- GMAT critical reasoning arguments reuse a small set of structures, and naming the structure before reading the choices is the whole method.
- Every structure has a predictable gap between evidence and conclusion, and the right answer to strengthen, weaken, assumption and evaluate questions lives in that gap.
- A prephrase is a one-line statement of the gap, written before the choices; the choices then get matched against it instead of read for appeal.
- Verbal is 23 questions in 45 minutes, 1:57 each, and a structure-first read brings most critical-reasoning questions in under 1:30.
- Log the structure of every critical-reasoning miss; verbal-weak candidates usually miss one or two structures repeatedly, not all nine.
Your Verbal score is the section holding your total down, and most of the damage is in critical reasoning, where you read all five answer choices, find two that sound plausible, and pick one on feel. The fix is to stop reading the choices first. GMAT critical reasoning arguments are built from a small set of structures, nine by our count, and once you can name the structure in the stimulus, you can state what the right answer has to do before you see a single choice.
That statement is the prephrase. This guide gives you the nine structures, an original example of each, and the prephrase for the most common question stems attached to it.
Why the structure comes before the choices
Verbal Reasoning on the Focus Edition is 23 questions in 45 minutes, which GMAC (mba.com) sets at about 1 minute 57 seconds per question across reading comprehension and critical reasoning (GMAC exam structure). A critical-reasoning question read choices-first takes most candidates 2:15 to 2:45, because every choice has to be evaluated on its own merits against a stimulus you are still holding loosely in your head. The same question read structure-first takes 1:15 to 1:30, because you are matching five choices against one sentence you already wrote.
The time matters, but the accuracy matters more. Wrong answers on this test are written to sound reasonable to someone who has not committed to what they are looking for. Four of the five choices are usually relevant to the topic, use the stimulus’s vocabulary, and do something that feels like strengthening or weakening. Only one does the specific thing the gap requires. If you arrive at the choices without a prephrase, the test’s writers have designed the wrong answers for you.
Name the structure, state the gap, and only then read the choices. The order matters.
The nine structures
Each entry below gives the shape, an original example, and the prephrase for the question types that most often attach to it. The question types are strengthen, weaken, assumption (what the argument requires), evaluate (what would help decide), and sometimes resolve the paradox or boldface.
One: correlation to cause
The evidence is that two things occur together, or one follows the other; the conclusion is that one causes the other. The gap is everything else that could explain the pairing.
Example: In the twelve months after a city added protected bike lanes to its downtown core, retail sales in the core rose 9%. The city council concluded that the bike lanes caused the increase.
Prephrase for weaken: something else happened in that year that would raise sales (a new office tower opened, a competing district closed), or the direction is reversed (sales were rising and attracted the investment). Prephrase for strengthen: rule out the alternatives, or show the same pattern elsewhere with nothing else changing. Prephrase for assumption: no other factor accounts for the rise.
Two: sample to population
The evidence describes a group that was studied; the conclusion applies to a larger or different group. The gap is whether the sample represents the population.
Example: A survey of customers who called a software company’s support line found that 70% were dissatisfied with the product’s search feature. The company concluded that most of its customers dislike the search feature.
Prephrase for weaken: the sample is skewed (people who call support are the people with problems). Prephrase for strengthen: the callers resemble the customer base, or a random sample agrees. Prephrase for evaluate: how were the respondents selected, and how do they compare to customers who did not call.
Three: plan to goal
Someone proposes an action to achieve a stated goal. The gap is whether the action produces the goal, and whether it produces side effects that defeat it.
Example: To reduce employee turnover, a regional bank plans to raise starting salaries for tellers by 8%. The bank expects turnover among tellers to fall substantially within a year.
Prephrase for weaken: turnover is driven by something the plan does not touch (schedules, management, commute), or the raise triggers a cost that undoes it. Prephrase for strengthen: exit interviews cite pay, or a comparable bank that raised pay saw turnover fall. Prephrase for assumption: pay is a significant reason tellers leave.
Four: analogy
Something true of one case is concluded to be true of another case because the two are similar. The gap is the relevant difference.
