Nova: Picture this. You have spent three months crafting the perfect training contract application. Every word polished. Every experience framed beautifully. You hit submit. And then, before a single human being ever reads it, a 30-minute online test decides whether your application goes straight into the bin. That is the Watson Glaser Test. And for thousands of aspiring solicitors every year, it is the gatekeeper standing between them and a career at firms like Clifford Chance, Linklaters, and Hogan Lovells.

Nova: : Three months of work, and it comes down to 40 multiple choice questions in half an hour? That sounds brutal, Nova. But also, honestly, it sounds like something you can prepare for if you know what you are doing.

Nova: Exactly. And that is what we are talking about today. There is a course called Beating the Watson Glaser Test, created by a man named Joe Mallet. Joe secured five training contract offers and scored in the 99th percentile on this test. He now teaches at The Corporate Law Academy, and his course has become one of the go-to resources for anyone serious about passing. Today we are unpacking what makes his approach different, what the test actually demands, and how you can learn to think like the test, not like yourself.

Nova: : I love that phrase. Think like the test, not like yourself. That already sounds counterintuitive and slightly uncomfortable, which means we are in for a good conversation. Let us get into it.

Nova: Let us start with the basics. The Watson Glaser Critical Thinking Appraisal is a pre-employment psychometric test used primarily in the legal sector. Forty questions, five sections, thirty minutes. It does not test legal knowledge. It does not test your grades. It tests one thing and one thing only: your ability to think critically.

Nova: : Okay, but critical thinking sounds like one of those buzzwords that everyone claims to have. What does the test actually measure?

Nova: It measures five distinct skills: your ability to draw inferences from information, to recognise hidden assumptions, to make logical deductions, to interpret evidence, and to evaluate the strength of arguments. Each of those gets its own section. And here is the thing that makes it so high-stakes: for many top law firms, the Watson Glaser is used as a filter. If you do not hit their percentile cutoff, your entire application, however brilliant, never gets read by a human.

Nova: : So it is literally a gatekeeper. But what kind of score do you need? Is there a universal pass mark?

Nova: There is not, and that is part of what makes it so tricky. Each firm has its own benchmark, and they rank you against a norm group. One candidate on the TCLA forum reported averaging between eighty and ninety percent on practice tests but only landing in the 65th percentile for Linklaters while hitting the 88th percentile for Hogan Lovells. The cutoff can be anywhere from the 20th to the 75th percentile depending on the firm and the applicant pool. Some firms reportedly use it as a standalone filter, others weigh it alongside your application. The uncertainty itself is stressful.

Nova: : And this is where Joe Mallet enters the picture. A guy who got five training contract offers and scored in the 99th percentile. That is not luck, that is a system.

Nova: Absolutely. Joe is now an associate at Pillsbury Winthrop Shaw Pittman, an international law firm, having trained at Vinson and Elkins. But his course, Beating the Watson Glaser Test, was born from his own experience navigating this exact gauntlet. He realised that most people fail not because they lack intelligence, but because they approach the test with the wrong mindset entirely.

Nova: Let us walk through the five sections, because each one has its own logic, and confusing them is one of the biggest reasons people lose marks. First up: Inferences. You get a short passage, then a series of statements. For each statement you have to decide: is it True, Probably True, Insufficient Data, Probably False, or False?

Nova: : Five options. That already feels more nuanced than a simple yes no. What is the trickiest part?

Nova: According to Joe Mallet and many others who have mastered the test, the hardest distinction is between Insufficient Data and Probably False. Candidates often let their own general knowledge seep in and push them toward Probably False, when the passage itself simply does not contain enough information to judge. The rule is: you may use commonly accepted knowledge, but you cannot bring in specific outside facts. If the passage does not address it, you cannot conclude it.

Nova: : So you have to become almost unnaturally disciplined about only using what is on the page.

Nova: Exactly. Section two: Recognition of Assumptions. Here you are given a statement and a proposed assumption. You decide whether that assumption is actually being made. This is a binary test, made or not made. Joe emphasizes that an assumption is something the statement presupposes or takes for granted. It must be necessary for the statement to hold. Something that might be true but is not essential to the argument? That is not an assumption being made.

Nova: : So it is about finding the hidden scaffolding holding up the argument.

Nova: Beautiful way to put it. Section three: Deduction. You get a set of facts and a proposed conclusion. The question: does this conclusion follow necessarily from the facts? Not probably. Not maybe. Necessarily. This is the highest bar in the entire test.

Nova: : Necessarily, as in, it would be logically impossible for the facts to be true and the conclusion false?

Nova: Precisely. Now contrast that with section four: Interpretation. Here you read a passage and decide whether a conclusion follows beyond reasonable doubt. That is a lower bar than necessarily. The distinction between deduction and interpretation trips up countless candidates. The Corporate Law Academy forum is full of threads where people discover this nuance for the first time and realise they have been approaching both sections identically.

Nova: : And finally, section five?

Nova: Evaluation of Arguments. You are given a statement and a proposed argument about it. You judge whether that argument is strong or weak. A strong argument is both relevant and directly addresses the statement. A weak argument might be irrelevant, rely on assumptions, or commit logical fallacies. Joe teaches students to watch out for straw man arguments, red herrings, and the classic correlation-equals-causation trap.

Nova: So what does Joe Mallet actually teach in his course that makes the difference? The course is structured with dedicated modules for each question type, where Joe works through practice questions and explains not just why the right answer is right, but why the wrong answers are wrong. That second part is critical.

Nova: : Because understanding the anatomy of a wrong answer helps you avoid similar traps in the real test.

Nova: Exactly. But the deeper insight running through the whole course is something we touched on earlier: you must learn to think like the test, not like yourself. Your intuition, your personal knowledge, your opinions about what seems reasonable in the real world, these will all work against you in the Watson Glaser.

