# SAT Inferences: the conclusion the passage forces

Inferences items end in a blank and ask which choice most logically completes the text. Why the tempting option is one step too far, three worked items to attempt, and the limiting phrase that decides most of them.

## The short answer

Inferences items end with a blank and ask which choice most logically completes the text. The passage hands you a few facts, and the answer is the conclusion those facts force. The option most people choose instead is reasonable in the world and one step past what the passage supports. Inferences is one of four Information and Ideas skills, about 26% of the section.

## Key facts

- **Section:** Reading and Writing
- **Domain:** Information and Ideas
- **Domain share:** ~=26% of the section, 12-14 questions
- **Passage length:** 25-150 words
- **Options:** 4 (A-D)
- **If you are unsure:** Guess rather than leave it blank

Question counts, timings and domain shares come from College Board's test pages and the Digital SAT Suite Specifications Overview, checked 28 September 2026.

## How it's scored

The blank almost always falls at the end of the passage, and the prompt is a standard line: which choice most logically completes the text?

Information and Ideas is about 26% of Reading and Writing, 12 to 14 of the 50 operational questions. The domain covers this skill, Central Ideas and Details, and both Command of Evidence skills.

> The right answer says no more than the passage allows. If an option needs one extra fact to be true, that missing fact is why it is wrong.

## Inferences examples

Worked examples written for this page. Answer each one before you open the model response: memorizing a response is scored as an irrelevant answer in the real test.

**Example 1 -- a word borrowed from the passage**

> Medieval European ironworkers made steel in small batches, and the carbon content of each batch varied widely. Blades were therefore judged by bending them rather than by any measure of what they contained. When crucible steel of consistent carbon content reached European workshops in the eighteenth century, toolmakers began naming a hardness in written contracts. This suggests that ______

Prompt: Which choice most logically completes the text?

- A. medieval ironworkers took little interest in the quality of their steel.
- B. written specifications became useful only once the material they described was predictable.
- C. bending was a more reliable test of steel than measuring carbon content was.
- D. crucible steel was harder than any steel made in medieval Europe.
Answer: B.

Model response: B. Contracts naming a hardness appear alongside steel of consistent carbon content. Before that, a number in a contract would have described only one batch.

Why it scores: D borrows the word hardness from the passage and moves past its logic. The contrast being drawn is consistency, and consistency is not the same thing as hardness.

**Example 2 -- turning not yet into never**

> A wetland restored in 2005 now holds more plant species than an untouched reference wetland nearby. Its insect community, though, is still about half as diverse as the reference site's. Soil samples show that carbon in the restored soil is rising and has not yet reached reference levels. Ecologists reviewing the site conclude that ______

Prompt: Which choice most logically completes the text?

- A. plant diversity can recover faster than the conditions other groups depend on.
- B. the restoration has failed to meet any of the goals set for it.
- C. insect diversity at the restored site will never match the reference site.
- D. planting more species is the fastest route to a full recovery.
Answer: A.

Model response: A. Plants are ahead of the reference site while insects and soil carbon lag behind. That is a difference in pace, not a verdict.

Why it scores: C turns not yet into never. The soil figures are still rising, which says the site is changing rather than stuck at half.

**Example 3 -- generalizing past the sample**

> A national survey asked firms whether they had raised prices in the past year, and 68 percent said they had. The survey sampled only firms with more than 500 employees. Smaller firms account for most of the country's output. An economist reading the result warns that ______

Prompt: Which choice most logically completes the text?

- A. the figure may not describe price setting across the economy as a whole.
- B. most firms in the country raised their prices in the past year.
- C. large firms raise prices more often than small firms do.
- D. price increases were smaller than 68 percent of firms had expected.
Answer: A.

Model response: A. The 68 percent covers large firms only, and the passage says smaller firms produce most of the output. So the figure cannot stand for the whole economy.

Why it scores: B is the sentence the number seems to say out loud. The sample holds only large firms, and C asks for a comparison the survey never made.

**Three questions is a warm-up, not practice** 16,000+ Digital SAT practice questions, 20+ full adaptive mocks and 20+ sectional tests on a Bluebook-style screen. One full Digital SAT mock and one sectional test per module are free, with question-bank practice and 100 AI scoring coins a month. No card required. [Start practicing free](https://app.testmigo.com/register)

## Common mistakes

| What costs marks | Do this instead |
| --- | --- |
| Choosing the option that sounds wisest. | The item is not asking for a good opinion. It wants the sentence the given facts make necessary. |
| Turning a current limit into a permanent one. | Watch for "not yet", "so far" and "still rising". None of them support the word never. |
| Matching a word instead of the logic. | An option that reuses a noun from the passage is not supported by that. Ask what the passage claims about the noun. |
| Generalizing past the sample. | Check who or what was measured. A result about large firms, or about one catchment, does not describe everything. |

**Turn correct answers into a score** The SAT score calculator maps correct answers per module onto a 400-1600 range with College Board's own table, and superscores across test dates. [Open the SAT score calculator](/sat/score-calculator)

## How to practice

1. Finish the sentence yourself before you look at the options. Write it down, even in five words.
2. For each option you reject, name the word that made it too much: every, never, all, most, because.
3. Underline the limiting phrase in every passage. There is usually one, and it usually decides the item.
4. Review each miss by asking what extra fact your option needed. That fact is the gap the item was testing.

## Common questions

### What does "most logically completes the text" mean?

It means the conclusion the passage supports, with nothing added. The blank comes at the end, and the wrong options ask you to supply a fact the passage never gave you.

### Is an inference the same as a main idea?

No. A main idea is stated across the passage. An inference is not stated at all: it is what follows from what is stated. Both skills sit in the Information and Ideas domain.

### How much of the SAT is Inferences?

College Board reports the domain rather than the skill. Information and Ideas is about 26% of the Reading and Writing section, 12 to 14 of its 50 operational questions, and Inferences is one of four skills inside it.

**Sources**
- [College Board: what's on the SAT Reading and Writing section (checked 28 September 2026)](https://satsuite.collegeboard.org/sat/whats-on-the-test/reading-writing)
- [College Board: Digital SAT Suite of Assessments Specifications Overview (PDF)](https://satsuite.collegeboard.org/media/pdf/digital-sat-test-spec-overview.pdf)
- [College Board: how SAT scores are calculated (checked 28 September 2026)](https://satsuite.collegeboard.org/sat/scores/understanding-scores/how-scores-are-calculated)

[Practice Inferences free](https://testmigo.com/sat)