What is the difference between command of evidence and inferences questions?

Inferences questions ask what conclusion the passage's information supports. Command of evidence questions give you the claim already and ask you to find the outside fact or data point that would prove it. The claim is fixed in command of evidence; you are just finding its support.

Do I need outside knowledge to answer these?

No. Everything you need is either in the passage's claim or in the answer choices themselves. You are reasoning about logical support, not recalling facts from outside the test.

Why is the most surprising answer choice usually wrong?

Because these questions are designed to test precision, not general knowledge. An eye-catching fact about the topic is often included specifically to tempt students who are scanning for relevance instead of checking whether the fact proves the exact claim stated.

How should I handle a table with several columns of numbers?

Identify the one comparison the claim is actually making, such as a difference between two groups or a change over one time period, and check only the columns needed for that comparison. Reading every number in the table before looking at the claim wastes time.

SAT Command of Evidence: How to Pick the Right Support

The reader wants to know what a command of evidence question is actually asking for and how to eliminate wrong answer choices that look like they could work.

What a command of evidence question looks like

You get a short passage that states a claim, hypothesis, or finding, often attributed to a researcher or study. The question then asks which of four options, if true, would most support that claim. In the data version, instead of four text quotations you get a table or graph, and the four answer choices describe different numbers or trends drawn from it.

Either way, the passage itself usually does not contain the answer. You are being handed a claim and asked to reason about what kind of evidence would actually back it up, or to read a data set and pick the row or trend that matches the claim stated in the passage.

The rule that decides the answer

Underline or restate the exact claim in your own words before you look at the options. Command of evidence questions are precise: the claim might be about one specific group, one specific time period, or one specific comparison, and an answer choice that supports a nearby but different claim is wrong.

Then treat each option like a yes or no test: does this specific piece of evidence make the specific claim more likely to be true? Throw out anything that is just related to the general topic. Throw out anything that would support a different, similar-sounding claim. What is left is the answer.

The trap most students fall into

The trap is picking the option that is the most interesting or surprising fact about the topic, rather than the fact that actually proves the claim in front of you. Test writers include at least one answer choice that is true, real-sounding, and about the same subject, but that does not logically connect to the specific claim being tested. If the claim is 'the new fertilizer increased crop yield in dry climates,' a true statement about the fertilizer's cost, or about crop yield in general regions, does not support that specific claim even though it is on-topic.

In the data version, the same trap shows up as an answer choice that describes a real number from the table but does not match the direction or the specific comparison the claim requires. If the claim is about a difference between two groups, an answer that only describes one group's number, without the comparison, does not support it.

Worked example

Passage: 'A researcher hypothesized that students who took short breaks every 25 minutes retained more information than students who studied in one long block.' Options describe four different findings from the researcher's study.

First, restate the claim: short breaks every 25 minutes leads to better retention than one long block, specifically compared to studying without breaks. An option saying 'students who took breaks reported enjoying studying more' does not support the claim, because enjoyment is not retention. An option saying 'students in both groups spent the same total number of minutes studying' does not support the claim either, it just controls a variable. An option saying 'students who took breaks scored higher on a retention test one week later than students who studied in one long block' directly supports the claim, because it measures retention and compares the two exact groups named in the hypothesis. That is the answer.

Reading the data version quickly

For a table or graph, first find what the passage's claim predicts should happen in the data, in plain terms like 'group A should be higher than group B' or 'the number should go up over time.' Then scan the table only for the specific rows or columns that test that prediction. Ignore rows that are true but irrelevant to the comparison the claim makes.

StudyPebble's practice set for this skill includes both the text and data versions, with each explanation walking through why the correct choice matches the claim and why the closest wrong choice falls just short of it.