Digital SAT Prep · Digital SAT Reading, Writing & Math

Command of Evidence

Selecting textual or quantitative evidence that supports a claim or conclusion.

Seven concepts. Digital SAT Command of Evidence rewards paraphrasing the claim, then selecting the quotation or data point that shares its subject, scope and direction — not a true fact about something else.

Digital SAT Prep
Medium level
7 concepts
5 practice questions

1Paraphrase the claim before you open the choices

Cover the answer choices, read the claim or hypothesis, and restate it in one plain sentence of your own. Only then hunt for the quotation or finding that says the same thing in different words. Paraphrasing first blocks options that are true, polished, or related but aimed at a different point.

If you cannot restate the claim, you do not yet know what the evidence must do — go back to the stem before browsing A–D.

Figure. Restate the claim in your own plain words before you touch the choices. A paraphrase locks subject, scope and polarity so shiny quotes cannot pull you off.

Claim then evidence

Cover choicesHide A–D so wording cannot pull you off the claim.
ParaphraseSay the claim in one plain sentence: who/what, what is asserted, and any comparison.
Hunt matchUncover options; pick the quote or finding that restates that sentence, not a neighbour topic.

Claim: Urban tree cover lowers summer street temperatures. Before looking at choices, which paraphrase is on-job?

Cities need more parks for recreation
More trees in cities make hot streets cooler in summer
Trees are beautiful and improve air quality
Summer heat is dangerous for pedestrians

The claim links tree cover to cooler summer streets. Option B keeps that link. Parks-for-recreation, beauty/air-quality and heat-danger are related urban topics that do not restate the temperature claim.

2The quotation must hit the same specific point

Textual evidence is the line that directly backs the claim's specific point — same actor, same action, same assertion. A nearby sentence about the same topic that never states that point is a planted near-miss.

Ask: if this quotation were the only sentence I had, would the claim still be licensed? If you still need an outside leap, the quote is related, not supporting.

Figure. Textual evidence must make the same point as the claim — not merely share a topic. Nearby true lines that talk about something else fail.

Point match

Name the pointFrom your paraphrase, isolate the one assertion the evidence must prove.
Test each quoteAsk whether the line alone forces that assertion — not merely shares a topic word.
Drop cousinsReject true lines about causes, side effects or other actors that never state the point.

Passage: "The committee delayed the vote because members lacked complete cost data. Public turnout that week was high. The committee meets every month." Claim: The vote was delayed because cost information was incomplete. Which sentence best supports the claim?

Public turnout that week was high.
The committee delayed the vote because members lacked complete cost data.
The committee meets every month.
Public turnout was high because the vote was delayed.

The delayed-because-incomplete-cost-data line alone licenses the claim's cause link. High turnout is a different variable. Monthly meetings are idle calendar fact. Turnout-because-delay reverses a causal direction the claim never asserts.

3Evidence must share the claim's subject and strength

Strong evidence has the same subject and the same strength as the claim. Overly broad data ("all cities") for a narrow claim ("this neighbourhood"), or a weak hedged finding for a strong certainty claim, fails the match even when the topic overlaps.

Scope mismatch is why a fact can feel supportive and still be wrong: it is about a bigger, smaller or stronger world than the claim allows.

Figure. Evidence must be as wide as the claim. A some-quote cannot carry an all-claim even when every word is true of the passage.

Scope traps
Claim shape	Mismatch	Why it fails
This school district…	National average…	Wrong subject width — country ≠ district
Often / tends to…	Always / proves…	Evidence overclaims certainty the stem never asks
Both species A and B…	Only species A…	Missing half the claimed subject
Raises yield under drought…	Raises yield in wet years…	Condition in the claim is not the condition tested

Claim: In this coastal town, weekend beach closures reduce local shop revenue. Which finding best matches the claim's scope?

Nationwide, tourism spending fell last year
In this coastal town, shop receipts dropped on weekends when the beach was closed
Coastal towns always lose money when any park closes
One shop owner prefers rainy weekends

Option B keeps this town, weekend closures and shop revenue. Nationwide tourism is too broad. Always/any park overclaims strength and widens the subject. One owner's preference is anecdote, not the revenue claim.

4Confirm support direction before you select

Evidence either makes the claim more likely (supports) or less likely (undermines). Read the stem's verb — supports, weakens, most strengthens — and check that the chosen finding moves the same way. A clean opposite trend is a frequent trap because it is about the right variables.

Direction is checked after subject and scope, not instead of them: right variables moving the wrong way still fail.

Figure. Polarity must match. A falling series cannot support a rose claim, and a rising series cannot support a fell claim.

Direction filter

Mark the verbSupports / strengthens versus weakens / undermines — write which way the answer must push.
Predict the moveState what result the claim needs (higher, lower, more often, less often).
Kill the flipDrop options that reverse that move even if they name the same variables.

Hypothesis: A tutoring program raises students' algebra scores. Which finding would undermine the hypothesis?

Students in the program scored higher than a matched group without tutoring
Students in the program scored lower than a matched group without tutoring
Students liked the tutors
Algebra is required for graduation

Undermine needs the opposite of raised scores → program students scored lower. Higher scores would support. Liking tutors and graduation rules do not test the score claim.

5For data claims, restate, locate, then verify direction

A graph or table plus a claim is still Command of Evidence: restate the claim, find the exact figure or trend the claim needs, and check that the number agrees in units and direction. Cover the answer choices while you read the data so option wording cannot invent a trend you did not see.

Supporting data must show the exact direction the claim predicts — a higher number for the treatment that is said to increase yield, not merely any number from the same study.

Figure. Locate the exact cell or axis value the claim needs, then check that the comparison word matches that number — not the trend you remember from a glance.

