GMAT FOCUS 2026

GMAT Critical Reasoning practice: 20 questions with explanations

13 November 2026 · 20 questions, about 40 minutes · by Jaclyn Caruana, MBA

Twenty original GMAT Focus Critical Reasoning questions, ten rated Medium and ten rated Hard, across the eight question types you will meet in Verbal Reasoning. Each question shows its type, then reveals the answer, the reasoning, the wrong answer most likely to catch you and why it fails. Your score updates as you go, and a results table at the end breaks it down by question type.

The short version

Critical Reasoning is one of the two question types in the GMAT Focus Verbal Reasoning section, which GMAC describes as 23 Reading Comprehension and Critical Reasoning questions in 45 minutes. Every Critical Reasoning question gives you a short argument and asks you to do one job with it: find its assumption, strengthen or weaken it, decide what would evaluate it, draw what must follow, explain a paradox, label its parts or name its flaw. Identify that job from the question stem first, predict what the answer has to do, then read the choices. The set below trains exactly that.

Where Critical Reasoning sits on the GMAT Focus

The GMAT Focus Edition has three sections: Quantitative Reasoning, Verbal Reasoning and Data Insights. GMAC's exam overview lists the whole exam at 2 hours 15 minutes and 64 questions, with Verbal Reasoning at 45 minutes and 23 questions that assess critical reasoning and reading comprehension skills.1

GMAC's scoring page describes the Verbal Reasoning section as 23 multiple-choice Reading Comprehension and Critical Reasoning questions, and says Critical Reasoning questions measure the ability to make arguments, evaluate arguments, and formulate or evaluate a plan of action. Verbal Reasoning is scored from 60 to 90, and it counts equally with the other two sections toward the Total Score of 205 to 805.2 To see how a Verbal score combines with your other two sections, try the GMAT Focus score calculator.

The time math matters: 45 minutes for 23 questions is just under 2 minutes per question on average, and that average has to cover Reading Comprehension passages too. If you regularly run short at the end of Verbal, read why GMAT test takers run out of time and how to fix it before you drill for speed.

1. GMAC, The GMAT Exam: Exam Structure, Content and Features.

2. GMAC, GMAT Exam Scores.

The eight question types in this set

Every Critical Reasoning question starts from the same raw material, a short argument, so the question stem is what tells you the job. The table gives the stem wording used in this set, the move that solves each type, and how many questions of each type follow. The mix is weighted toward Strengthen, Weaken, Evaluate and Boldface, because those are the types where wrong choices can sit closest to the right one.

TypeTypical stemThe moveIn this set
AssumptionThe argument depends on which of the following assumptions?Name the gap between evidence and conclusion, then negate each choice. The one whose negation sinks the argument is the answer.2
StrengthenWhich of the following, if true, most strengthens the argument?Find the exact link the conclusion rests on and shore up that link, often with a comparison group.4
WeakenWhich of the following, if true, most seriously weakens the argument?Break the same link: a rival cause, groups that already differed, or a step in the plan that fails.4
EvaluateWhich of the following would it be most useful to determine in order to evaluate the argument?Pick the question whose two possible answers push the conclusion in opposite directions.3
InferenceIf the statements above are true, which of the following must also be true?Combine the facts and accept only what is forced. Reject anything that adds a cause or a quantity.2
ParadoxWhich of the following, if true, most helps to explain the facts above?Find the choice that makes both facts true at once, not just one of them.1
BoldfaceIn the argument, the two portions in boldface play which of the following roles?Find the conclusion first, decide whether the author agrees with each portion, then check both halves of a choice.3
FlawThe argument is most vulnerable to criticism on the grounds that itName the reasoning error in your own words before you read the choices.1

How to use this practice set

The first ten questions are Medium and the last ten are Hard. All scenarios, companies, towns and studies are invented for practice.

Medium: questions 1 to 10

Question 1 of 20AssumptionMedium

Velmar County plans to replace its diesel school buses with battery-electric buses. A county analyst argues that the switch will not raise the yearly cost of running the bus fleet, a cost made up of fuel, maintenance and insurance. Her reasoning: the routes and the miles driven each year will stay the same, and per mile, the electricity the new buses use will cost the county less than the diesel the current buses burn.

