The most common question type in the Reading and Writing section (15.3% of the bank). Covers both its textual and quantitative forms, stating the claim in your own words before reading any choice, reading axis labels and units before a data choice, and rejecting choices that are true and on topic but support a different claim — with six worked examples, four traps and three checks.
Subject: SAT Prep · 61 slides · applied lesson
Open the interactive version of this deck
Title
SAT Reading & Writing · Type 1 of 10
15.3% of the question bank — 258 of 1688 questions
Objectives
You are given a claim, a hypothesis or a conclusion, and asked which piece of evidence would most strongly support it. The evidence is either a quotation from the text or a data point from a table or graph. Both versions test one thing: whether you can hold a claim precisely enough to measure four candidates against it.
One sentence governs the type: the claim is the ruler, and accuracy is not the same as relevance.
Naruhodo Tutoring — SAT Reading and Writing question bank export (sat-english-question-index.json) — 258 tagged questions of this type in the site's bank
Section
Section 1
Concept
You are given a claim, a hypothesis or a conclusion, and asked which piece of evidence would most strongly support it. The evidence is either a quotation from the text or a data point from a table or graph. Both versions test one thing: whether you can hold a claim precisely enough to measure four candidates against it.
You will see it phrased in these ways:
The two forms differ only in where the evidence comes from. In both, the claim is the ruler and every choice is measured against it.
Picture it
The same three-panel card as the survey deck, so the shorthand carries over: what identifies it, what you write first, and what is built to catch you.
Figure (svg): Command of Evidence: the tell, the move, and the trap
Note what the red panel does not say. The wrong answers are not false and not off-topic — they are accurate statements answering a question nobody asked.
Prediction
The distinction the whole type rests on.
Predict first
A claim states that a new fertiliser increases yield. Which finding supports it?
Correct: Fields treated with the fertiliser produced 20 per cent more grain than untreated fields
Why: The claim is specifically about YIELD, so only a finding about how much was produced can support it. The other three are entirely plausible facts about the fertiliser — cost, ease of use, regulatory approval — and none of them speaks to yield at all. This is the whole type in miniature: three true, on-topic statements that support different claims, and one that addresses the stated one.
Concept
Two faces, one skill.
| variant | where the evidence comes from | the extra step |
|---|---|---|
| Textual | a quotation from the passage | check the quotation actually asserts the claim |
| Quantitative | a table, bar chart or graph | read the axes and units before the choices |
| Complete the statement | data, filling a blank in a sentence | the sentence sets the claim; the data must fit it |
The quantitative version is the one students fear and the easier of the two, because a number either matches the claim or it does not.
Definition probe
The wrong answers fail in recognisable ways.
Sort into buckets
A claim says a bird species nests earlier in warmer springs. Why does each choice fail or succeed?
Discrimination
A claim states that a lake's fish population has declined since 2010.
Sort into buckets
What does each finding do?
Warm-up
Do not read the choices yet.
Discussion prompt
A text says: researchers hypothesised that the difference in species counts arose largely from disagreement about how to classify microorganisms. In your own words, what must a supporting finding show?
Hint: What exactly is being asserted?
Answer:
It must show that classification decisions, rather than genuine biological change, account for the difference in counts.
So a supporting finding would be something like: the two teams applied different classification criteria to the same organisms, and the gap narrows when a single criterion is applied.
What would NOT support it, however true: that the surveys were ten years apart, that new species have been discovered, or that one team surveyed a larger area.
Each of those is a genuine fact and each would explain the difference by a DIFFERENT mechanism — which is precisely how the distractors are built.
Writing that one sentence before reading the choices is the whole technique.
Pattern
Three steps, and the first happens with the choices covered.
Write the claim in your own words as one sentence, before reading any choice.
Why: The claim is the standard every choice will be measured against, and reading the choices first quietly reshapes it.
For a data question, read the axis labels, units and categories before anything else.
Why: A choice can misread the graph in a way that is invisible unless you know what the axes actually measure.
Test each choice against your sentence, asking not is this true but does this prove THAT.
Why: Every wrong choice is true. The question is entirely about relevance to the specific claim.
