Inference questions belong to Information and Ideas.

A common stem is:

Which choice most logically completes the text?

The trap is thinking “inference” means guessing what is probably true.

On the SAT, the best inference is the choice most strongly supported by the passage, even if the passage never states it word-for-word.

The minimum-proof rule

Use this standard:

Choose the answer that requires the fewest unsupported assumptions.

The correct inference should feel like a small step beyond the text—not a leap.

Explicit vs inferred

Explicit:

The lake froze three weeks earlier than its historical average.

Reasonable inference:

Conditions that year favored earlier freezing.

Unsupported leap:

Climate change has permanently reversed in the region.

The passage does not justify that conclusion.

The evidence chain

For every inference, identify:

1. Fact 1
2. Fact 2
3. Logical link
4. Conclusion

Example:

Fact 1: Birds exposed to urban noise sang at higher frequencies.

Fact 2: High-frequency sounds were less masked by low-frequency traffic noise.

Inference: The frequency shift may help the birds' songs remain detectable in noisy environments.

That inference links the two facts directly. That kind of logical link functions much like the connecting words covered in our transitions guide.

Avoid “could be true”

SAT wrong answers are often plausible.

The question is not:

Could this happen?

It is:

Does this passage support it?

A choice can be scientifically reasonable and still unsupported by the text.

Pay attention to degree

Passage:

The treatment increased growth in 7 of 10 samples.

Supported:

The treatment increased growth in most samples.

Too strong:

The treatment always increases growth.

Inference questions punish exaggeration.

Conditional language

Words such as:

• may
• suggests
• likely
• could
• tends to

often match cautious evidence.

That does not mean the weakest answer is always correct. It means the answer's strength should match the text.

Numerical inference

Suppose:

Group A averaged 14.2 units. Group B averaged 9.7 units.

Supported:

Group A's average exceeded Group B's.

No advanced interpretation needed.

But if the passage does not give variability or significance testing, do not automatically infer that the difference is statistically significant.

Experimental design inference

Text:

Plants receiving fertilizer and plants receiving no fertilizer were grown under identical light and watering conditions. Fertilized plants grew taller on average.

Reasonable:

The results are consistent with fertilizer contributing to greater growth under the tested conditions.

Too strong:

Fertilizer guarantees greater growth in all plant species.

Literature inference

Inference is not only scientific.

If a character:

• repeatedly avoids a topic;
• changes the subject when asked;
• hides a letter;

the passage may support an inference that the character is reluctant to discuss the subject.

But do not diagnose motives beyond what the text supports.

Use contradiction elimination

For each answer ask:

Can I point to something in the text that makes this hard to believe?

Eliminate:

• contradictions;
• unsupported causes;
• scope expansions;
• extreme language;
• answers using correct topic vocabulary but wrong relationships.

Common traps

Trap 1: outside knowledge

You know something about the topic, so you choose it.

Ignore what you know unless the passage supports it.

Trap 2: causal leap

Correlation → causation.

Trap 3: universal claim

One study → all cases.

Trap 4: emotional overreading

A fictional character hesitates → “is terrified.”

Maybe. But does the text prove that level of emotion?

Trap 5: reverse logic

If A predicts B, do not automatically claim B predicts A.

Worked example

Text:

Researchers found that seedlings exposed to intermittent wind developed thicker stems than seedlings grown in still air. The windy and still-air groups received equal light, water, and nutrients.

Best inference:

Mechanical stimulation from wind may contribute to thicker stem development under these conditions.

This is directly tied to the controlled difference.

Too strong:

Wind is the only factor that determines stem thickness.

Original mini-practice

1

Text:

A historian compared store advertisements from 1910 and 1930. The later advertisements devoted substantially more space to illustrations and less to dense descriptive text.

Which inference is best supported?

A. Consumers in 1930 could not read long texts.
B. Advertising design shifted toward greater visual emphasis between the two periods.
C. Illustrations were cheaper to print in 1910.
D. All 1930 advertisements contained no descriptive text.

2

Text:

In a laboratory test, material X retained 96% of its initial strength after 1,000 bending cycles, while material Y retained 71%.

A. Material X was more resistant to strength loss under the tested bending conditions.
B. Material X is superior for every engineering application.
C. Material Y cannot bend.
D. The materials initially had identical strength.

3

Text:

When asked about the missing notebook, Aria immediately began discussing the weather and then left the room.

Best inference?

A. Aria definitely stole the notebook.
B. Aria may be reluctant to discuss the notebook.
C. Aria dislikes all conversations.
D. The notebook never existed.

Answers

1. B
2. A
3. B

A fast inference checklist

Before choosing:

1. Can I underline evidence for this?
2. Am I adding a cause not established?
3. Is the answer stronger than the passage?
4. Does it expand one case into all cases?
5. Is it merely plausible, or actually supported?

How to practice

After each inference question, write:

Evidence → conclusion

Example:

later ads use more illustrations → greater visual emphasis.

If you cannot make that arrow clearly, your answer may be too speculative. The same evidence-to-conclusion habit carries over to rhetorical synthesis questions, where you combine several bullet points into one accurate claim.

What to learn next

• Command of Evidence
• Central Ideas
• Cross-Text Connections
• Words in Context

If sentence-level grammar is a separate weak spot, see the sentence boundaries guide. Browse the rest of our Reading & Writing guides for more Information and Ideas skills.

Official references

• College Board, content domains: https://satsuite.collegeboard.org/practice/content-domains
• College Board, Question Bank Reading and Writing skills: https://satsuite.collegeboard.org/practice/student-question-bank/reading-writing

---

Trademark note: SAT® is a registered trademark of College Board. MastaPrep is not affiliated with, endorsed by, or sponsored by College Board.