The analyst's argument depends on which of the following assumptions?

Answer: (C)

Why (C) is right. The conclusion is about a total: fuel plus maintenance plus insurance. The evidence covers only fuel. For the total not to rise, the other two parts cannot climb by more than the fuel savings. Negate (C): maintenance and insurance rise by more than the fuel savings, and the yearly cost goes up, so the conclusion collapses.

The most tempting wrong answer: (A). The purchase price sounds like the obvious hidden cost, but the analyst defines the yearly running cost as fuel, maintenance and insurance. A higher sticker price does not touch that total, so the argument does not need (A).

Why the other choices fail

  • (B) Comfort is not part of the cost the conclusion is about.
  • (D) The argument needs electricity to stay cheaper per mile, not to get cheaper. Negate (D) and the conclusion still stands.
  • (E) Results in other counties are neither required nor about Velmar's own costs.

Tip: When a conclusion is about a total, list every part of that total and check which parts the evidence never mentions.

Question 2 of 20StrengthenMedium

Last year the town of Orrin Falls repainted the crosswalks at twelve intersections near its schools with wide, high-visibility stripes. In the twelve months after the repainting, collisions involving pedestrians at those twelve intersections fell by 30 percent compared with the previous twelve months. The town's traffic office concludes that the new stripes caused the decline.

Which of the following, if true, most strengthens the traffic office's conclusion?

Answer: (A)

Why (A) is right. The office infers cause from a before-and-after drop. The danger is that collisions fell everywhere for some other reason. (A) supplies a comparison group: where nothing changed, collisions held steady, so a town-wide trend does not explain the drop at the repainted intersections.

The most tempting wrong answer: (B). A record-high year just before the change suggests the drop may be a return to normal levels, the pattern statisticians call regression toward the mean. That gives a reason other than the stripes for the decline, so (B) weakens the conclusion.

Why the other choices fail

  • (C) A lower speed limit is a rival cause of the decline, so it weakens the conclusion.
  • (D) Upkeep cost says nothing about whether the stripes reduced collisions.
  • (E) Fewer pedestrians is another rival explanation for fewer pedestrian collisions, so it weakens.

Tip: The strongest support for a causal claim is a comparison group that did not get the change and did not improve.

Question 3 of 20WeakenMedium

Last year, employees at Dunmore Insurance who enrolled in the company's voluntary fitness program took 40 percent fewer sick days than employees who did not enroll. Dunmore's human resources director concludes that taking part in the fitness program reduces the number of sick days an employee takes.

Which of the following, if true, most seriously weakens the director's conclusion?

Answer: (D)

Why (D) is right. Enrollment was voluntary, so the people who joined may have differed from the start. (D) shows they did: the future enrollees already took far fewer sick days before the program existed. The gap last year is then better explained by who chose to join than by the program itself.

The most tempting wrong answer: (A). A company-wide dip affected both groups, so it cannot explain why enrolled employees took 40 percent fewer sick days than the others. The gap between the groups, which is the director's evidence, is untouched.

Why the other choices fail

  • (B) Cost is a side issue. The conclusion is about sick days, not value for money.
  • (C) Job satisfaction is a different outcome and does not explain the sick-day gap.
  • (E) Counting non-attenders as enrolled would, if anything, hide part of the program's effect, so it does not weaken the claim.

Tip: When people choose their own treatment, ask whether the groups already differed before it started.

Question 4 of 20InferenceMedium

At Harlow Tech's annual sales conference, every attendee chose exactly one of two afternoon workshops: negotiation or analytics. Every attendee from the company's European office chose negotiation. More than half of all attendees chose analytics.

If the statements above are true, which of the following must also be true?

Answer: (B)

Why (B) is right. More than half chose analytics, so fewer than half chose negotiation. Every European attendee is inside the negotiation group, so the European group can be no larger than that group. It is therefore fewer than half of all attendees.

