LSAT Evaluate the Argument Question Tips

An evaluate question asks what you would need to know before you could judge an argument. The winning move is the variance test: take each answer choice as a question, answer it both ways, and keep the one where the two answers push the argument in opposite directions. If both answers point the same way, or neither changes anything, the choice is wrong.

How do you answer an LSAT evaluate the argument question?

First, find the conclusion and the gap in the argument, the same way you would for a strengthen or weaken question. Then, for each answer choice, ask: what would I learn if the answer were yes, and what would I learn if the answer were no? The correct choice is the one where yes helps the argument and no hurts it, or the reverse. That two-way swing is the whole test. Everything else on this page is just a way to run that test faster and avoid the traps.

What an evaluate question actually asks

Evaluate questions come in two stems. You will see one of these wordings, or something close to it:

• "Which one of the following would be most useful to know in order to evaluate the argument?"
• "In order to evaluate the argument, it would be most useful to know whether..."

Notice the shape of the answers. They are not claims. They are questions, usually starting with "whether." You are not being asked to supply a fact that proves the conclusion. You are being asked to name the missing fact that matters most. That missing fact almost always sits in the gap between the premises and the conclusion.

Evaluate questions are rare. Tutors often describe them as appearing at most about once per test, and some tests have none at all. That rarity is good news and bad news. Good news: they will not decide your score. Bad news: students who have never practiced the type freeze on it, because it looks like nothing else on the section. Learn the variance test once, and the rare question becomes a free point.

The variance test: the one method that decides these questions

Here is the method, stated precisely. For each answer choice:

1. Turn the choice into a yes-or-no question.
2. Suppose the answer is yes. Does the argument get stronger, weaker, or stay the same?
3. Suppose the answer is no. Does the argument get stronger, weaker, or stay the same?
4. Keep the choice only if yes and no push in opposite directions.

Three ways a choice fails the test:

• Both answers help. The information is good news either way, so it cannot settle whether the argument works. Eliminate it.
• Both answers hurt. Same problem in reverse. Eliminate it.
• Neither answer matters. The information is about something the argument does not depend on. Eliminate it.

The most common error is picking a choice that sounds relevant but only moves the argument one way. Relevance is not enough. The question asks what would be most useful in evaluating the argument, and only a two-way swing actually evaluates it.

A worked example

Consider this original argument, written in LSAT style:

"A town that replaced its downtown parking meters with a mobile payment app saw a 20 percent rise in downtown shopping visits over the following year. The app therefore caused the increase in shopping visits."

The conclusion is causal: the app caused the rise. The gap is everything the premises leave out. Maybe a new shopping center opened that year. Maybe visits were already rising. Maybe the count changed. Now consider two candidate evaluate questions:

• "Whether any other change affecting downtown shopping happened during that year." If yes, the causal claim weakens, because something else could explain the rise. If no, the claim strengthens, because the main rival explanation is gone. Two-way swing. This is the shape of a correct answer.
• "Whether shoppers liked the app." If yes, the story feels nicer, but it does not show the app caused the rise. If no, the rise still happened, so the claim is barely dented. No real swing. This is a trap answer: related to the topic, useless for the evaluation.

Run every choice through that same yes-and-no exercise. The right answer usually announces itself within seconds once you stop asking "is this interesting?" and start asking "does the answer change my verdict?"

The 4-step routine

1. Find the conclusion. Ask what the author wants you to accept. If you evaluate the wrong claim, every choice will look plausible. Our guide to finding the conclusion covers the hard cases, including conclusions buried mid-argument.
2. Name the gap in plain words. Say it as a sentence: "This assumes X." The gap is usually a missing link: a rival cause, a shaky sample, a term that shifted meaning, or a necessary condition nobody checked.
3. Prephrase the question you would ask. Before reading the choices, finish this sentence: "I would need to know whether..." Your prephrase will often match the correct choice almost word for word.
4. Run the variance test on the survivors. Two or three choices usually survive a first pass. Apply the yes-and-no test to those, and pick the one with the cleanest swing.

On timing, treat evaluate questions like any other Logical Reasoning question: roughly 80 to 90 seconds on average, with more available if you banked time on easier early questions. If one stalls you past about 90 seconds, mark your best guess, flag it, and move on. Since the type is rare, it is never worth sacrificing two later questions for one evaluate question.

Evaluate vs. its lookalikes

Evaluate questions borrow language from three neighboring types. Knowing the difference keeps you from importing the wrong method.

The useful connection: the correct evaluate answer often points at the argument's biggest unstated assumption. If you can spot the necessary assumption, you can usually predict the evaluate answer. The difference is only the packaging. A necessary assumption answer states the assumption as a fact. An evaluate answer asks whether it is true. Our necessary vs. sufficient assumptions guide explains that distinction in full, and our flaw question guide covers the gap-spotting skill that powers all of these types.

Traps that catch good students

• The interesting-but-irrelevant fact. A choice can be about the argument's topic and still fail the variance test. Shoppers liking the app is interesting. It does not decide whether the app caused the rise.
• The one-way helper. A choice whose yes answer strengthens the argument but whose no answer changes nothing is a strengthen answer in disguise. It fails the test.
• The premise re-test. A choice that asks about something the stimulus already tells you. If the information is already given, learning it again evaluates nothing.
• The scope shift. A choice about a different group, time period, or place than the conclusion covers. The answer could swing something, just not this argument.
• The extreme version. A choice that asks about an exaggerated version of the gap ("whether the app was the only possible cause"). The argument rarely needs something that strong, so the choice overreaches.

How to practice evaluate questions

Because the type is rare, do not drill it in isolation for hours. Instead, fold it into your normal Logical Reasoning practice:

1. When you review any strengthen, weaken, or flaw question, ask yourself the evaluate version: "What would I need to know to settle this argument?" That turns every LR question into evaluate practice.
2. For each miss, write the gap in one sentence in your miss log, then write the evaluate question that would test that gap. Our mistake-tracking guide shows the log format that makes this stick.
3. Once you can run the variance test without thinking, stop drilling the type. Your time is better spent on flaw, strengthen, and assumption questions, which appear far more often. The full type list lives in our Logical Reasoning question types guide.

As of October 2026, the current LSAT format has two scored Logical Reasoning sections, so the rare evaluate question can appear in either one. The method does not change between sections. Find the conclusion, name the gap, prephrase the question, and run the variance test.