Example: A chain of coffee shops succeeded by opening locations inside grocery stores in suburban Dallas. The chain’s leadership concluded that the same approach will succeed in downtown Chicago.
Prephrase for weaken: a difference between the two settings that bears on the outcome (downtown shoppers do not drive to grocery stores, or downtown already has a coffee shop on every corner). Prephrase for strengthen: the two settings match on the factor that drove the first success. Prephrase for evaluate: whether the factors behind the Dallas result exist in Chicago.
Five: trend to prediction
Evidence of a past pattern supports a conclusion about the future. The gap is whether the conditions that produced the pattern will continue.
Example: Enrollment in a university’s evening MBA program has grown about 6% a year for five years. The dean projects that enrollment will continue to grow at that rate for the next five.
Prephrase for weaken: a condition that drove the growth is ending (a large local employer that paid tuition is cutting the benefit), or a constraint is being hit (the program is near capacity). Prephrase for strengthen: the drivers of growth persist. Prephrase for assumption: no change in the relevant conditions is expected.
Six: percentages and absolute numbers
Evidence is given in one form (a percentage, a rate, a share) and the conclusion is drawn in the other (a count, a total), or the reverse. The gap is the missing base.
Example: The share of a hospital’s emergency-room visits that result in admission fell from 22% to 17% over three years. An administrator concluded that fewer patients are being admitted from the emergency room than three years ago.
Prephrase for weaken: the base changed (total emergency-room visits rose enough that 17% of the new total exceeds 22% of the old). Prephrase for strengthen: total visits held steady or fell. Prephrase for assumption: the number of visits did not rise substantially.
This structure is the one that quant-strong candidates miss least and verbal-weak candidates miss most, and the fix is the same two words every time: what base. Ask it every time.
Seven: cost and revenue
The argument concludes something about profit, loss or financial benefit from evidence about only one side of the ledger. The gap is the other side.
Example: A publisher cut the price of its e-books by 30% and saw unit sales rise 40% in the following quarter. The publisher concluded that the price cut increased its e-book profits.
Prephrase for weaken: revenue per unit fell more than volume rose (30% off and 40% more units is actually a revenue decline of 2%: 0.7 times 1.4 is 0.98), or costs rose with volume. Prephrase for strengthen: margins per unit were high enough that the volume gain more than covered the price cut. Prephrase for assumption: the increase in volume outweighed the reduction in price per unit.
That example is worth working with a pencil once. Many candidates see 40% and 30% and assume the larger number wins.
Eight: explaining a result
The argument observes a fact or a discrepancy and offers one explanation for it. The gap is every other explanation. Resolve-the-paradox questions are the mirror image: they give the discrepancy and ask you to supply the explanation.
Example: A regional airline’s on-time arrival rate improved from 78% to 86% in the year after it replaced its scheduling software. The airline attributed the improvement to the new software.
Prephrase for weaken: another explanation (the airline also dropped its most delay-prone route, or weather that year was unusually mild). Prephrase for strengthen: rule out the alternatives (routes and weather were comparable). Prephrase for a paradox version (say, the software was installed but delays rose): a factor that reconciles the two facts, such as a surge in flights that the schedule did not anticipate.
Nine: necessary and sufficient conditions
The evidence establishes that one thing is required for, or guarantees, another; the conclusion treats the relationship as going the other way. The gap is the direction of the conditional.
Example: Every candidate the firm has promoted to partner in the past decade had completed a rotation in the firm’s London office. Dana has just completed a London rotation, so Dana will be promoted to partner.
Prephrase for weaken: the London rotation is necessary but not sufficient (many who completed it were not promoted). Prephrase for assumption: completing the rotation is enough on its own, or Dana meets whatever else is required. Prephrase for a flawed-reasoning stem: the argument confuses a requirement with a guarantee.
This structure also shows up inside the others. A plan argument that says “we will reach the goal because we have done the necessary thing” is structure nine wearing structure three’s clothes.
How the prephrase changes the way you read choices
Once the prephrase is written, the five choices are read for one property: does this choice do the thing the prephrase says, or not. That turns the standard wrong-answer types into fast eliminations.