The most tempting wrong answer: (A). (A) reverses the conditional. The statements say every European attendee chose negotiation, not that every negotiation attendee was European. Non-European attendees could also have chosen negotiation.

Why the other choices fail

  • (C) Nothing is said about the size of any other office's delegation.
  • (D) The negotiation group could be entirely European, so (D) is not forced.
  • (E) This contradicts the statement that every European attendee chose negotiation.

Tip: A group that sits entirely inside a minority must itself be a minority. Chain the sizes before you read the choices.

Question 5 of 20BoldfaceMedium

Some members of the Ardley town council propose banning gasoline-powered leaf blowers. Gasoline leaf blowers are undeniably loud, and their exhaust adds to local air pollution. But most leaf blowing in Ardley is done by small landscaping crews, and the battery blowers that crews say could replace their gasoline models cost several times as much. A ban would therefore raise what residents pay for yard care while doing little for air quality, since landscaping produces only a small share of Ardley's emissions. The council should reject the proposed ban.

In the argument, the two portions in boldface play which of the following roles?

Answer: (E)

Why (E) is right. The argument opposes the ban. The first portion admits a real point on the ban's side: the blowers are loud and pollute. The argument never disputes it; it outweighs it with cost and with the small share of emissions. The second portion is the final recommendation, so it is the main conclusion.

The most tempting wrong answer: (B). The argument never says the blowers are quiet or clean. It grants both facts and then argues they do not justify a ban. A point that is admitted and outweighed is a concession, not a claim shown to be false.

Why the other choices fail

  • (A) The first portion supports the ban, the position the argument rejects, so it is not evidence for the conclusion.
  • (C) The first portion is a premise the argument concedes, not its conclusion.
  • (D) Nothing follows from the second portion; it is the final conclusion, not an intermediate one.

Tip: Before you label a boldface portion, ask whether the author agrees with it. Admitted but outweighed means concession.

Question 6 of 20EvaluateMedium

Teller County replaced the coin-operated parking meters in its downtown with a system that lets drivers pay by phone. Downtown parking revenue in the six months after the switch was 15 percent higher than in the same six months one year earlier. A council member concludes that the pay-by-phone system is responsible for the increase.

Which of the following would it be most useful to determine in order to evaluate the council member's conclusion?

Answer: (B)

Why (B) is right. Run the swing test. If the hourly rate rose by about 15 percent, the rate increase alone explains the extra revenue and the system gets no credit. If the rate did not change, that rival explanation is gone and the conclusion is stronger. The answer to (B) moves the conclusion in either direction.

The most tempting wrong answer: (A). Seasonality is a real confound in general, but this comparison already handles it: the six months after the switch are compared with the same six months a year earlier. Both periods contain the same season, so the answer to (A) does not change the evaluation.

Why the other choices fail

  • (C) Other counties' contracts say nothing about what caused Teller County's revenue change.
  • (D) Cost recovery is about whether the system was worth it, not about what raised revenue.
  • (E) Driver age does not bear on whether the system caused the increase.

Tip: Before you chase a confound, check what the comparison already controls; then look for the one it does not.

Question 7 of 20ParadoxMedium

Between 2022 and 2025, the number of Corvan City residents holding a public library card rose by 25 percent. During that period the library system did not close any branches or shorten its opening hours. Yet the number of physical books checked out from the system's branches fell steadily over the same three years.

Which of the following, if true, most helps to explain the facts above?

Answer: (D)

Why (D) is right. The puzzle has two halves: more cardholders, yet fewer physical checkouts. (D) explains both. The e-book app requires a card, which draws new cardholders, and borrowing that moves to e-books no longer shows up as physical checkouts.

The most tempting wrong answer: (B). Computer-only cardholders explain why more cards did not add checkouts, but they do not explain why checkouts fell. With (B) alone, checkouts should have stayed flat, so half the puzzle remains.

Why the other choices fail

  • (A) Population growth explains a little of the card rise and makes the drop in checkouts more puzzling, not less.
  • (C) More visits make falling checkouts harder to explain.
  • (E) Who borrows the books says nothing about why borrowing fell.

Tip: A paradox answer must make both facts true at once. Test each choice against each half separately.