Nova: : That sounds almost philosophical. Can you give me a concrete example?

Nova: Sure. Imagine a passage says: "A study found that employees who work from home report higher job satisfaction than office-based workers." A proposed inference might be: "Working from home causes higher job satisfaction." Your common sense might say, yeah, that tracks, probably true. But the passage only reports a correlation. It says nothing about causation. The correct answer could be Insufficient Data because the passage does not establish causation. Joe teaches you to catch these distinctions automatically.

Nova: : So it is about developing a kind of mental checklist you run through for each question type.

Nova: That is exactly the approach. The Oxford University Careers Service article on the Watson Glaser calls these thinking algorithms. For the Evaluation of Arguments section, for example, you might use an ITDN table, checking whether the argument is Important, True, Directly relevant, and Not assuming. For Recognition of Assumptions, you apply the negative test: if the assumption were false, would the statement collapse? If yes, it is an assumption being made. If the statement could still stand, it is not.

Nova: : And Joe Mallet walks students through these algorithms section by section?

Nova: He does, and he goes further. His course includes a workbook with additional practice questions so you can apply the strategies yourself. He also addresses the meta-skill of learning how to learn for this test. The course opens with modules on your overall approach, how to structure your preparation, and the right mindset. This is not just about cramming practice questions. It is about rewiring how you process arguments.

Nova: : I can see why someone with five training contract offers and a 99th percentile score would have credibility here. But is there evidence it works for other people?

Nova: Yes. On TCLA forums, users who have taken the course report significant score improvements. One user noted having completed seventy-eight percent of the course and described their subsequent success as attributable to it. The course is consistently listed among top resources by applicants who successfully secured training contracts. It is not magic, it is systematic preparation.

Nova: Let us talk about the most common ways people sabotage themselves on this test. The number one trap, and I cannot emphasize this enough, is bringing in outside knowledge.

Nova: : Wait, but I thought the inference section specifically says you can use commonly accepted knowledge. That seems contradictory.

Nova: It is the number one source of confusion on TCLA forums. Here is the distinction. Commonly accepted knowledge means things like knowing that water is wet, or that most parents care about their children. It does not mean bringing in specific facts you happen to know about the topic of the passage. If the passage is about trends in American remote work, you cannot use your personal knowledge about remote work statistics to fill in gaps. You have to work only with what the passage gives you.

Nova: : So the boundary is between universal common sense and specific factual knowledge.

Nova: Exactly. Trap number two: confusing the standards across sections. As we discussed, deduction requires a conclusion to follow necessarily, while interpretation requires it to follow beyond reasonable doubt. Treating them the same will cost you points. Trap number three: falling for logical fallacies. The test loves to present arguments that sound persuasive but are logically flawed. The classic ones are confusing correlation with causation, straw man arguments, and red herrings.

Nova: : I have to ask. Is there any role for intuition at all, or is this purely mechanical?

Nova: That is a great question. Joe Mallet does not say to abandon intuition entirely. In fact, for the inferences section, some intuition is necessary to distinguish between Probably True and Probably False. But he wants you to develop a trained intuition, one that has internalised the rules through deliberate practice, rather than relying on your untrained gut feeling. The goal is to make the thinking algorithms so automatic that they feel like intuition.

Nova: : Practice until the systematic approach becomes second nature.

Nova: Right. And one more practical tip that comes up repeatedly: do not linger on tricky questions. The test is timed, thirty minutes for forty questions. If you get stuck on one inference question for three minutes, you are sabotaging the rest of your test. Joe advises flagging difficult questions and returning to them if time allows. Pace management is a skill in itself.

Nova: : What about the mental side? Test anxiety, impostor syndrome, that kind of thing?

Nova: The TCLA forums are full of this. Candidates talk about impostor syndrome constantly. One forum user said they were shocked every time they progressed to the next stage, completely discounting all their effort. Joe addresses this indirectly through the course structure. When you have a systematic method, when you know exactly what each section is looking for, when you have practised enough that the patterns become familiar, the anxiety diminishes. Confidence comes from competence. That is the underlying philosophy.

Nova: So let us bring this together. The Watson Glaser Test is a gatekeeper used by the UK's most prestigious law firms. It tests five dimensions of critical thinking: inference, assumptions, deduction, interpretation, and evaluation of arguments. And it is designed to make your intuition fail you.

Nova: : Which is why Joe Mallet's course, Beating the Watson Glaser Test, is built around a fundamentally different approach. Do not rely on what feels right. Learn the specific rules of each section. Develop thinking algorithms. Practise until the systematic approach becomes automatic. And crucially, learn to think like the test, not like yourself.

Nova: The three biggest takeaways I would offer any listener are these. First, understand the distinction between deduction, which requires a conclusion to follow necessarily, and interpretation, which requires it to follow beyond reasonable doubt. That nuance alone can transform your score. Second, for assumptions, apply the negative test. If the assumption were false, would the argument collapse? That is your answer. And third, treat every practice session as deliberate training, not just random quizzing. Understand why the wrong answers are wrong, not just why the right ones are right.

Nova: : And if you are serious about a career in commercial law, invest in preparation. Whether it is Joe Mallet's course through The Corporate Law Academy, free resources from Clifford Chance's website, or comprehensive practice packs from JobTestPrep, the time you spend mastering this test is tiny compared to the career it unlocks.

Nova: Beautifully said. The Watson Glaser is not testing how smart you are. It is testing whether you have learned to think in a particular disciplined way. And that, unlike raw intelligence, is something you can absolutely train. Joe Mallet proved that. So can you.

Nova: : This is Aibrary. Congratulations on your growth!