The ASSUMPTION Is the Whole Game in Critical Reasoning

By EMPOWERgmat on September 12, 2017

Here is a scene I watch play out every single week. A Test Taker reads a Critical Reasoning argument once, half skimming, and jumps straight to the five answer choices. Two of them 'sound good.' He flips between them. He rereads the argument. He flips again. Then he picks the one that feels right, gets it wrong, and has NO idea why.

That is not a reading problem. It is an order of operations problem. He went shopping for an answer before he knew what he was shopping for.

What Critical Reasoning is really testing

Critical Reasoning (CR) is one of the three question types in the Verbal section, next to Sentence Correction and Reading Comprehension. Each CR question gives you a short argument, usually a few sentences, and then asks you to do a job with it: find the assumption, strengthen it, weaken it, or name the flaw. The Verbal section gives you 41 questions in 75 minutes, and it is adaptive question by question, so once you confirm an answer you cannot go back and change it. You can read the full layout on GMAC's page on the sections and timing.

Almost every CR argument is built the same way. There is evidence (the facts the author gives you), a conclusion (the claim the author wants you to buy) and a gap between them. The assumption is the unstated idea that has to be true for that evidence to actually prove that conclusion. The author never says it out loud. That is exactly why it is a gap.

Now stop and think about this for a second. An Assumption question asks you to find the gap. A Strengthen question asks you to fill it. A Weaken question asks you to pry it open. A Flaw question asks you to name it. Same gap, four different jobs. So if you can find the gap, you have already done the hard part of a HUGE share of the CR you will face.

Two cliff edges joined by a wooden bridge that is missing its middle plank, with a person standing beside the gap holding that plank, ready to fit it in. — The evidence and the conclusion are the two ends of the bridge. The assumption is the missing plank.

Find the gap BEFORE you look at the answers

I'll bet that on your last practice test, for most of the CR questions, you could not have told me the gap in one sentence before you read choice A. Go check your notepad. If it is blank next to those questions, there is your answer.

The four wrong answers are not random. They are written by people who know exactly what sounds good to a tired Test Taker. If you walk into the answer choices without your own prediction, they will pull you around. If you walk in with the gap written down, you are hunting for ONE thing, and most of the choices fall away fast.

Here is the process:

Read the question so you know the job.
Read the argument once and write the conclusion in a few words.
Write the evidence in a few words.
Stop. Ask: what has to be true for this evidence to prove this conclusion? Write it as 'Gap:'.
Only now go to the answers, looking for your gap.

Try it

In the two years since the town of Millbrook added bike lanes to its three busiest streets, the number of accidents involving cyclists on those streets has fallen by one third. The town council concludes that the bike lanes have made cycling on those streets safer.

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

In nearby towns that added bike lanes, accidents involving cyclists also fell.
Bike lanes are the most effective safety measure a town can add to its streets.
Accidents involving cyclists tend to cause more serious injuries than other traffic accidents.
The number of people cycling on those three streets has not fallen by one third or more over the past two years.
Some cyclists in Millbrook ride on streets that have no bike lanes.

Show the answer

Step 1, notes. On paper: 'C: lanes made cycling safer. E: cyclist accidents down 1/3.'

Step 2, find the gap. 'Safer' means less risk for each rider. Fewer accidents only means that if about the same number of people are still riding. Let's TEST VALUES. Say there were 300 riders and 30 accidents before. That is 1 accident for every 10 riders. Accidents fall by a third, so now there are 20. If there are still 300 riders, that is 1 in 15. Safer. But if riders fell to 200, then 20 out of 200 is 1 in 10 again. No change at all. And if riders fell to 150, then 20 out of 150 is worse than 1 in 10. MORE dangerous. So the note is: 'Gap: number of riders did not drop as much as accidents did.'

Step 3, hunt. Choice (D) says riders have not fallen by a third or more. That is our gap, almost word for word.

Step 4, confirm with the flip test. Turn (D) into its opposite: the number of riders HAS fallen by a third or more. Now the drop in accidents could just come from fewer people riding, and the council's conclusion falls apart. When the opposite of a choice wrecks the argument, that choice is the assumption. The answer is (D).

The trap. Most people who miss this pick (A). It sounds supportive and it might even be true in real life. Flip it: in nearby towns, accidents did NOT fall. Does that prove Millbrook's lanes failed? No. Millbrook's conclusion can still stand. The argument does not NEED (A). (B) is too strong: flip it and the lanes are not the single best measure, but they can still have made riding safer. (C) and (E) never touch the gap between 'fewer accidents' and 'safer.'

The mistakes that keep CR scores flat

Reading the answer choices before you know the gap. This is the big one. You let the test writer set the agenda.
INSTEAD: Write 'C:', 'E:' and 'Gap:' on your pad for every CR question before you look at choice A. Every one. No exceptions.
Picking the answer that is 'true' instead of the answer that is NEEDED. Choice (A) above is the classic. Plenty of wrong answers are perfectly reasonable facts.
INSTEAD: Ask one question of each choice: does the argument fall apart without this? Use the flip test on your final two.
Doing it all in your head. 'I can see the gap, I don't need to write it down.' That is pride talking, and pride is remarkably expensive on a test where you cannot go back.
INSTEAD: Write it. A few words. The note keeps you from drifting toward a shiny wrong answer after you have read all five.
Falling for big words. 'Always', 'most', 'only', 'never'. An assumption is usually the smallest thing the argument needs, not the biggest thing that would help it.
INSTEAD: When a choice sounds extreme, flip it. If the argument survives the flip, cross it off.
Blaming the clock. 75 minutes for 41 questions works out to a little under 2 minutes each on average, and people panic about that. But if you are rereading arguments three times, the clock is not the disease. The missing notes are.
INSTEAD: One careful read, notes as you go, gap written, then answers. That route is FASTER than skim, guess and reread.

When you review, look for the gap you missed

After a practice test, do not just mark a CR question wrong and move on. Go back and ask: did I write a gap? If I did, was it the right one? If I didn't, would I have caught the trap if I had? That is real analysis, and it is the kind I wrote about in Self-Analysis of Your GMAT Performance MUST Include Actual Analysis. 'I got it wrong' is not a diagnosis. 'I picked a true answer that the argument did not need' is.

Also train this method under real conditions. Since July 11, 2017 you can pick one of three section orders, so Verbal might come first or last on your Test Day. Either way, your note taking has to hold up when you are tired. If it falls apart late in the test, read Big Factors in Why Verbal Scores Can Drop on Test Day and fix that in practice, not on the real thing.

Your challenge for today

Pull 10 CR questions you have ALREADY done, ideally from GMAC's official practice materials. Cover the answer choices. For each one, write 'C:', 'E:' and 'Gap:' in a few words. Then uncover the choices and check how many times your gap pointed straight at the right answer.

If it matched fewer than half, good. You just found the exact skill to train, and it is a trainable one. Do another 10 tomorrow. The gap is always there. Your job is to see it before the test writer shows you something shinier.