Question 8 of 20StrengthenMedium

Pellam Suites, a hotel chain, will place a card in each room asking guests who stay more than one night to hang up any towel they are willing to reuse. Pellam's managers predict that the cards will reduce the number of towels the chain launders per occupied room each night.

Which of the following, if true, most strengthens the managers' prediction?

Answer: (A)

Why (A) is right. The plan has two steps: guests hang towels, and staff leave those towels alone. If housekeepers swapped hung towels out anyway, the cards would change nothing. (A) secures that second step, so a guest's choice to reuse actually reduces the towels sent to the laundry.

The most tempting wrong answer: (C). Durable towels lower what Pellam spends replacing towels, which sounds like the same goal. The prediction, though, is about how many towels are laundered, and towel durability does not change that number.

Why the other choices fail

  • (B) If anything, a large share of one-night guests limits how many guests the cards can affect, so it does not strengthen.
  • (D) This explains why Pellam cares about laundry, not whether the cards will reduce it.
  • (E) Other chains adopting cards does not show the cards reduced their laundry.

Tip: A plan's prediction gets stronger when a choice secures the step most likely to break it.

Question 9 of 20InferenceMedium

Every product that Brantley Home launched in the past five years and that went on to exceed its first-year sales target was promoted mainly through online video. Some of the products Brantley launched in that period were promoted mainly through online video and were also sold at a discount at launch. None of the products Brantley launched at a discount in that period exceeded its first-year sales target.

If the statements above are true, which of the following must also be true of the products Brantley launched in the past five years?

Answer: (E)

Why (E) is right. Some products were both video-promoted and discounted. No discounted product exceeded its target. So those video-promoted, discounted products did not exceed their targets, which means at least one video-promoted product missed its target.

The most tempting wrong answer: (D). (D) sounds like the lesson of the data, but it is a causal claim. The statements only describe which products hit their targets. They cannot show what video promotion did or did not contribute, so (D) is not forced.

Why the other choices fail

  • (A) Contradicted: the discounted video products missed their targets.
  • (B) Products could miss their targets without a discount; nothing rules that out.
  • (C) We know only that some discounted products used video, not most.

Tip: On must-be-true items, link the 'some' statement to the 'none' statement first; that pairing usually forces the answer.

Question 10 of 20WeakenMedium

To ease rush-hour congestion on the Mill Street Bridge, the city of Kenlow will begin charging a toll to cross the bridge between 7 and 9 a.m. and between 4 and 6 p.m. on weekdays. City planners predict that the toll will substantially reduce the number of vehicles crossing the bridge during those hours.

Which of the following, if true, most seriously weakens the planners' prediction?

Answer: (C)

Why (C) is right. A toll reduces traffic only if the people deciding whether to cross feel the cost. If most rush-hour drivers are fully reimbursed, the toll costs them nothing, and they have little reason to change when or whether they cross.

The most tempting wrong answer: (B). Drivers who can easily move their trips outside the toll hours are exactly the drivers a toll would push out of those hours. (B) supports the prediction; it is a wrong-direction choice.

Why the other choices fail

  • (A) How the revenue is spent does not affect whether drivers avoid the toll.
  • (D) Light traffic at other hours does not bear on the toll-hour count.
  • (E) Opposition is not evidence that drivers will keep crossing at the same rate.

Tip: A price changes behavior only if the person making the choice actually pays it.

Hard: questions 11 to 20

Question 11 of 20AssumptionHard

At Tolland University, admitted students who visited campus before the enrollment deadline were twice as likely to enroll as admitted students who did not visit. To raise the share of admitted students who enroll, the admissions office plans to offer every admitted student a free round-trip ticket to visit campus. The office concludes that, by bringing more admitted students to campus, the plan will raise that share.

The office's conclusion depends on which of the following assumptions?

Answer: (D)

Why (D) is right. The plan works only if the ticket changes someone's behavior and the visit then changes that person's decision. Negate (D): no student who would otherwise stay home takes the trip and becomes more likely to enroll. Then the tickets go only to students who would have visited anyway, or change nothing, and the share cannot rise through the plan.

