Watson Glaser Test: 5 Sections, Rules and Sample Questions

The Watson Glaser test is 40 critical thinking questions in 30 minutes across five sections. Learn each section's rules with worked sample questions.

The Watson Glaser test, formally the Watson-Glaser Critical Thinking Appraisal, is a multiple choice critical thinking assessment published by TalentLens (Pearson). The standard version is 40 questions in 30 minutes, split across five sections: Inference, Recognition of Assumptions, Deduction, Interpretation and Evaluation of Arguments. Longer versions of 80 questions in 60 minutes exist. It is used heavily in legal recruitment, including firms such as Linklaters, Clifford Chance and Hogan Lovells, and by the Government Legal Service. The test is not a general knowledge test and it is not a maths test. Every answer must come from the passage in front of you. The single biggest cause of low scores is importing what you already know about the topic instead of judging only what the text supports.

TL;DR

Standard format is 40 questions in 30 minutes, about 45 seconds per item.
Five sections, each with its own answer scale and its own rule.
Inference uses a five point scale: True, Probably True, Insufficient Data, Probably False, False.
Assumptions, Deduction and Interpretation are yes or no judgements with different strictness levels.
Evaluation of Arguments asks only if an argument is strong or weak, assuming it is true.
Answer strictly from the passage. Outside knowledge is allowed only in Inference, and only common knowledge.

What does each Watson Glaser section actually ask?

The sections look similar but the rules differ, and mixing them up is how candidates lose marks. This table is the summary worth memorising before you sit the test.

Section	Your task	Answer options	Strictness
Inference	Judge how likely a conclusion is	True, Probably True, Insufficient Data, Probably False, False	Common knowledge allowed
Recognition of Assumptions	Decide if the statement takes something for granted	Assumption made: Yes or No	Must be necessary to the statement
Deduction	Decide if a conclusion necessarily follows	Conclusion follows: Yes or No	Strict logic only
Interpretation	Decide if a conclusion follows beyond reasonable doubt	Conclusion follows: Yes or No	Slightly looser than Deduction
Evaluation of Arguments	Decide if an argument is strong or weak	Strong or Weak	Must be important and directly related

Before you drill Watson Glaser items, it helps to be comfortable with timed multiple choice reasoning in general. Our logical reasoning test practice page covers the wider family of tests, and if a numerical stage sits alongside your critical thinking test you can warm up with five free timed numerical questions.

Sample question 1: Inference

Statement. A city council installed 500 new cycle racks in 2024. Bicycle theft reports in the city rose by 12 percent in 2025. Reported cycle commuting rose by 30 percent in the same year.

Judge each inference.

More people cycling to work in the city led to more bicycles being available to steal.
The new cycle racks caused the rise in bicycle theft.
The council installed the racks in every district of the city.

Worked solution. Item 1 is Probably True. The facts give a 30 percent rise in cycle commuting and a 12 percent rise in thefts. It is common knowledge that more bicycles in use means more targets, so the inference is more likely true than false, but the facts do not prove it, so it is not True.

Item 2 is Insufficient Data. Nothing in the statement links the racks to theft levels. A correlation in time is not evidence of cause, and there is no basis for judging one way or the other.

Item 3 is Insufficient Data. The statement gives a total of 500 racks and says nothing about their distribution.

The trap in Inference is the gap between Probably True and Insufficient Data. Ask yourself: does ordinary common knowledge tilt this one way? If yes, use Probably True or Probably False. If nothing tilts it, use Insufficient Data.

Sample question 2: Recognition of Assumptions

Statement. “We should move the team meeting to Tuesday so that everyone can attend.”

Decide whether each assumption is made.

At least one person cannot attend the meeting on its current day.
Tuesday is the most productive day for the team.
Everyone is available on Tuesday.

Worked solution. Item 1: assumption made, Yes. The phrase “so that everyone can attend” only makes sense if someone currently cannot.

Item 2: assumption made, No. The statement is about attendance. Productivity is never referenced, so it is not taken for granted.

Item 3: assumption made, Yes. Moving the meeting to Tuesday to let everyone attend takes for granted that Tuesday works for everyone.

The rule: an assumption must be necessary for the statement to hold, not merely plausible alongside it.