Step 3 is where the discipline lies. A choice that is accurate and interesting must still be rejected if it proves something else.
Section
Section 2
Concept
Evidence is judged only by whether it supports the specific claim stated, not by whether it is true or interesting.
This is why the type is called command of evidence rather than comprehension. You are being asked to judge fit, not accuracy.
Concept
On a quantitative question, establish what the graph measures, in what units, for which categories.
| read this first | why it matters |
|---|---|
| The horizontal axis label | tells you what the categories or times are |
| The vertical axis label | tells you what is being measured |
| The units | thousands, percentages and rates change every reading |
| The legend | distinguishes multiple series that look alike |
The commonest quantitative distractor is a real number from the graph paired with the wrong label.
Prediction
Both, or the choice fails.
Predict first
A claim says a museum's evening opening increased attendance. Which finding supports it?
Correct: Attendance on evenings the museum opened late was 30 per cent higher than on comparable days
Why: The claim is about ATTENDANCE, so only a finding measuring attendance can support it. Positive reviews, staff costs and expansion to other cities are all plausible facts that speak to satisfaction, expense and strategy — three different claims. This is the standard construction: one choice on the stated variable, three on adjacent ones.
Concept
A finding that would EXPLAIN the claim is not the same as a finding that shows the claim is true.
Ask whether the choice would convince someone who doubted the claim. A mechanism does not settle whether the thing happened.
Prediction
Check the direction explicitly.
Predict first
A claim says a river's water quality has improved since 2015. Which finding CONTRADICTS it?
Correct: Pollutant concentrations measured in 2024 were higher than in 2015
Why: Higher pollutant levels mean worse water quality, which cuts directly against the claim of improvement. The second and third both support it, and the fourth describes an action that might cause improvement without showing it occurred. Direction is the check: does this make the claim more or less likely?
Concept
A relationship can be confirmed at either end, so evidence from the opposite extreme still supports it.
Ask whether the finding is CONSISTENT with the claimed relationship, not whether it repeats the claim's phrasing.
Concept
One choice usually cuts against the claim, and it is easy to select when reading quickly.
Under time pressure, on-topic can feel like supporting. Check the direction explicitly on every choice.
Prediction
The units decide the reading.
Predict first
A graph's vertical axis is labelled sales in thousands of units. A bar reaches 45. What does that represent?
Correct: 45,000 units
Why: The axis label states both the quantity, units sold, and the scale, thousands, so a bar at 45 means 45,000 units. Reading it as 45 ignores the scale, and the other two change the quantity entirely — one to a percentage and one to money. Reading the axis label before any choice makes all three errors visible.
Concept
A finding consistent with the claim is not the same as one that supports it strongly.
Specific measurements beat general impressions, and direct statements beat indirect ones.
Concept
When data fills a blank, the surrounding sentence tells you exactly what the data must show.
This variant is the most mechanical on the type: the sentence states the claim explicitly, so no interpretation is required.
Prediction
Explaining is not showing.
Predict first
A claim says a town's cycling rate has risen. Which is evidence FOR the claim rather than a possible cause of it?
Correct: Counts at twelve junctions recorded 25 per cent more cyclists than five years ago
Why: Only the first measures cycling itself. New lanes, fuel prices and a campaign are all plausible reasons why cycling MIGHT have risen, and none of them shows that it did. This is a favourite construction because explanatory choices feel highly relevant while addressing a different question entirely.
Two truths and a lie
Three of these statements about evidence questions are correct. The one left standing is false.
Eliminate the wrong options
Which statement is FALSE?
Survives elimination: c
Why: A possible cause is not evidence of an effect. If the claim is that a population declined, learning that its habitat shrank offers a mechanism but does not show the decline happened. Evidence must speak to the claimed fact itself. This distractor is effective precisely because an explanatory finding feels tightly connected to the claim while addressing a different question.
Check
State the claim before reading the choices.
Check your understanding
A text argues that a composer's use of silence became more deliberate over his career. Which quotation best supports that claim?
Answer: A
Why: The claim concerns his treatment of SILENCE and whether it became more deliberate. Only the first choice describes how rests were notated and contrasts early with late work, which is exactly the claimed change.