The most tempting wrong answer: (A). (A) aims at the right worry, that visitors may simply be the already keen students, but it asks too much. The plan needs the visit to matter for some students, not for the visit to be the main reason visitors enroll. Negate (A) and the visit can still tip some students, so the conclusion survives.

Why the other choices fail

  • (B) The plan needs some students held back by cost, not most of them. Negate (B) and the plan can still work.
  • (C) The conclusion is about the enrollment share, not about whether the plan pays for itself.
  • (E) Like (A), this rules out all self-selection. Visitors can be keener to begin with and the visit can still add to that, so (E) is not required.

Tip: On necessary-assumption questions, prefer the weakest statement that still bridges the gap; strong words like 'mainly' or 'no more' usually overshoot.

Question 12 of 20StrengthenHard

Since an invasive crayfish species appeared in Lake Merin a decade ago, the lake's population of a native minnow has fallen by half. Researchers studying the lake conclude that the crayfish have reduced the minnow population chiefly by eating minnow eggs, which the minnows lay in shallow gravel beds.

Which of the following, if true, most strengthens the researchers' conclusion?

Answer: (B)

Why (B) is right. The conclusion has two parts: the crayfish caused the decline, and they did it chiefly by eating eggs. (B) tests the mechanism directly. Keep crayfish away from the eggs and hatching holds steady; leave the eggs exposed and hatching falls. That isolates egg predation as the route.

The most tempting wrong answer: (A). Similar declines in other invaded lakes support the idea that the crayfish harm the minnows, but they say nothing about how. The researchers' claim is specifically about eggs, and (A) leaves that link untouched.

Why the other choices fail

  • (C) A second feeding route makes it less likely that eggs are the chief one, so this weakens the conclusion.
  • (D) Fewer minnows caught restates the decline; it adds nothing about the cause.
  • (E) More crayfish fits the general claim of harm but, like (A), does not touch the egg mechanism.

Tip: When a conclusion names a mechanism ('chiefly by X'), the strongest strengthener blocks X and shows the effect disappears.

Question 13 of 20WeakenHard

Last year Larch Bank redesigned its mobile banking app. In a survey of customers who had logged in to the app during the past month, the redesigned app received an average rating of 4.6 out of 5, higher than the old app ever received. Bank executives conclude that the redesign has not driven customers away from the bank.

Which of the following, if true, most seriously weakens the executives' conclusion?

Answer: (E)

Why (E) is right. The survey reached only people still logging in, so it cannot hear from anyone the redesign drove away. (E) supplies evidence from outside that sample: closings doubled after the redesign, and most of the departing customers dropped the app right after it changed. That points to exactly the loss the executives deny.

The most tempting wrong answer: (B). (B) starts like a sampling-bias weakener, then defuses itself: rare users have rated the app about as highly as frequent users. Their absence from the survey therefore does not hide a pool of unhappy customers.

Why the other choices fail

  • (A) Slightly different ratings among current users say nothing about customers who left.
  • (C) A lower rating than competitors does not show that anyone actually left.
  • (D) This questions the comparison with the old app, but the conclusion is about customers leaving, not about whether the app improved.

Tip: When the evidence comes only from people who stayed, look for an answer that brings in the people who left.

Question 14 of 20EvaluateHard

Last year the Ashby water utility installed smart meters in every home it serves. A meter sends a household an alert when it detects continuous flow of the kind a leak produces. In the year after installation, average water use per household fell by 8 percent compared with the year before. The utility concludes that the meters reduced water use chiefly by prompting households to find and fix leaks sooner.

Which of the following would it be most useful to determine in order to evaluate the utility's conclusion?

Answer: (A)

Why (A) is right. The claim is about the route: leak alerts. If most of the decline came from households that got alerts, the route fits. If it came mostly from households that never got an alert, something other than leak repair drove the drop, and the conclusion fails. Only (A) swings the conclusion both ways.

The most tempting wrong answer: (C). Matching declines elsewhere would suggest that smart meters reduce water use, but the utility's claim is about how they do it. Meters could cut use in other ways, for example by showing households their usage, so (C) leaves the leak mechanism undecided.