Choices that are about the topic but not about the gap go first. In the bike-lane example, a choice about how many cyclists used the lanes is on topic and irrelevant to whether sales rose for another reason. Choices that do the opposite of what the stem asks (strengthen on a weaken question) go next, and they are common because a candidate without a prephrase cannot tell direction apart under time. Choices that restate the evidence or the conclusion in different words do nothing to the gap and go third. What remains is usually one choice and sometimes two, and the two differ on whether they reach the gap directly or only sort of.
The “sort of” choice is the test’s favorite wrong answer. It addresses the gap in a weaker or more conditional way than the right answer, and it tends to sound more cautious and therefore more reasonable. Pick the one that does the job, not the one that sounds measured.
Match, don’t evaluate. Evaluating is what the wrong answers want.
Learning the nine in four weeks
Week one is recognition without the clock. Take 15 to 20 critical-reasoning questions a night, cover the choices, and write the structure number and the gap before uncovering. Do not answer yet. Check the gap against the correct answer’s explanation and note where your gap was too vague (“the sample might be biased” is vague; “support callers have problems” is a gap). Most candidates can name the structure on sight by the end of the week.
Week two adds the answer and the prephrase for the specific stem. Same 15 to 20 a night, still untimed, now writing the prephrase, then answering, then logging by structure whether you got it. By the end of the week your log shows which structures you miss. Verbal-weak candidates almost always have two, and six (what base) and nine (direction of the conditional) are the most common.
Week three is timed, 1:45 a question, the same drill with the noteboard shorthand (“six, what base”) replacing full sentences. Week four mixes critical reasoning into full 23-question Verbal sections at 1:57 average and checks whether the recognition holds when a reading-comprehension passage has just consumed six minutes.
Log the structure of every miss for the full four weeks. The log, not the total, is what tells you if the method took.
Rick’s note: the candidates who struggle with this are not the ones who can’t reason. They are the ones who have been told for years that reading carefully is the skill, so they read the five choices carefully and get outreasoned by a test writer who knew exactly what careful reading would find appealing. The structure is how you stop playing the writer’s game.
What to do this week
- Copy the nine structures with their one-line gaps onto a single index card and keep it beside you for every Verbal set.
- Tonight, do 15 critical-reasoning questions with the choices covered, writing the structure number and the gap before you look. Do not answer yet.
- Tomorrow, repeat with the prephrase for the stem written before uncovering, then answer and log each miss by structure.
- By Saturday, identify your two most-missed structures and do 20 questions of those alone.
- Next week, time the same drill at 1:45 a question and check whether the structure still appears on the noteboard before the choices.
Questions people ask
How many critical reasoning questions are on the GMAT Focus Verbal section?
GMAC does not publish a fixed split, but the 23-question Verbal section is divided between reading comprehension and critical reasoning, and in the forms we see critical reasoning is roughly half. That makes it around 10 to 12 questions, each of which is one of the nine structures below.
What is a prephrase?
A one-line statement, written or said before you look at the answer choices, of what the right answer must do. For a weaken question on a causal argument, the prephrase is ‘something else caused the effect, or the cause and effect are reversed.’ Then you read the choices looking for that, and the four wrong choices become easy to discard because they do not do it.
What if the argument does not fit any of the nine?
It almost always fits one, sometimes two at once (a plan argument that rests on a causal claim is common). If you genuinely cannot name it, fall back on the universal move: find the conclusion, find the evidence, and state the gap between them in your own words. That is what every structure’s prephrase is doing anyway.
Do boldface and inference questions use the structures too?
Boldface questions ask about the role of a sentence, so recognizing the structure tells you immediately whether a bold portion is evidence, conclusion, counter-position or concession. Inference questions are different: they ask what must be true from the statements and do not have a gap to fill, so the structures matter less there. Treat inference as its own task.
How long should this take to learn?
Two weeks to learn the nine with their prephrases and about two more to make the recognition automatic under the clock. Verbal-weak candidates we work with usually see their critical-reasoning accuracy move first and their reading-comprehension timing second, because the time saved on critical reasoning is what buys the passages room.