Choice C is the closest distractor because restraint feels connected to silence. Naming the claimed variable as the deliberate use of rests keeps them apart.
Section
Section 3
Worked example
A text argues that a poet's later work became markedly more concise. Which quotation best supports that claim?
Figure (svg): A claim with three candidate quotations measured against it
State the claim: her later poems are SHORTER or more compressed than her earlier ones.
Why: Concise is about length and compression, which fixes what evidence must measure.
Test each quotation against that sentence, not against the topic generally.
Why: All three quotations are about the poet's later work, so topic alone does not discriminate.
Only the line-count comparison measures length, so it is the evidence.
Why: A specific measurement of the claimed quantity is the strongest possible support.
Notice that the winning choice contains numbers. On this type, a measurement of the claimed quantity nearly always beats a description of it.
Verify: check the rejected choices are genuinely true: both are, and both concern different properties.
Why: Confirming they are true rather than false is what proves the discrimination was about relevance.
Answer: the quotation comparing average line counts
Worked example
A bar chart shows recycling rates for four cities. A student claims that City B recycles a greater share of waste than City D. Which choice uses the data to support this?
Figure (svg): A comparative claim with three data statements measured against it
Read the axes first: the vertical axis is percentage of waste recycled, and the categories are the four cities.
Why: Percentage rather than tonnage is exactly the distinction the third choice exploits.
State the claim: B's PERCENTAGE exceeds D's PERCENTAGE.
Why: The claim is comparative and about share, which fixes two requirements.
Only the first choice names both cities and compares their percentages.
Why: A comparative claim needs both terms of the comparison present.
Two separate traps appear here: a choice that omits half the comparison, and a choice that switches from share to total. Reading the axis label catches the second.
Verify: check the second choice: it is accurate about City B and silent about City D.
Why: Accuracy without the comparison cannot support a comparative claim, confirming the rejection.
Answer: the choice comparing 48 per cent with 31 per cent
Worked example
A researcher claims that a marsh's bird population has grown since restoration began. Which finding is evidence, rather than a possible cause?
Figure (svg): A claim about population growth with one measurement and two causes
State the claim: the NUMBER of birds is higher now than before restoration.
Why: The claimed fact is a population count, so evidence must count birds.
Ask of each choice: does this measure the population, or explain why it might change?
Why: This single question separates evidence from mechanism.
Only the count of birds measures the claimed quantity.
Why: Habitat area and nesting platforms are inputs; the bird count is the outcome.
The sceptic test is the cleanest version of this distinction: would this convince someone who doubted the claim? A mechanism would not.
Verify: check whether the rejected choices would convince a sceptic that numbers rose.
Why: They would not — a doubled water area is compatible with a population that fell, which confirms they are not evidence.
Answer: the choice reporting the rise in annual bird counts
Step zero
Before reading a single choice.
Discussion prompt
An evidence question presents a hypothesis and four findings. What is your first action, and why does it have to come before the choices?
Hint: It involves writing rather than reading.
Answer:
Write the claim in your own words, as one sentence.
Why before the choices: each choice is a plausible, true statement about the same topic, and reading them first quietly reshapes what you think the claim was.
Once your sentence is fixed, the choices are measured against it rather than the other way round.
Be specific about the variable. Not the study is about birds, but the claim is that nesting happens EARLIER in warmer springs.
On a data question, add one step: read the axis labels and units before the choices too, for the same reason — a choice can quietly attach a real number to the wrong quantity.
Worked example
A text claims that a policy reduced traffic congestion. Which choice CONTRADICTS the claim?
Figure (svg): A claim with a supporting choice, a contradicting choice and a neutral one
State the claim: congestion is LOWER after the policy than before.
Why: The direction of the claimed change is what evidence must confirm.
Check the direction of each choice: does it make congestion look better or worse?
Why: A choice can be tightly on topic and point the wrong way.
Rising journey times mean worse congestion, so that choice contradicts the claim.
Why: It is evidence against, which is the opposite of what a support question asks for.
Under time pressure, on-topic reads as relevant. Checking the direction of each choice takes two seconds and catches this entirely.
Verify: check the neutral choice: the date of introduction says nothing about congestion either way.