Why the other choices fail

  • (B) Satisfaction with alerts does not show that alerts produced the savings.
  • (D) Payback time concerns the investment, not the cause of the decline.
  • (E) Where leaks waste more water does not show whether fixing leaks produced this decline.

Tip: When the conclusion names a mechanism, the decisive question splits the data by whether that mechanism was even present.

Question 15 of 20BoldfaceHard

Fenwick Manufacturing must choose one of two suppliers for a component used in its main product. Supplier X charges 12 percent less per unit than Supplier Y. But X's components fail Fenwick's inspection about three times as often as Y's, and each failed component stops the assembly line until a replacement is fitted. Fenwick's records show that, at X's failure rate, the line would stand idle for several hours a month, and a single hour of idle line costs Fenwick more than the entire monthly price difference between the two suppliers. Fenwick should therefore choose Supplier Y.

In the argument, the two portions in boldface play which of the following roles?

Answer: (C)

Why (C) is right. The argument concludes that Fenwick should choose Y; that conclusion is the last sentence, which is not in boldface. The first portion, X's lower price, favors X, the option the argument rejects. The second portion is a premise for choosing Y: idle time caused by X's failures costs more than X saves.

The most tempting wrong answer: (A). The second portion is persuasive and comes near the end, so it reads like the conclusion. But it is a cost fact offered as support; the recommendation itself, choose Supplier Y, is the unbolded final sentence.

Why the other choices fail

  • (B) The argument accepts that X is cheaper; it weighs the price gap, it does not deny it.
  • (D) The first portion favors X, so it cannot be evidence for choosing Y.
  • (E) The idle-line cost counts against X, the option the first portion favors.

Tip: Find the conclusion before labeling anything. If the conclusion is not in bold, neither portion can be 'the main conclusion.'

Question 16 of 20WeakenHard

The Glassfield electric utility will give each household it serves free LED bulbs to replace its incandescent bulbs. Because an LED bulb uses at least 75 percent less electricity than an incandescent bulb that gives the same amount of light, the utility predicts that the electricity its customers use for lighting will fall by roughly 75 percent once the bulbs are installed.

Which of the following, if true, most seriously weakens the utility's prediction?

Answer: (E)

Why (E) is right. The 75 percent forecast assumes households use their lights for the same hours as before. (E) says they do not: when light gets cheaper, people use more of it, a pattern economists call the rebound effect. More hours and more fixtures eat into the savings per bulb, so lighting electricity falls by less than 75 percent.

The most tempting wrong answer: (A). (A) makes the savings matter less to a household's total bill, but the prediction is only about electricity used for lighting. A 75 percent cut in a small slice is still a 75 percent cut in that slice.

Why the other choices fail

  • (B) Extra winter heating affects heating energy, not the electricity used for lighting.
  • (C) Bulb lifespan does not change how much electricity the lights draw.
  • (D) The funding source has no bearing on electricity use.

Tip: Efficiency forecasts assume behavior stays fixed. When a resource gets cheaper to use, look for people using more of it.

Question 17 of 20StrengthenHard

Around 900 BCE, the pottery found at the inland settlement of Tel Varo changed abruptly to a style typical of a culture on the coast. Some archaeologists hold that Tel Varo's residents simply began copying coastal pottery they obtained through trade. Others argue that the change shows that people from the coast settled at Tel Varo.

Which of the following, if true, most strengthens the argument that people from the coast settled at Tel Varo?

Answer: (B)

Why (B) is right. Copying through trade explains why traded wares would change, but home-made cooking pots that were rarely traded should keep the local tradition unless the people making them changed. A switch in those everyday pots favors newcomers who brought their own habits over locals imitating imports.

The most tempting wrong answer: (A). Local clay rules out imported pots, which feels like a point for settlers. But locals copying a coastal style would also use local clay, so (A) fits both explanations equally and does not favor either.

