Watson Glaser Test Explained: What It Is and How to Actually Pass It (2026)

If a law firm, government department or professional services firm has just sent you a Watson Glaser invitation, here is the short version: it is a 30-minute, 40-question critical thinking test, it is probably the single most decisive online hurdle in your application, and it is very learnable, but not by revising harder. It is passed by understanding a handful of rules about how the test wants you to think, rules that are genuinely counter-intuitive the first time you meet them.

This guide explains what the Watson Glaser measures, walks through all five sections with the trap in each one, and finishes with a preparation plan that fits in a week. It pairs with our companion piece on which UK firms use the Watson Glaser and what they test; this one is about the test itself.

What the Watson Glaser actually is

The Watson Glaser Critical Thinking Appraisal is one of the oldest psychometric tests still in mainstream use; it has been refined for nearly a century and is currently published by Pearson TalentLens. The modern online version most UK candidates sit runs to roughly 40 questions in around 30 minutes, drawn from a question bank, and is typically sat unsupervised online, sometimes re-verified with a shorter supervised version later in the process.

Employers use it because the thing it measures maps almost exactly onto legal and analytical work: can you separate what a text actually establishes from what it merely suggests, and both of those from what you happen to believe? Pearson describes the underlying skillset as the RED model: Recognise assumptions, Evaluate arguments, Draw conclusions.

The score that matters is usually a percentile against a norm group of other graduates or trainee applicants, and firms set their own cut-offs. You will rarely be told the threshold. What you can control is straightforward: candidates who understand the five question types outperform candidates who walk in cold, reliably and by a wide margin.

The five sections, and the trap in each one

1. Inferences

You get a short factual passage, then statements to rate on a five-point scale: true, probably true, insufficient data, probably false, false. The trap is the scale itself. Almost everything you want to call "true" is merely probably true, because "true" is reserved for statements the passage establishes beyond reasonable doubt. Meanwhile "insufficient data" is a correct answer far more often than instinct suggests. Calibrate hard against your own confidence: the question is not "is this sensible?" but "how far does this exact text carry me?"

2. Recognition of Assumptions

A statement is followed by proposed assumptions; you mark whether each one is genuinely presupposed by the statement. The trap: an assumption is not "something the speaker would probably agree with," it is something the statement cannot stand without. "We should hire more engineers to ship faster" assumes engineers affect shipping speed. It does not assume engineers are hard to find, however reasonable that sounds. Ask: if this assumption were false, would the original statement collapse? If it survives, it was not assumed.

3. Deductions

Premises, then a conclusion; you say whether the conclusion follows. Follows means logically forced: if the premises are true, the conclusion cannot possibly be false. Plausible, likely and sensible are all traps, and the traps run in both directions: conclusions that sound like a stretch sometimes follow with iron necessity, while conclusions that sound obviously right often do not follow at all. Work the logic mechanically rather than by feel, and treat the premises as a sealed world, even when they are plainly false in real life. "All cats are green; Rex is a cat" forces "Rex is green," and the test expects you to say so.

4. Interpretation

Similar shape to Deductions, but the standard softens: does the conclusion follow beyond reasonable doubt from the passage? You are no longer looking for iron necessity, just for whether the reading is properly supported rather than merely available. Most candidates find this the gentlest section, provided they resist importing outside knowledge.

5. Evaluation of Arguments

A question ("Should the UK do X?"), then arguments for or against, each to be rated strong or weak. The trap is agreement. A strong argument is one that is both directly relevant to the question and important in scale, regardless of which side it argues or whether you personally buy it. A weak argument is trivial, tangential or merely rhetorical, however much you agree with its conclusion. Candidates with strong opinions on the topic systematically underrate the other side's good arguments; the test is counting on that.

How to actually pass it

Learn the rules before you practise. One untimed pass through each section type, checking every answer against the explanation, teaches you the marking logic. Practising before you understand the five traps just rehearses your mistakes.

Then practise timed, in section blocks. Forty questions in thirty minutes is about 45 seconds each. The time pressure is mild once the rules are automatic and brutal while you are still translating. Two or three timed sets is where most people's scores jump.

Build the "page only" reflex. The single highest-value habit across all five sections is catching yourself importing outside knowledge. When a passage says something you know is false, the test is deliberately checking whether you can reason inside it anyway.

Keep a miss log. Watson Glaser errors are unusually diagnosable: nearly every wrong answer is one of the five traps above. Two lines per miss (which trap, what the tell was) turns a practice set into a syllabus.

Sit it fresh, not last. It is a concentration test as much as a reasoning test. Do it early in the day, on a laptop, with the phone elsewhere, and never at the end of a deadline all-nighter. Test invitations usually arrive within days of applying and carry one-week deadlines, so bank your practice before you submit the application, not after the email lands.

The Watson Glaser feels alien for about two days of practice, and then it clicks, permanently. Put in the two days.

Frequently asked questions

What is the Watson Glaser test used for? It screens candidates for roles built on critical reasoning, most famously training contracts and vacation schemes at UK law firms, along with the Government Legal Profession and various professional services employers. It sits early in the application process and typically operates as a hard filter: score below the firm's cut-off percentile and the application usually stops there, regardless of the rest of your form.

How long is the Watson Glaser test and how many questions does it have? The standard online version most UK applicants sit contains around 40 questions with a time limit of about 30 minutes, split across five sections: Inferences, Recognition of Assumptions, Deductions, Interpretation and Evaluation of Arguments. Exact counts vary slightly by version because questions are drawn from a bank. That works out at roughly 45 seconds per question, which is comfortable once the question types are familiar.

What is a good score on the Watson Glaser? Scores are usually reported as a percentile against a comparison group of similar applicants, and each employer sets its own pass mark, which is almost never published. Competitive law firms are generally understood to expect comfortably above-median performance, so treat the 70th percentile as a working target rather than a guarantee. Since you cannot see the bar, the practical strategy is the same for everyone: practise until the five question types feel mechanical.

Can you prepare for the Watson Glaser or is it an IQ test? It is very preparable, and that surprises people. The test measures a style of disciplined reading and reasoning, not general knowledge or raw IQ, and each of its five sections has explicit rules that untrained intuition gets wrong. Learning those rules and running a few timed practice sets typically produces a substantial score improvement, which is precisely why firms assume serious candidates have prepared.