Sample question 3: Deduction

Statement. All qualified auditors in the firm hold a practising certificate. Some staff in the firm are qualified auditors. Treat both statements as true without exception.

Decide whether each conclusion necessarily follows.

Some staff in the firm hold a practising certificate.
All staff who hold a practising certificate are qualified auditors.
No unqualified staff hold a practising certificate.

Worked solution. Conclusion 1: follows, Yes. Some staff are qualified auditors, and all qualified auditors hold a certificate, so those staff hold a certificate.

Conclusion 2: follows, No. This reverses the first premise. The premise says auditors imply certificates, not that certificates imply auditors.

Conclusion 3: follows, No. Nothing rules out non auditors holding a certificate for another reason.

In Deduction, “some” means at least one and possibly all. Reversal of a conditional is the most common wrong answer on the whole test.

Sample question 4: Interpretation

Statement. In a survey of 1,200 graduate applicants, 720 said they had taken an online aptitude test before applying. Of those 720, 480 said they had practised for less than one hour in total.

Decide whether each conclusion follows beyond reasonable doubt.

Most of the surveyed applicants had taken an online aptitude test before applying.
Two thirds of the applicants who had taken a test practised for less than one hour.
Practising for more than one hour improves test scores.

Worked solution. Conclusion 1: follows, Yes. 720 out of 1,200 is 60 percent, which is a majority of those surveyed.

Conclusion 2: follows, Yes. 480 divided by 720 equals two thirds exactly.

Conclusion 3: follows, No. The survey reports practice time only. It reports no scores, so nothing about improvement can be concluded.

Interpretation rewards a quick arithmetic check. Do the division before you judge, because roughly half of these items turn on a percentage or a fraction.

Sample question 5: Evaluation of Arguments

Question. Should employers replace all graduate interviews with online aptitude tests? Treat each argument as true and decide if it is strong or weak.

No. Interviews let an employer assess communication skills that a multiple choice test cannot measure.
Yes. Online tests are cheaper to administer than a full interview day.
No. Some candidates find online tests stressful.

Worked solution. Argument 1 is Strong. It is directly related to the question and it identifies something material that would be lost by the change.

Argument 2 is Strong. Cost of hiring is important to employers and it bears directly on the proposal.

Argument 3 is Weak. Stress is real but it is not directly related to whether tests should replace interviews, since interviews are stressful too. The argument is of minor importance to the question at issue.

For an argument to be strong it must be both important and directly related to the question. Being true is assumed and is never enough on its own.

How do I prepare in the week before the test?

Timing first, content second. At 40 questions in 30 minutes you have around 45 seconds per item, so you cannot afford to reread a passage three times. Practise reading the passage once carefully and each proposed conclusion twice.

Learn the five rule sets cold. Write the table above from memory until you can do it in two minutes. Under time pressure, applying an Interpretation rule to a Deduction item is the failure mode that costs whole blocks of marks.

Suppress your own opinion. The Evaluation of Arguments section is designed to reward candidates who can judge an argument they disagree with. If a topic makes you react, that is a signal to slow down for five seconds, not to answer faster.

If your employer pairs the Watson Glaser with a verbal reasoning stage, the same passage discipline applies, and our SHL verbal reasoning test guide covers that format. For overall process planning across several tests in one application, read how to pass online assessment tests. If your employer uses a general cognitive test instead, see the CCAT test guide.

Frequently Asked Questions

How long is the Watson Glaser test?

The standard version is 40 questions in 30 minutes. Some editions run 80 questions in 60 minutes. Your invitation email should state which one you are sitting.

What is a good Watson Glaser score?

There is no universal pass mark. Employers set their own threshold and compare you to a norm group, so the same raw score can clear one firm and miss another.

Can I use outside knowledge?

Only in the Inference section, and only widely shared common knowledge. Deduction, Interpretation, Assumptions and Evaluation of Arguments must be answered from the passage alone.

Is the Watson Glaser only for lawyers?

No. It is most common in legal recruitment, but it is also used for professional and managerial selection in other sectors.

Can I retake it?

That is the employer’s decision, not the publisher’s. Many firms allow one attempt per application cycle. Ask the recruiter before you sit it.