Watson Glaser Critical Thinking Test — Complete Guide

The Watson Glaser Critical Thinking Test (often called the Watson Glaser Assessment or WGA) measures your ability to think logically, analyse information, evaluate arguments, and draw sound conclusions. It's used heavily by law firms (for training contract and graduate recruitment), consulting firms, and accounting firms. The test assumes no specific knowledge — it measures your thinking style, not your knowledge of law, business, or any particular field.

Firms use Watson Glaser because it predicts success in their graduate roles better than general interviews. Lawyers need to analyse cases and judge arguments. Consultants need to extract insights from data. Accountants need to spot logical flaws. Watson Glaser assesses these capabilities objectively.

The test is challenging because it operates at a meta-level: you're not solving problems, you're assessing the quality of reasoning. For example, a question might ask: "Is the following argument strong or weak?" You must evaluate whether the logic is sound, not whether you agree with the conclusion. Many candidates struggle because they conflate logical strength with agreement.

Score interpretation: there's no fixed "passing" score. Instead, your score is compared to other candidates. Law firms might use Watson Glaser to shortlist the top 30% of test-takers. Your goal is to score in the top quartile of candidates.

These questions ask: what unstated assumptions underlie this argument? You're given an argument (a claim and supporting reasoning) and must identify whether certain assumptions are made. You answer: (1) Assumption made — the argument relies on this assumption to be sound, or (2) Assumption not made — the argument doesn't require this assumption.

Example argument: "We should hire more junior staff because salary costs would decrease." Potential assumption: "Salary costs are currently too high." Answer: Assumption made. The argument only makes sense if the arguer believes salary costs are too high.

Another potential assumption: "Junior staff are less experienced than senior staff." Answer: Assumption made. The argument assumes junior staff cost less because they're less experienced.

The skill here is identifying what must be true for the argument to work. Some assumptions are explicit, others implicit. For example, "We should require job candidates to pass a test" assumes (implicitly) that passing the test is a good indicator of job performance. If the test didn't predict performance, the recommendation wouldn't make sense.

Strategy: for each potential assumption, ask yourself: "Would the argument still make sense if this assumption were false?" If yes, it's not an assumption. If no, it's an assumption the argument requires.

Interpretation questions present data (charts, statistics, written information) and ask what it shows. "What does the data illustrate?" You answer based on what the data actually demonstrates, not what it might suggest. These test careful reading and avoiding over-interpretation.

Example: "Sales increased from £100m to £150m. Conclusion: The company is performing well." The data shows sales increased, not whether performance is "good" (good relative to what? Relative to targets? Relative to competitors?). The interpretation must be grounded in what the data literally shows.

Argument evaluation questions ask whether an argument is strong (logically sound, based on relevant evidence) or weak (logical fallacy, irrelevant evidence, overgeneralisation). Example argument: "We should ban university entrance tests because research shows that essays are better predictors of success." Answer: This argument might be weak. It assumes essays are the only alternative and doesn't address whether tests have other benefits. It also doesn't prove essays are significantly better.

Strategy for interpretation: stick to what the data shows. Avoid assumptions. For argument evaluation: identify logical fallacies, check whether evidence is relevant, note if the conclusion is too broad for the evidence provided.