Watson-Glaser example questions, worked

Each Watson-Glaser section asks a different question about a short passage: is an inference true, is something assumed, does a conclusion follow, or is an argument strong. Below are ten original examples, two from each of the five sections, each with the answer and the reason, then a free timed set of 20.

Last checked

Then try 20 on the clock

  • 20 questions
  • 15 minutes
  • Every answer explained

Free, all five sections, the reason behind every answer.

Start the free test

By the LifeScore Editorial Team · Updated · How we check facts · How the questions are made

What does a Watson-Glaser inference question look like?

A passage of facts, then a possible inference to rate on five points: True, Probably true, Insufficient data, Probably false or False. It is the only section where common knowledge counts, and only to judge how likely something is.

The rule: Take the passage as true. Decide whether the proposed inference is true, probably true, false or probably false, or whether there is not enough information. Use common knowledge only to judge how likely something is.

Example 1 · Inference

A research firm surveyed 2,000 UK adults online and found that 41% had read at least one printed book in the previous month, down from 47% when it asked the same question of a similar sample five years earlier. Among respondents aged 18 to 24, the figure rose from 38% to 44% over the same period. How to answer: Take the passage as true. Decide whether the proposed inference is true, probably true, false or probably false, or whether there is not enough information. Use common knowledge only to judge how likely something is.

Proposed inference: "Among respondents aged over 24, the share who had read a printed book fell between the two surveys."

  • A. True
  • B. Probably true
  • C. Insufficient data
  • D. Probably false
  • E. False
Answer A. The overall figure is a mix of the under-25 and over-24 figures. Earlier, a mix of 38% and the over-24 figure gave 47%, so the over-24 figure was above 47%. Now a mix of 44% and the over-24 figure gives 41%, so it is below 41%. It must have fallen.

Example 2 · Inference

In a study of 400 new employees at a logistics firm, those who completed an optional two-day induction course were 40% less likely to leave within their first six months than those who did not. The course was offered to every new employee, and about half chose to take it. The firm's HR director has since proposed making the course compulsory. How to answer: Take the passage as true. Decide whether the proposed inference is true, probably true, false or probably false, or whether there is not enough information. Use common knowledge only to judge how likely something is.

Proposed inference: "The HR director thinks the course helps the firm keep new employees."

  • A. True
  • B. Probably true
  • C. Insufficient data
  • D. Probably false
  • E. False
Answer B. Proposing to make the course compulsory straight after a result like this suggests the director believes it works. The passage does not state the director's reasons, so it is probable, not certain.

What does a recognition of assumptions question look like?

A statement, then a proposed assumption: is the speaker taking it for granted? The trap is marking anything sensible or true as an assumption. Ask instead: would the statement still make sense if this were false? If it would, the assumption is not made.

Example 1 · Recognising assumptions

Statement: "Our new library app will cut the queues at the front desk, because members will be able to renew their books from their phones." How to answer: An assumption is something taken for granted. Decide whether the person making the statement is taking the proposed assumption for granted.

Proposed assumption: "Every library member owns a smartphone."

  • A. Assumption made
  • B. Assumption not made
Answer B. The claim needs only some members to renew by phone. It does not depend on every member having one.

Example 2 · Recognising assumptions

Statement: "Because applications to our graduate scheme doubled this year, we will add an online critical thinking test before the interview stage." How to answer: An assumption is something taken for granted. Decide whether the person making the statement is taking the proposed assumption for granted.

Proposed assumption: "The current selection process would struggle to handle twice as many applicants without an extra step."

  • A. Assumption made
  • B. Assumption not made
Answer A. The doubling is given as the reason for adding a step, which only makes sense if the present process would not cope with the extra volume.

What does a Watson-Glaser deduction question look like?

Premises, then a conclusion that either must follow or does not. Treat the premises as true even when they are odd, and read "some" as "at least one". The two classic slips are reversing an "all" statement and reading a rule backwards.

Example 1 · Deduction

All the trainees in the London office passed the firm's critical thinking test. Some of the trainees in the London office studied history at university. How to answer: Take the statements as true, even if they clash with what you know. A conclusion follows only if it must be true whenever the statements are.

Proposed conclusion: "Everyone who passed the critical thinking test is a trainee in the London office."

  • A. Conclusion follows
  • B. Conclusion does not follow
Answer B. The statements say all London trainees passed, not that only London trainees passed. Reversing an "all" statement is the classic slip.

Example 2 · Deduction

Every project that overruns its budget is reported to the board. Project Kestrel was not reported to the board. How to answer: Take the statements as true, even if they clash with what you know. A conclusion follows only if it must be true whenever the statements are.