Why the other choices fail

  • (C) Coastal wares spreading inland through trade supports the copying explanation, so it weakens.
  • (D) The stimulus already says the change was abrupt; restating it adds nothing.
  • (E) A history of borrowing styles through trade makes the copying explanation more plausible.

Tip: When two hypotheses compete, an answer helps only if it fits one and not the other. Ask what each hypothesis predicts.

Question 18 of 20BoldfaceHard

In the quarter after Corrin Outdoor began sponsoring a popular trail-running podcast, sales of its hiking boots rose 18 percent. Several Corrin executives take this rise as evidence that the sponsorship is bringing in new customers. Yet the company's sales records show that nearly all of the additional boot sales went to customers who had bought Corrin boots before. In the same month the sponsorship began, Corrin also launched a discount for returning customers. The rise in sales, then, more likely reflects that discount than the podcast.

In the argument, the two portions in boldface play which of the following roles?

Answer: (D)

Why (D) is right. The argument's conclusion is the last sentence: the rise more likely reflects the returning-customer discount. The first portion is the executives' reading, which the argument rejects. The second portion is the sales-record fact used against that reading: the extra sales came from returning customers, not new ones.

The most tempting wrong answer: (B). The second portion begins after 'Yet,' so it feels like the argument's answer. It is a finding from the sales records, though. The conclusion the argument reaches instead is the final sentence about the discount.

Why the other choices fail

  • (A) The argument rejects the executives' reading, so the first portion cannot be its conclusion.
  • (C) The argument relies on the second portion, not the first, and it never rejects the sales-record finding.
  • (E) The argument does not accept any part of the executives' reading; it attributes the rise to the discount.

Tip: 'Yet' often introduces evidence, not the conclusion. Keep reading until you find the claim everything else supports.

Question 19 of 20EvaluateHard

A study of startup companies in the city of Varden found that 70 percent of the founders whose companies reached $10 million in annual revenue within five years had previously worked at a large corporation. An investor concludes that experience at a large corporation makes a founder more likely to build a company that reaches that level of revenue.

Which of the following would it be most useful to determine in order to evaluate the investor's conclusion?

Answer: (A)

Why (A) is right. 70 percent means nothing without a baseline. If about 70 percent of the founders who fell short also had corporate experience, that experience is just common among founders and predicts nothing. If the figure among them is far lower, corporate experience does go with success. The answer to (A) swings the conclusion.

The most tempting wrong answer: (D). A sector split sounds like a hidden confound, but on its own it decides nothing. Without knowing whether corporate veterans cluster in technology, the answer to (D) neither supports nor undermines the investor.

Why the other choices fail

  • (B) Time spent at the corporation is a detail about the successful group only; it gives no comparison.
  • (C) Founders' opinions of their own experience are not evidence that it made them more likely to succeed.
  • (E) More funding could be one route by which experience helps, so a yes fits the conclusion and a no does not refute it.

Tip: A share among winners is evidence only next to the same share among non-winners. Ask for the base rate.

Question 20 of 20FlawHard

Five years ago, 40 percent of the surgeries performed at Wexley Hospital were outpatient procedures, after which patients go home the same day; the rest required an overnight stay. This year, 55 percent of Wexley's surgeries were outpatient procedures. Hospital administrators conclude that Wexley now performs fewer surgeries that require an overnight stay than it did five years ago.

The administrators' reasoning is most vulnerable to criticism on the grounds that it

Answer: (C)

Why (C) is right. The share of overnight surgeries fell from 60 percent to 45 percent, but the conclusion is about their number. If total surgeries grew by more than a third, 45 percent of the new total exceeds 60 percent of the old one, and overnight surgeries rose. The administrators treat a falling share as a falling count.

The most tempting wrong answer: (A). (A) names the right concept, share versus total, but in the wrong direction. If the total shrank, overnight surgeries would have fallen even further, which helps the conclusion. Only growth in the total can sink it.

Why the other choices fail

  • (B) The conclusion is only about Wexley, so it does not generalize to other hospitals.
  • (D) The argument makes no claim about why the share changed.
  • (E) Later readmissions are not surgeries that require an overnight stay, so they do not affect the count in the conclusion.

