LSAT Flaw questions: naming the mistake in the abstract

A Flaw question shows you a broken argument and asks you to say what is broken about it. The stem usually reads "the reasoning is most vulnerable to criticism on the grounds that it" or "the argument is flawed in that it". You are not asked whether the conclusion is true. You are asked what the author did wrong on the way to it.

That distinction is the whole difficulty. A conclusion can be perfectly correct and still be reached by terrible reasoning, and this section only ever cares about the reasoning. Students who argue with the content — "but that IS how groundwater works" — miss these consistently, because the credited answer is never about groundwater.

Why the answers are written so abstractly

Flaw answers describe the error in general terms: "takes a condition sufficient for an outcome to be necessary for it", "draws a conclusion about a population from a sample unlikely to represent it", "treats the failure to prove a claim as proof that it is false". None of them mention the topic of the passage. This is deliberate, and it is why the section rewards vocabulary.

The consequence for study is direct. If you can spot that an argument moved from a correlation to a cause, but you cannot recognize the sentence "presumes, without warrant, that one of two correlated phenomena must be the cause of the other" as describing that move, you will read past the right answer. Learning the abstract phrasings is not vocabulary for its own sake — it is learning to recognize your own thought written in someone else's words.

The flaws that repeat

A small set of errors covers most of what the section tests. They are worth knowing by name, because naming one before you read the choices turns a five-way comparison into a lookup.

• Correlation treated as causation — two things happen together, so one is said to cause the other. The counter is always an alternative explanation, a reversed direction, or a common cause behind both.
• Unrepresentative sample — a conclusion about everyone drawn from a group unlikely to stand in for everyone. Self-selected respondents are the classic version.
• Necessary confused with sufficient — treating a requirement as a guarantee, or the reverse. Watch for "only if" doing quiet work.
• Part and whole — what is true of each piece is assumed true of the assembly, or the reverse.
• Equivocation — a word changes meaning between the evidence and the conclusion, and the argument only works if you fail to notice.
• Attacking the source — the arguer's motive, character, or hypocrisy is offered where an answer to their claim was needed.
• Absence of evidence — nobody has shown it, therefore it is false. Or: nobody has disproved it, therefore it is true.
• Percentage against number — a rate rises while the count falls, or the reverse, and the argument treats them as interchangeable.
• False dilemma — two options are presented as exhaustive when a third is available.
• Circularity — the conclusion appears among the evidence, usually reworded well enough that it takes a second read to catch.
• Missing comparison — something is called high, effective, or improved with nothing stated to compare it against.

The habit that makes these fast

Read the argument, then say the flaw out loud in ordinary language before touching the answers. "He surveyed people who already liked the product." "She's assuming nothing else could have caused the drop." Ugly phrasing is fine. The point is to fix your own judgment before five well-written sentences get a chance to move it.

Then read the choices looking for a translation of what you already said. Four of them will describe errors the argument did not make — often real errors, described accurately, just not this argument's. That is the most common trap in the whole question type: an answer that names a genuine fallacy which simply is not present here.