Free Courses › Logical Reasoning

The big question types · Lesson 8 of 14

Flaw questions

After this lesson, you can run the 5-step flaw method, from finding the conclusion to matching your predicted flaw to an answer choice, and label the break in an argument's logic in 2 or 3 words.

26 min video, plus notes and 1 practice question below

In this lesson

  • Recognize a flaw question from its stem before you even read the stimulus
  • Run the 5-step process: conclusion, evidence, the golden question, prediction, matched answer
  • Label the flaw in 2 or 3 words instead of a full sentence, so you can predict the right answer instead of reacting to it
  • Recognize the 6 flaw patterns that account for most flaw questions, including bad comparisons and correlation mistaken for causation
  • Catch the 5 trap answer types the LSAT reuses to pull points from readers who skip the prediction step

Why flaw questions are worth mastering first

Flaw questions show up 5 to 8 times across both scored logical reasoning sections, up to 20% of your logical reasoning points coming from one question type. They're grounded in a single skill, seeing where an argument breaks down, which makes them teachable and repeatable in a way vaguer question types aren't.

What they're really testing is whether you can tell a conclusion doesn't follow logically from its evidence, even when the argument sounds persuasive. Every argument gives you evidence, a fact or observation, and a conclusion, the judgment the author wants you to accept. Your job is to check whether the evidence actually proves that conclusion, and if it doesn't, find the assumption holding the gap together.

Recognize the question before you read the stimulus

Flaw questions use direct language: "which one of the following most accurately describes a flaw in the reasoning above," "the argument is most vulnerable to criticism on the grounds that," or "the reasoning in the argument is most flawed because." Some swap in softer phrasing like "vulnerable to criticism" or "questionable technique" instead of the word "flaw" itself, but they're asking the same thing.

Read the question stem before the stimulus. Once you know you're looking at a flaw question, read the argument like a detective looking for where it falls apart, not for whether you agree with it. That mental shift alone can save 10 to 15 seconds per question.

The 5-step process

Step 1, ID the conclusion: the one sentence the author is trying to prove, usually flagged by words like "therefore," "thus," "so," or "it follows that." Step 2, find the evidence used to support it, treating every premise as true even if you'd disagree with it in real life. Step 3, ask the golden question: does the evidence actually prove the conclusion? If not, what assumption is the author making without proving it? That gap is the flaw, and it helps to write a 2 or 3 word label for it right on your scratch paper.

Step 4, read every choice like a lawyer, asking only whether it matches your prediction, not whether it sounds smart. Step 5, choose the answer that best describes the gap, even if it doesn't use your exact words. For example, an argument that says "I studied every day for a month and got a 170, so if you study every day you'll get a 170 too" assumes your situation matches theirs in every way that matters. Nothing proves that, and that unproven assumption is the flaw.

6 flaw patterns that show up again and again

The lesson on LSAT flaw patterns walks through 10 flaws that show up across logical reasoning generally. These are the 6 that turn up most in flaw questions specifically, some the same, some new, and the 5-step method above finds any of them, named pattern or not. Bad comparisons treat 2 different things as equivalent, like concluding cooking at home is better than eating out just because it's cheaper. Correlation mistaken for causation assumes one thing caused another just because they happen together, like assuming playing an instrument improves LSAT scores because instrument players score higher. Conditional logic mixups treat a necessary condition as if it guarantees the result, like assuming doing well on the LSAT means someone went to Harvard.

A biased or unrepresentative sample draws a big conclusion from a small or skewed group, like judging what most law students think from 5 classmates. Confusing numbers and percentages treats a higher percentage as proof of a higher raw count: a firm that extends offers to 80% of 5 interns still hires fewer people than one that extends offers to 60% of 100 interns. Circular reasoning uses the conclusion as its own evidence, like saying a law is unfair because it's unjust, which just restates the claim instead of proving it.

5 trap answers that steal points from careful readers

The wrong flaw describes a real logical error, just not the one in this argument, so always check it against your own prediction. Out of scope answers introduce an idea the argument never raised, like nutrition showing up in an argument about cost. Too strong or too weak answers push a real issue to an extreme with words like "always" or "never," or water it down with "might" or "could" when the argument's own language was more direct.

An answer that's true but not actually flawed points out something accurate about the argument that doesn't weaken it, like noting an assumption that turns out to be reasonable. And an answer that mislabels the argument's structure describes a survey or sample that never appears in the stimulus at all. Predicting the flaw before you read the choices is what makes all 5 traps easy to spot instead of tempting.

Try it yourself

A company's CEO argues: our new office chairs are more comfortable than our old ones, so employee productivity will increase this year. Which one of the following most accurately describes a flaw in the reasoning above?

Show the answer

Answer: The argument assumes that comfort is the only factor affecting productivity, when other factors like workload, staffing, or training could also explain a change in productivity.

The evidence (the new chairs are more comfortable) only supports a claim about comfort, not one about productivity. The argument never rules out other explanations for the productivity change, the same correlation-does-not-equal-causation flaw as the instrument-and-LSAT-score example in the lesson.

The short version

  • When an argument sounds confident and well-written, that's not proof it's valid, don't let good writing talk you out of hunting for the gap.
  • Read the stem before the stimulus, every single time, even when you're in a groove and tempted to skip it. That's the habit that saves you 10 to 15 seconds a question.
  • After every flaw question you miss, log which of the 6 patterns it was, or note that it wasn't any of them. The pattern that keeps showing up in your log is the one to drill next.

Free download

The Logical Reasoning companion pack

Every lesson’s notes, key rules, and practice questions in 1 PDF, emailed to you. The videos teach. The pack is what you review from.

I’ll also send you my best LSAT strategies. Unsubscribe anytime.