Why: Three distinct relationships appear here — support, contradiction and irrelevance — which is the full range.
Answer: the choice reporting that journey times rose
Worked example
A table shows average rainfall for four months. A sentence reads: rainfall was heaviest in the month of blank, when it reached blank millimetres. Which choice completes it?
Figure (svg): A sentence-completion goal with three candidate data fillings
Read the full sentence first: it asserts that rainfall was HEAVIEST in the named month.
Why: The sentence itself is the claim, so no interpretation is needed.
Find the highest value in the table and note both the month and the figure.
Why: The claim requires a maximum, and both blanks must come from the same row.
Only the choice pairing October with 142 satisfies both blanks correctly.
Why: The other two either name the wrong month or mismatch the month with its value.
The mismatch distractor — right month, wrong number — is specific to this variant and is caught by checking both blanks against the same row.
Verify: check that 142 really is the largest of the four values.
Why: Confirming the maximum is what validates the word heaviest in the sentence.
Answer: October, when it reached 142 millimetres
Faded example
From memory. Four lines that carry the type.
Fill in the blanks
Write the claim in your own words before reading any choice. Every wrong choice will be true but about a different claim. On a data question, read the axis labels before the choices. And a finding that would EXPLAIN the claim is not evidence that the claim is true.
Why: The second blank is the one that changes behaviour. Once you accept that every choice is true, you stop checking accuracy and start checking fit — which is the only thing that can actually distinguish them.
Worked example
A claim says a training programme improved reaction times. Two choices are both relevant. Which supports it more strongly?
Figure (svg): A claim with a measured comparison, a self-report and a neutral fact
State the claim: reaction times are FASTER after training.
Why: The claimed effect is a measurable quantity.
Both the measurement and the self-report are relevant to reaction speed.
Why: Neither is off topic, so relevance alone does not decide.
The measured comparison is far stronger, because it quantifies the effect and includes a control group.
Why: The question asks which MOST strongly supports, which is a comparison of strength.
When two choices are both relevant, prefer the one that is specific, measured and comparative. Those three properties nearly always identify the intended answer.
Verify: apply the sceptic test to each: a doubter would accept the measurement and question the self-report.
Why: Feeling quicker is compatible with no actual improvement, so it is weak evidence.
Answer: the choice reporting 190 against 240 milliseconds
Fill the middle
Wrong answers fail in recognisable ways.
Fill in the blanks
A choice can fail by being about a different variable, by pointing in the opposite direction, by offering a cause rather than an outcome, or by being relevant but too weak.
Why: Four failure modes, and every distractor on this type is one of them. Naming which one a choice commits is faster than arguing with it, and it protects against being persuaded by a statement that is perfectly true.
Estimation
Most choices can be dismissed on the variable alone.
Predict first
A claim concerns the SPEED of a chemical reaction. Which choice could possibly be relevant?
Correct: The reaction completed in 12 seconds at 40 degrees and 31 seconds at 20 degrees
Why: Only one choice mentions time at all, and speed is a claim about time. Colour, cost and history are properties of the reaction that cannot bear on how fast it runs. Scanning for the claimed VARIABLE — here, time — eliminates three choices in a few seconds without reading any of them carefully.
Check
Read the axis before the choices.
| City | Recycling rate | Total waste (tonnes) |
|---|---|---|
| Alton | 48% | 120,000 |
| Bexley | 31% | 310,000 |
| Carden | 52% | 95,000 |
| Denmoor | 27% | 410,000 |
Check your understanding
A student claims that Carden recycles a greater SHARE of its waste than any other city listed. Which choice best supports this?
Answer: A
Why: The claim is about SHARE, which is the recycling rate column. Carden's 52 per cent is higher than 48, 31 and 27, so the claim holds and this choice states exactly that comparison.
Three of the four choices are accurate readings of the table. Only one is about the column the claim actually concerns, which is why reading the headings first matters.
Section
Section 4
Trap
The trap. A claim concerns a species' nesting TIMING, and a choice reports that the species has expanded its range northward.
The choice is true, drawn correctly from the text, and clearly about the same species and the same warming.