Proposed conclusion: "Project Kestrel did not overrun its budget."

  • A. Conclusion follows
  • B. Conclusion does not follow
Answer A. If Kestrel had overrun, it would have been reported. It was not reported, so it did not overrun.

What does an interpretation question look like?

A passage of evidence, often with figures, then a conclusion that must follow beyond reasonable doubt. Some conclusions are simple arithmetic on the figures given; the trap ones jump from two numbers rising together to one causing the other.

Example 1 · Interpretation

Over the past ten years, the number of registered cyclists in Fenford has risen every year, and so has the number of reported cycling accidents. The council opened the town's first protected cycle lane three years ago. How to answer: Take the passage as true. A conclusion follows only if it follows beyond reasonable doubt from the information given.

Proposed conclusion: "The number of accidents for each registered cyclist has risen over the ten years."

  • A. Conclusion follows
  • B. Conclusion does not follow
Answer B. Both numbers rose, but the passage does not say which rose faster, so the rate per cyclist could have risen, fallen or stayed the same.

Example 2 · Interpretation

In a trial, 300 volunteers with mild back pain were split at random into two equal groups. One group did a 20-minute stretching routine every morning for eight weeks; the other made no change. At eight weeks, 61% of the stretching group and 34% of the other group reported less pain than at the start. How to answer: Take the passage as true. A conclusion follows only if it follows beyond reasonable doubt from the information given.

Proposed conclusion: "More than 90 people in the stretching group reported less pain at eight weeks."

  • A. Conclusion follows
  • B. Conclusion does not follow
Answer A. Each group had 150 people. 61% of 150 is 91.5, and since a rounded 61% can only be 91 or 92 people, more than 90 reported less pain either way.

What does an evaluation of arguments question look like?

A yes-or-no question, then an argument to rate Strong or Weak. Take the argument as true. It is strong if it is important and bears directly on the question, whether or not you agree with it; weak if it is trivial, too general or about something else.

Example 1 · Evaluating arguments

Question: Should the company let staff work from home two days a week? How to answer: Treat each argument as true. A strong argument is important and bears directly on the question; a weak one is trivial or not directly related.

Argument: "Yes: the company's own six-month pilot found productivity unchanged and staff turnover lower among those who took part."

  • A. Strong argument
  • B. Weak argument
Answer A. Evidence from the company itself on productivity and turnover goes to the heart of the decision.

Example 2 · Evaluating arguments

Question: Should universities make a critical thinking module compulsory for all first-year students? How to answer: Treat each argument as true. A strong argument is important and bears directly on the question; a weak one is trivial or not directly related.

Argument: "No: students should not have to study anything they did not choose."

  • A. Strong argument
  • B. Weak argument
Answer B. This would rule out every compulsory course. It says nothing about this module in particular, so it does not bear directly on the question.

Which words decide most Watson-Glaser answers?

A handful of words carry most of the logic. Underline them as you read; the answer usually turns on one of them.

Our guidance, from the rules each section states.

  1. All, every

    Deduction

    No exceptions, but only in one direction: "all A are B" says nothing about every B

  2. Some

    Deduction

    At least one, possibly all

  3. Only, unless

    Deduction, assumptions

    A condition that cannot be skipped

  4. Probably, likely

    Inference

    More likely than not, not certain

  5. Caused, because, led to

    Interpretation

    A claim of cause, which figures moving together do not prove

  6. Should

    Assumptions, arguments

    A recommendation: look for what it takes for granted

Frequently asked questions

Are these real Watson-Glaser questions?

No. They are original questions written by LifeScore in the five Watson-Glaser question types. Pearson publishes a few official examples on the TalentLens site, which are worth reading too.

What is the difference between True and Probably true?

True means the inference follows beyond reasonable doubt from the facts given. Probably true means it is more likely than not, given the facts and common knowledge, but not certain.

What is the difference between deduction and interpretation?

Deduction is strict logic: a conclusion follows only if it must be true whenever the premises are. Interpretation asks whether a conclusion follows beyond reasonable doubt from evidence, often with figures.

Is a strong argument one I agree with?

No. A strong argument is important and directly related to the question, taken as true. You can disagree with it and still mark it strong.

Where can I practise more?

The free practice test gives you 20 timed questions across all five sections, drawn from the same bank as these examples, with every answer explained.

LifeScore is not affiliated with Pearson or TalentLens. Watson-Glaser is a trademark of Pearson. These are original practice questions in the five Watson-Glaser question types, not real Watson-Glaser questions, and our score is not a Watson-Glaser percentile. Reviewed 12 November 2026 by the LifeScore Editorial Team.