V0 · How the Verbal Section Works

Fourteen questions. Thirty minutes. Three question types, mixed together in no fixed order.

That is the whole specification, and two things follow from it immediately.

You have a little over two minutes per question. Nobody spends the average — some questions take forty seconds and some take four minutes — but the total is fixed and it is not generous. A candidate who reads carefully and thinks thoroughly about every question will not finish, and unanswered questions score zero.

The types are mixed. You do not get all the Sentence Correction together. You will move from a grammar question to a passage to a logic question and back, and each demands a different mode of reading. Part of what this course builds is the ability to switch quickly and without a run-up.

The section is delivered in two modules, and how you perform in the first determines the difficulty of the second. Within a module you can move freely — skip a question, come back, change an answer — and once you submit a module you cannot return to it. The strategic consequences of that are the subject of the Exam Strategy chapter, and if you read only one other chapter before starting here, read that one.

Here is one. Read it before reading anything below it.

The stimulus is the paragraph. It is short — three sentences here, rarely more than six. It contains an argument: some facts, and a conclusion drawn from them.

In this one, the facts are that the lighting changed and assaults fell. The conclusion is the last sentence: the lighting caused the fall.

The question stem is the line beginning "Which of the following…". It tells you what to do with the argument. This one says undermines, so you are looking for the fact that makes the conclusion less believable.

There are eight varieties of stem, and they are the subject of chapter V2. For now the point is that the stem changes everything, and reading it before the stimulus is worth two seconds.

The five options are all plausible-sounding statements about the situation. Exactly one does the job the stem asked for. The other four are wrong for reasons that are often subtle — true but irrelevant, relevant but pointing the wrong way, or addressing a claim the argument did not make.

Notice what you are not asked. You are not asked whether LED lighting is a good idea, whether the authority is competent, or whether assaults actually fell. Everything in the stimulus is given as true, and your opinion about the subject matter is irrelevant.

What you are asked is whether this reasoning supports this conclusion. And it has a gap: two things happened in sequence, and the argument says one caused the other. It could be true. It could also be that something else caused the fall, and the lighting contributed nothing.

Option B is the answer, and here is why it is decisive. If assaults fell by the same proportion where the lighting did not change, then whatever caused the decline was not the lighting — it happened equally where the supposed cause was absent.

Option E is the second-best answer and worth understanding, because it does weaken the argument: extra police patrols are an alternative explanation. But B is stronger. An alternative cause says something else might explain this; a control says the effect happened without the cause at all.

A, C and D are all things you might reasonably think about if you worked for the city. None of them touches whether the lighting reduced assaults.

That is the shape of every Critical Reasoning question: find the gap between what you were told and what you are asked to believe, then do to it whatever the stem says.

A passage, followed by three or four questions about it. Passages run 200 to 350 words and are drawn from academic writing — history, social science, natural science. You will not need any prior knowledge of the subject.

Here is a short one, with a single question.

The passage is dense and it is not on your side. Read that second paragraph again and notice what it does not do: it does not say "in contrast", it does not summarise, it does not tell you which reading the author prefers. Business writing signposts constantly. This does not.

The questions come in six varieties, and they divide into two groups that matter more than the six names. Three of them — main idea, function, tone — are about the passage as a whole and can be answered from a mental map, without looking back. Three — detail, inference, application — are about specific content and should never be answered without looking back.

This question is a main-idea question. The answer is C.

Why the others fail is the useful part. A takes one of the two readings and offers it as the passage’s own view — but the passage endorses neither. B is true of archaeology in general and would fit a thousand other passages; if an option would work as the main idea of a passage you have not read, it is too broad. D describes a detail from the second paragraph. E states something the second reading implies, not what the passage is doing.

Three of the four wrong answers are true statements. That is normal here, and it is the single most important thing to understand about Reading Comprehension: the wrong answers are rarely false. They are about something slightly different from what was asked.

Not memory. The passage stays on the screen; you can look at it as often as you like.

What is tested is whether you can see the passage’s structure — what it is arguing, what each paragraph is doing, where the author stands — from one careful reading. That is a different skill from the extractive reading you do professionally, where you open a document knowing what you want and take only that. Here you do not know what you will be asked, so you cannot read selectively.

A sentence with part of it underlined, and five versions of the underlined part. Option A always repeats the original exactly, so choosing A means no change needed.

Option A is the original. This matters more than it sounds. It means one of your five options is always "leave it alone", and roughly one item in eight is correct as written. Candidates who assume something must be wrong invent errors that are not there.

The error is usually outside the underline. Here the underlined portion is just the verb phrase, and there is nothing wrong with "are being catalogued by two graduate students" as a phrase. The problem is its relationship to a word fourteen words earlier.