Tip: Percent and number are different claims. When a conclusion is about a count, ask what happened to the total.

Score: 0 correct of 0 answered20 left · Results by type

Your results by question type

Answer questions above and your results fill in here.

Question typeIn setAnsweredCorrect
Assumption200
Strengthen400
Weaken400
Evaluate300
Inference200
Paradox100
Boldface300
Flaw100
LevelIn setAnsweredCorrect
Medium (1 to 10)1000
Hard (11 to 20)1000

Get full GMAT Focus practice in 2027

PrepDrills GMAT, the GMAT Focus practice app from PrepDrills and Epic Exam Prep, opens in 2027 with full-length practice exams, a Mistake Log and speed drills. Join the waitlist to hear when it opens.

Join the GMAT app waitlist

How to review the questions you missed

A miss is only useful once you can name why it happened. For each question you got wrong, read the most tempting wrong answer first: if that is the choice you picked, the explanation names the exact trap, such as a reversed conditional, a cause that supports the right claim through the wrong link, or a boldface portion read as the conclusion. Then redo the question cold a day later without looking at the explanation.

Look at your results table by type. Two or more misses in one type usually means the method for that type is not yet automatic, not that the questions were unlucky. Assumption misses often come from choosing a statement that is too strong; Strengthen and Weaken misses often come from supporting or attacking the wrong link; Boldface misses usually start with not finding the conclusion first. If your misses cluster at the end of a timed run instead, the issue may be pacing, which the guide on running out of time on the GMAT covers in detail.

Frequently asked questions about GMAT Critical Reasoning

Is Critical Reasoning part of the GMAT Focus Edition?

Yes. GMAC describes the Verbal Reasoning section as 23 multiple-choice Reading Comprehension and Critical Reasoning questions, to be completed in 45 minutes. In GMAC's words, Critical Reasoning questions measure the ability to make arguments, evaluate arguments, and formulate or evaluate a plan of action.

What types of Critical Reasoning questions are there?

Prep materials, including this set, sort Critical Reasoning questions by what the question stem asks you to do. The eight types in this set are Assumption, Strengthen, Weaken, Evaluate, Inference, Paradox, Boldface and Flaw. They all start from the same kind of short argument, so identify the type from the stem before you read the answer choices.

How much time should I spend on each Critical Reasoning question?

The Verbal Reasoning section gives you 45 minutes for 23 questions, which works out to just under 2 minutes per question on average. That average also covers Reading Comprehension, so treat it as a pace to practice against rather than a fixed limit for every question.

How do you solve GMAT assumption questions?

Find the conclusion, find the evidence, and name the gap between them. Then use the negation test: reverse a choice and ask whether the argument falls apart. If it does, the argument depends on that choice. If the argument survives, it does not. Be wary of choices with strong words such as mainly or most, which often claim more than the argument needs.

Are these official GMAT questions?

No. All 20 questions are original, written by PrepDrills in the style of GMAT Focus Critical Reasoning, and every company, town and study in them is invented. GMAC does not endorse them. For official questions, use GMAC's own practice materials.

How is the Verbal Reasoning section scored?

According to GMAC, Verbal Reasoning scores range from 60 to 90. The GMAT Focus Total Score, which ranges from 205 to 805, is built from the Quantitative Reasoning, Verbal Reasoning and Data Insights section scores, and each section is weighted equally.

What do Medium and Hard mean in this set?

They are PrepDrills difficulty labels, not official GMAC ratings. On a Medium question, the standard method for the question type leads cleanly to the answer. On a Hard question, at least one wrong choice sits close to the right one, for example a choice that supports the conclusion through the wrong link, or a boldface label that fits only one of the two portions.

GMAT and GMAT Focus are trademarks of the Graduate Management Admission Council (GMAC). GMAC does not endorse, and is not affiliated with, PrepDrills or the practice questions on this page.

JC

Jaclyn Caruana, MBA

Co-founder of Epic Exam Prep, founded in 2010, with offices in Barcelona, Madrid, Milan and Zurich plus online. The Epic Exam Prep team brings 25 years of combined expertise in test preparation.

Read full bio ›