How a critical thinking test works, and what your score actually says
Most reasoning tests ask whether you can find a pattern. A critical thinking test asks something harder and far less comfortable: whether you can tell the difference between what a piece of writing has proved and what it has merely made you feel. Those two things overlap much less often than people assume. A paragraph can be true, well written, and entirely persuasive while still failing to support the conclusion sitting at the end of it, and the whole point of this format is to catch the moment you stop noticing.
The twenty items on this page are grouped into the five sections that have defined the format since the middle of the last century: recognising assumptions, inference, deduction, interpretation and the evaluation of arguments. They were all written for this site. Nothing here is lifted from a commercial appraisal, and the scenarios are ordinary ones — a bus fleet, a bakery, a hospital ward, a cycling club — precisely so that no specialist knowledge helps you and no specialist knowledge holds you back.
The five sections and what each one traps
Recognising assumptions gives you a short argument and asks what it is quietly taking for granted. The skill being measured is necessity: an assumption is something the argument needs in order to work, not merely something that would strengthen it if true. Candidates lose points here by choosing the option that would help the argument most, rather than the one without which the argument collapses.
Inference presents a passage of fact and asks which conclusion the facts support. The trap is the conclusion that is probably true in real life but is not established by the text. If a passage says early opening days always sold out, it tells you nothing whatever about the days the shop opened late — a point that seems obvious written down and disappears the moment it is embedded in a plausible story.
Deduction is the strictest section: you accept the premises as given, however odd, and decide what follows with certainty. This is where contrapositives and modus tollens live, and where the classic errors are affirming the consequent and denying the antecedent. Nothing about the plausibility of the premises matters. If the premises say every member of the choir reads music, and you are told only that Petra plays piano, then the honest answer is that the premises settle nothing about Petra.
Interpretation works on numbers and asks which reading follows beyond reasonable doubt. The traps are almost always the same three: proportions read as totals, an average read as a claim about every individual, and a change over time read as a cause. A rise from 41 per cent to 48 per cent means more adults read a book; it does not mean the average reader read more books, and it does not mean most adults read.
Evaluating arguments asks you to weigh reasons rather than derive them. A strong argument bears directly on the question and rests on something the situation actually establishes. A weak one is off-topic, appeals to what other people do, restates the conclusion, or trades on an emotional reaction. Popularity, tradition and personal preference are the three that catch most people, because in ordinary conversation they are perfectly acceptable reasons.
Reading your score
Your raw total out of twenty is mapped onto the familiar hundred-point scale so that it can sit beside your other results on this site. Twenty items is enough to place you confidently in a band and not enough to separate you from someone two or three points away, so read the band rather than the number.
| Score | Correct out of 20 | What it suggests |
|---|---|---|
| 130 and above | 19–20 | You reliably refuse conclusions the text has not earned |
| 120–129 | 17–18 | Strong; the misses are usually in interpretation of figures |
| 110–119 | 15–16 | Comfortably above the typical graduate applicant |
| 90–109 | 11–14 | Typical range; causal traps are catching you |
| 80–89 | 9–10 | You are answering from plausibility rather than from the passage |
| Below 80 | 8 or fewer | Worth retrying untimed before drawing any conclusion |
Practical ways to raise your score
- Read the question stem before the passage. Whether you are hunting an assumption or a valid deduction changes what you should be looking for entirely.
- For any tempting option, ask what extra fact it needs. If you have to supply that fact yourself, the option is wrong.
- Treat every word like most, some, all, only and never as load-bearing. Swapping some for most is the single most common way a correct statement is turned into a wrong option.
- When a passage reports a change and an option reports a cause, the option is nearly always the trap — unless the passage explicitly ruled the alternatives out.
- Answer everything. Nothing is deducted for a wrong answer, so an untouched item is strictly worse than a guess between two survivors.
How this differs from the other reasoning tests here
The deductive reasoning test stays inside formal logic: premises in, valid conclusion out, no judgement required. The logical reasoning test mixes puzzles with constraints to satisfy. This test is the one closest to the work people actually do with documents, because four of its five sections require you to decide how far evidence stretches rather than to compute an answer. If you score well here but poorly on the matrix-based tests, that is a genuinely informative gap: verbal-analytic reasoning and fluid pattern reasoning are related but far from identical.
What the test cannot tell you
A twenty-minute screen exercise measures critical thinking under artificial conditions: short passages, clean options, one defensible answer, no stake in the outcome. Real critical thinking happens where the evidence is incomplete, the sources have motives, and you would rather the conclusion went one way than the other. Nobody has built a good test for that, and a high score here is best understood as evidence that you have the tools, not proof that you use them when the answer matters to you personally.
Frequently asked questions
What does a critical thinking test measure?
It measures how carefully you separate what a passage actually establishes from what merely sounds reasonable. The five sections cover unstated assumptions, inference from evidence, formal deduction, interpretation of data, and the strength of an argument.
How long is this critical thinking test?
Twenty questions with a twenty minute limit, which works out at one minute per item. Most people finish with a few minutes to spare, and the test scores itself automatically if the clock runs out first.
Is this a published commercial appraisal?
No. It is an independent practice test written from scratch in the same five-section style. Published commercial appraisals have their own norms and licensing, and no item here is taken from it.
Why are the correct answers often the cautious ones?
Because critical thinking is mostly the discipline of not going beyond the evidence. Options that add a cause, a motive or a prediction the passage never supplied are the classic traps, and on this test they are always wrong.
Do employers use tests like this?
Yes. Law firms, consultancies, banks and graduate schemes commonly use a critical reasoning stage because the skill predicts performance in work built on documents, evidence and argument. Practising the format removes most of the surprise from a real sitting.
Can I improve my critical thinking score?
The underlying skill improves slowly, but test performance improves quickly once you learn the recurring traps: reversed causation, sufficient conditions treated as necessary, proportions confused with totals, and conclusions that are true in the world but not supported by the passage.