The options tell you what is being tested. Look at how they differ: are versus is versus have. That split announces the item is about subject–verb agreement, and you can settle that question and eliminate three options before thinking about anything else. This is the most efficient move in the format and most candidates never make it — they read all five options in full instead.

The answer is C. Strip the sentence to its skeleton by crossing out every prepositional phrase and every clause set off by commas:

What remains is The collection … is being catalogued. The subject is collection, singular. Manuscripts and volumes are both plural, both sit between the subject and the verb, and both pull your ear toward are.

Not your writing. You write well; that is not what is at issue.

What is tested is a specific list of errors that competent English speakers make and do not hear. Your ear will pass over a subject–verb mismatch across fourteen words of intervening clause because you know what the sentence means. The exam is built precisely on that gap.

The good news is that the list is short. Almost every item turns on one of eight recurring axes, and chapter V5 is a catalogue of them.

They look like three different subjects — logic, reading, grammar. Underneath they measure something closer together than that.

Critical Reasoning measures whether you can tell the difference between what follows and what merely sounds reasonable. In a meeting, "sales rose after we changed the packaging, so the packaging worked" passes without comment. Here it is the entire question.

Reading Comprehension measures whether you can extract structure from prose that does not offer it. Not speed, and not retention — the text is in front of you. What is tested is whether you can say what the author is doing and how committed they are to it.

Sentence Correction measures whether you can identify a specific set of structural errors under time pressure. It is the most mechanical of the three and the most learnable, which makes it the highest return per hour of study.

The common thread: all three punish reading that is fast, confident and approximate — the reading that has served you well for fifteen years. Each rewards a slower, more literal reading of what is actually on the page, applied for a very short time.

That is why this section is harder for experienced professionals than the raw material suggests. The material is not difficult. The habit is.

One more thing before the techniques, because it accounts for a lot of early frustration.

The prose is deliberately unhelpful. Compare two ways of writing the same thing:

Same content. The first tells you where the turns are, what matters, and what to conclude. The second makes you do all three.

This is not the exam being unfair. Academic prose genuinely reads like the second version, and the ability to work through it is part of what the exam measures. But it means that a candidate who reads quickly and well in a business context can feel unexpectedly slow here, and conclude that something is wrong with them. Nothing is. It is a different register and it takes practice.

Numbers are written into sentences rather than set out. No bullets, no bold on the figures, no tables unless the format calls for one. You decide what is relevant.

The sentences are long. Not to obscure the meaning, but because the ideas have qualifications attached, and the qualifications are frequently where the answer is.

Fourteen questions in thirty minutes. The averages by type, from experience rather than from arithmetic:

Notice the shape. A Reading Comprehension set is a large fixed cost followed by cheap questions, so never abandon a passage after reading it — you have paid the expensive part. Sentence Correction is where you buy back time, because a recognised item takes under a minute.

Do not compute your pace during the exam. Pick two checkpoints and know where you should be: roughly five questions done at ten minutes, ten at twenty. Two numbers, no arithmetic.

Everything else about time — the ninety-second decision to abandon, guessing with a method, what to do in the last five minutes — is in the Exam Strategy chapter, and it applies to all three sections.

The Verbal course is six chapters and they are meant to be read in order.

V1 · Anatomy of an Argument. The single skill underneath all Critical Reasoning: find the conclusion, find the premises, name the gap between them. Everything in V2 is an application of this, so V2 will not work without it.

V2 · The Question Types. All eight Critical Reasoning stems, and why each is the same question about the gap asked a different way.

V3 · How to Read. Reading Comprehension before the questions: reading for structure, building a four-line map, catching the turn that has no signal word.

V4 · The Question Types. The six Reading Comprehension question types, and which of them you answer from the map versus from the passage.

V5 · Sentence Correction. The eight recurring axes, from subject–verb agreement to idiom.

Each chapter has practice attached — drills that isolate one skill at a time, items that combine them, and a diagnosis table that reads your errors and tells you what to repeat. The practice belongs to the chapter and is linked from the section that teaches it. Do the practice before moving on. A chapter you have read and not practised is a chapter you will have to read again.

Three things worth carrying into V1.

Everything in a stimulus is true. You will spend the whole section not questioning the facts you are given. That is the opposite of your professional instinct, and it takes deliberate effort at first.

The wrong answers are usually true statements. They are wrong because they answer a slightly different question. When two options both look right, the question is almost never which is more true — it is which one answers what was asked.

One in eight Sentence Correction items needs no change. If you find yourself changing every sentence, that is a habit costing you points, and it is the cheapest thing on this list to fix.