Critical Thinking Test Practice

What the real test usually looks like

The Watson-Glaser Critical Thinking Appraisal, published by Pearson TalentLens, is the best-known critical thinking test. Its publisher lists these details (TalentLens):

• 40 items in 30 to 40 minutes, timed, drawn from an item bank.
• Online, either unsupervised for screening or in a supervised setting, with results available at once.
• Five subtests: inference, recognizing assumptions, deduction, interpretation and evaluating arguments.
• For graduates, managers, supervisors, professionals and executives, in recruitment and development.

The RED model. Pearson's 2020 efficacy report explains that inference, deduction and interpretation have been reported together as "Draw Conclusions" since Watson-Glaser II in 2010. Each test taker gets 40 items: 12 on recognizing assumptions, 12 on evaluating arguments and 16 on drawing conclusions (Pearson).

Each section has its own answer format. In Pearson's practice booklet, inference items use five answers: true, probably true, insufficient data, probably false and false. Assumption items ask whether an assumption is made. Deduction asks whether a conclusion "necessarily follows". Interpretation asks whether it follows "beyond a reasonable doubt". Argument items ask whether an argument is strong or weak (Pearson practice booklet).

Where you may meet it. It is often an early filter. For example, a recent Clifford Chance listing for London training contracts puts the Watson Glaser test straight after the online application, and only those who pass have their application reviewed (Clifford Chance).

Question types with worked examples

One worked example for each Watson-Glaser-style section, plus the line most people find hardest.

1. Inference: probably true. Facts: a cafe's sales of hot drinks rose sharply in the week the temperature fell below freezing. Inference: the cold weather helped drive the rise. Answer: probably true. Common knowledge says people buy more hot drinks in cold weather, which the inference section allows you to use, but the facts do not rule out other causes, such as a promotion. So not definitely true.

2. Inference: insufficient data. Same facts. Inference: the cafe is in a city center. Nothing in the facts points either way, and common knowledge does not help. Answer: insufficient data. Trap: choosing "probably false" because the inference feels unlikely. Probably false needs some reason in the facts to doubt it.

3. Recognizing assumptions. Statement: "Let's run the training online so that more staff can attend." Proposed assumption: some staff who cannot attend in person could attend online. Made, because the plan only works if that is true. Proposed assumption: online training is better than training in person. Not made: the statement is about attendance, not quality. Trap: accepting anything the speaker would probably agree with.

4. Deduction. Statements: "All projects over budget are reviewed. Project Delta was reviewed." Conclusion: Project Delta was over budget. Does not follow: projects may be reviewed for other reasons. The rule runs one way only. Our syllogisms guide has more of these patterns.

5. Interpretation. Passage: staff turnover was 18% in 2024 and 12% in 2025; flexible hours were introduced at the start of 2025. Conclusion: turnover was lower in 2025 than in 2024. Follows beyond a reasonable doubt. Conclusion: flexible hours caused the fall. Does not follow: the timing fits, but the passage does not rule out other causes. Trap: reading cause into a sequence of events.

6. Evaluating arguments. Question: "Should the city make its buses free to ride?" Argument: "Yes; free buses would cut car traffic in the city center and reduce congestion." Strong: Pearson's booklet says a strong argument must be "both important and directly related to the question", and you treat the argument as true. Argument: "No; buses are sometimes late." Weak: lateness has nothing to do with whether fares should be charged.

How to prepare

1. Learn each section's rules before you practice for speed. Each section has a different answer format, and using one section's logic in another is the most common way to lose marks.
2. Know your pace. 40 items in 30 to 40 minutes is 45 to 60 seconds an item (TalentLens). Your invitation will give the exact time; practice to the shorter end.
3. Use common knowledge only in inference. Pearson's booklet allows "commonly accepted knowledge" for inference. In the other sections, treat the statements as true and use only what they say.
4. Master the probably true versus insufficient data line. Probably true needs the facts to point that way. Insufficient data means the facts give no basis at all.
5. Watch "some", "all", "only" and "most". These words decide most deduction items. "Some" means at least one, nothing more.
6. Judge arguments on relevance and importance, not on whether you agree. Treat every argument as true, then ask whether it bears directly on the question.
7. Take the free official practice test. TalentLens offers one, which helps you learn the instructions and screens (TalentLens).
8. Read the instructions on the day. Each section restates its rules. A few seconds there saves several wrong answers.

How scoring usually works

Raw score. Pearson's efficacy report explains that each of the 40 items is right or wrong, giving a raw score from 0 to 40, and that raw scores alone say little (Pearson).

Percentile against a norm group. The raw score is compared with a norm group. The same report gives an illustration: a raw score of 35 against its manager norm would be a percentile rank of 86, meaning the person scored as well as or better than about 86% of managers. That is the publisher's example, not a pass mark. The report says the Watson-Glaser comes with 18 norms built by Pearson, covering occupations such as accountant and engineer and levels from entry to executive, and employers can build their own.

Reports. TalentLens lists number correct, T-score, stanine and sten scores, and Profile, Development and Interview reports. Reports can show your overall result and a score for each of the three RED areas (TalentLens).

There is no universal pass mark. Each employer sets its own target for each role.

Our practice score is the number of our own questions you got right, with a section breakdown. It is not a Watson-Glaser score and does not predict a percentile.

Common mistakes on Watson-Glaser-style tests

• Using outside knowledge in deduction or interpretation. Only inference allows common knowledge.
• Treating "probably" as "definitely". True in the inference section means it follows beyond reasonable doubt. Most inferences that sound right are only probably true.
• Marking an assumption as made because it is reasonable. It must be something the statement relies on, not something the speaker might also believe.
• Rejecting a strong argument you disagree with. Treat the argument as true. If it is important and directly related, it is strong.
• Reading "some" as "most". "Some managers are engineers" says nothing about most managers.
• Rushing the inference section. It has five answers instead of two, so it needs a little more time per item. Budget for it.

How this practice compares with the real test

What we match. All five Watson-Glaser-style question types, each with its own answer format, and a pace close to the real 45 to 60 seconds per item. Every answer is explained in terms of the section's rule.

What we cannot match. Pearson's items and norms are its own. We cannot give you a percentile, and our questions are not taken from the Watson-Glaser. TalentLens's own free practice test is also not a live test, and TalentLens says its results cannot be scored meaningfully because the items have not been trialled like live questions (TalentLens).

How to use both. Take the official practice test once to learn the screens. Use our questions and full mocks to build accuracy on each section's rules and to practice at the real pace.

A one-week Watson-Glaser study plan

• Day 1: Take the free sample above. Note which sections cost you the most.
• Day 2: Inference only. Practice the five answers, especially probably true versus insufficient data.
• Day 3: Assumptions and arguments. Say out loud why each assumption is or is not needed.
• Day 4: Deduction and interpretation, with the syllogisms guide and the deductive reasoning examples.
• Day 5: A full timed mock. The 30-Day Pass includes full critical thinking mocks, each with explanations.
• Day 6: Take TalentLens's free practice test to learn its screens, then review your weakest section.
• Day 7: A second mock in the morning, then rest. Check your device and quiet room for test day.