is True, False or Cannot tell on the basis of the passage alone. "True" means the passage confirms the statement; "False" means the passage contradicts it; "Cannot tell" means the passage does not give enough information either way.

The third option is what makes the format demanding. Many plausible statements are neither confirmed nor contradicted, and candidates who answer from general knowledge or common sense, rather than from the text, tend to choose True or False when the right answer is Cannot tell. This mirrors a well-established finding in the psychology of reasoning: Evans, Barston and Pollard showed that people's prior beliefs pull their judgements away from strict logic, especially when a conclusion is believable but does not actually follow.^[2] Workplaces need people who resist that pull when reading contracts, policies, reports or evidence.

For employers, the format has a practical advantage: it reflects how written information is used at work. Few jobs ask people to define words, but many ask them to read a document and decide what it does and does not commit the organisation to. Good items are written so that the correct answer depends only on the passage, with statements carefully balanced across the three options so that guessing or a habit of always choosing one answer does not pay.

What the research shows

Verbal reasoning is a facet of general mental ability, which Schmidt and Hunter identified as one of the strongest single predictors of job performance and training success.^[3] A UK meta-analysis by Bertua, Anderson and Salgado included verbal tests among the specific ability measures and found that both general and specific ability tests predicted job performance and training success, with higher validity in more complex occupations.^[4]

As our article on ability tests in recruitment explains, Sackett and colleagues have since shown that the classic estimates were inflated by overcorrection for range restriction.^[5] Ability tests still predict, but more modestly than once believed, so a verbal score is best treated as one strong piece of evidence rather than a verdict.

When to use a verbal reasoning test

Use one when job analysis shows that people must regularly read, interpret and act on written information with consequences for getting it wrong. Legal, compliance, policy, HR, management, consulting and many graduate roles fit that profile, as do safety-critical roles where procedures and instructions must be followed exactly.

Be more cautious where written material is peripheral to the job. A verbal test for a role that is mainly practical or numerical adds little predictive value and can disadvantage candidates for reasons unrelated to the work. Under section 19 of the Equality Act 2010, a practice that puts a protected group at a particular disadvantage must be shown to be a proportionate means of achieving a legitimate aim; a clear link to job requirements is the foundation of that case.^[6]

Fairness and language

Verbal tests are sensitive to language background. A candidate who reasons well but is working in a second language may score lower than their underlying ability, particularly under time pressure. Where you assess candidates internationally or in more than one language, the ITC Guidelines for Translating and Adapting Tests set out the steps needed for scores in different languages to be comparable; simple translation is not enough.^[7]

Other practical points apply too. Candidates with dyslexia or visual impairments may need extra time or alternative formats, and our guide to reasonable adjustments covers how to handle these consistently. Compare scores with a relevant norm group and interpret them in bands from Very Low to Very High alongside interview and other evidence, rather than as a single cut-off.