Data-support routine

RestateWrite the claim as a predicted comparison (new > old, dawn > dusk, and so on).
LocateFind the cells, bars or points that name both sides of that comparison.
VerifyCheck units and direction; pick the option that matches your reading, not a true fact about another series.

Claim: A new fertilizer increased crop yield. Study yields: old fertilizer 3.0 t/ha; new fertilizer 4.2 t/ha. Which result best supports the claim?

Old fertilizer produced 3.0 t/ha
New fertilizer produced 4.2 t/ha, higher than the old fertilizer's 3.0 t/ha
Both fertilizers were tested in the same soil type
Farmers preferred the new fertilizer's packaging

The claim needs higher yield under the new fertilizer. 4.2 > 3.0 supplies that direction. Stating only the old yield, naming a shared soil type, or citing packaging preference does not show an increase.

6Match evidence to both terms of the comparison

Many hypotheses are comparisons: more at dawn than at dusk, higher with treatment than without. Supporting evidence must speak to both terms. A finding about total songs per day, or about another species, never tests the dawn-versus-dusk claim even if it is true.

When the stem says "which finding would support", prefer the option whose result matches the predicted outcome on both sides of the comparison.

Figure. Higher-than is a two-way comparison; highest is a full ranking. Evidence for one does not automatically carry the other.

Both sides

List the polesName both sides the hypothesis compares (dawn vs dusk, with vs without).
Require bothKeep only findings that measure or contrast those same two poles.
Drop one-sidedReject totals, other species, or other times that omit a pole.

A researcher hypothesizes that a bird species sings more at dawn than at dusk. Which finding would best support this?

The birds produced twice as many songs at dawn as at dusk
The birds sang hundreds of songs each day
A different species sang mostly at dusk
Dawn temperatures were cooler than dusk temperatures

Twice as many at dawn as at dusk hits both poles in the predicted direction. Daily totals omit the comparison. Another species is the wrong subject. Temperature is a different variable.

7Reject evidence that is true but about the wrong variable

A choice can be factually accurate and still fail: it discusses a different variable than the claim. Discard accurate lines about popularity, cost, preference or a side outcome when the claim is about yield, scores or frequency.

Irrelevance is not "sounds unrelated" — it is "does not move the probability of this claim." True-but-idle options are planted because they reward skimming for familiar words.

Figure. A fact can be accurate and still fail if it answers a different variable than the claim. Discard true lines about cost or popularity when the claim is about yield.

True-but-idle patterns
Claim about	Idle true option	Why idle
Yield / score / rate	People liked the program	Attitude ≠ measured outcome
One species' behaviour	Another species' behaviour	Wrong subject
Cause named in the claim	Unrelated concurrent event	Same time, different variable
Predicted direction	Study methods were careful	Quality of study ≠ the result

Claim: Longer library hours increased student use of study rooms. Which option is true-but-irrelevant to the claim?

Study-room bookings rose after hours were extended
Students said they liked the new paint colour in the lobby
Evening bookings outnumbered morning bookings after the change
Study-room use stayed flat when hours were left unchanged in a comparison term

Paint colour can be true and still never addresses study-room use. Bookings rose, evening vs morning after the change, and the unchanged-hours comparison all speak to use or to the hours manipulation.

Notes

Textual Evidence: Given a hypothesis or claim, pick the quotation that directly supports it; the evidence must address the same specific point, not a related one.
Quantitative Evidence: A graph or table plus a claim requires the data point or trend that logically completes or backs the claim, matching units and direction.
Scope Matching: Strong evidence has the same subject and strength as the claim; overly broad or narrow data weakens the match.
Direction Check: Confirm the evidence moves the same way as the claim (supports vs. undermines) before selecting.

Formulas

Support test: chosen evidence must make the claim more likely and share its subject and scope.
Data-support routine: restate the claim, locate the exact figure, verify the number agrees with the claim's direction.
Reject evidence that is true but irrelevant to the specific claim being tested.
For 'which finding would support', prefer the option whose result matches the predicted outcome of the hypothesis.

Exam traps & shortcuts

Paraphrase the claim first, then hunt for evidence that says the same thing in different words.
In graph questions, cover the answers, read the data yourself, then pick the option that matches your reading.
Discard choices that are factually accurate but discuss a different variable than the claim.

Reference tables

Run claim paraphrase and match tests before browsing polished wording.

Command of Evidence playbook (night before)
Move	Do this	Avoid
Paraphrase	Restate the claim in one plain sentence with choices covered	Letting option wording redefine the claim
Same point	Quote alone must license the assertion	Topic-neighbour sentences that never state it
Scope	Keep subject width and strength aligned	National data for a town claim; always for often
Direction	Supports needs the predicted move; undermines needs the flip	Right variables moving the wrong way
Both poles	Comparison claims need evidence on both sides	Totals or other subjects that omit a pole
Relevance	Evidence must move this claim's probability	True facts about liking, cost or another variable

Recap

Keep these on exam day; each one kills a planted distractor.

Paraphrase: Cover choices; restate the claim in one plain sentence; then hunt.
Same point: The quote alone must license the claim — topic overlap is not enough.
Scope: Same subject width and strength; broad or overstrong data fails.
Direction: Supports vs undermines — kill the clean opposite trend.
Data: Restate → locate the figure → verify units and direction.
Both poles: Dawn vs dusk needs both times; totals and other species are traps.

Practise Command of Evidence

Reading is free and needs no account. Practice, mocks and progress live in the app.

5 exam-style questions on this topic, with explanations
A 5-question practice set that ends the chapter
Timed mocks scored with the real marking scheme
Readiness tracked per topic, kept on your device

Continue with Google — freeNo card, no trial. Works offline once installed.