It supports a different claim — about range — and says nothing about when the birds nest.
The fix. Name the claimed VARIABLE explicitly in your one-sentence restatement.
Then check each choice mentions that variable at all before considering anything else.
Trap
The trap. A claim says congestion fell, and you select a choice reporting that journey times rose.
The choice is squarely on topic and quantitative, which under time pressure reads as relevant.
But it cuts against the claim, which is the opposite of what a support question wants.
The fix. Check the direction of every choice: does it make the claim more likely or less likely?
This is a two-second question and it separates support from contradiction completely.
Error analysis
A student's reasoning on an evidence question. Everything asserted is true.
Annotate
On: \( \text{Claim: nesting is earlier in warmer springs.} \;\Rightarrow\; \text{Chose: the range has expanded north.} \)
The error is not one of comprehension. The student understood the text correctly and measured the choice against a blurred version of the claim, which is exactly what reading the choices first produces.
Trap
The trap. The claim is that a population grew, and a choice reports that its habitat doubled in area.
A larger habitat is a plausible reason a population MIGHT grow, so the choice feels highly relevant.
It does not show that the population actually did grow, which is what the claim asserts.
The fix. Apply the sceptic test: would this convince someone who doubted the claim?
A mechanism would not, because the mechanism is compatible with the claim being false.
Elimination
A claim states that a new bus route reduced average commuting TIME for residents of a district.
Eliminate the wrong options
Which finding supports it? Three fail on the variable alone.
Survives elimination: a
Why: The claim names a specific variable — commuting time — so only a finding measuring time can support it. Ridership, satisfaction and cost are all genuine outcomes a transport authority would care about, and each supports a different claim. Scanning for the claimed variable eliminated three choices before any of them was read carefully.
Trap
The trap. A choice reports that City D recycled more, quoting a real number from the chart — but the number is total tonnes while the claim was about percentage share.
Both quantities appear in the material, and the number itself is correct.
Without having read the axis label, there is nothing to catch the switch.
The fix. Read the axis labels, units and legend before evaluating any choice.
Then check that each choice's number belongs to the quantity the claim is about.
Counterexample
A rule students apply too broadly.
Discussion prompt
A student says: if a choice is true and mentions the topic, it supports the claim. Give a counterexample.
Hint: Think of a true statement that points the wrong way.
Answer:
Counterexample: the claim is that a river's water quality improved, and the choice reports that pollutant concentrations rose.
That statement is true and squarely on topic — same river, same measure, same period — and it CONTRADICTS the claim.
A second counterexample in a different direction: the claim is that a bird population grew, and the choice reports that its habitat doubled.
True, on topic, and not evidence — a mechanism is compatible with the population having fallen anyway.
The corrected rule: a choice supports a claim only if it makes that specific claim more likely. Truth and topicality are necessary and nowhere near sufficient.
Edge cases
Evidence must address the claim. How closely?
Discussion prompt
When does indirect evidence still count, and how do you choose between two choices that both bear on the claim?
Hint: The question usually asks which supports it MOST strongly.
Answer:
Indirect evidence can count when it bears on the claimed relationship even if it does not restate it. Evidence from the opposite extreme is the clearest case: a cold spring bringing later nesting supports a warmth-and-timing claim.
Two relevant choices are separated by strength, and three properties decide it: specific beats vague, measured beats reported, and comparative beats absolute.
The sceptic test is the general tool: which choice would be hardest to explain away if the claim were false?
A self-report is genuinely weak because feeling faster is compatible with no change in actual speed.
And a single anecdote is weaker than an aggregate, because one case can be exceptional while an average cannot.
The one thing that never counts is a choice about a different variable, however strong it is on its own terms.
Check
Apply the sceptic test.
Check your understanding
A conservation group claims that otter numbers along a river have recovered since 2010. Which finding is the strongest evidence?
Answer: A
Why: The claim is about NUMBERS recovering, so a count at two points in time measures exactly the claimed quantity and shows the increase directly.
Choice D is the instructive one. It is real evidence of something, and far weaker than a count — which is why the question asks for the STRONGEST support rather than any support.
Section
Section 5
Matching
The claim: a school's new reading programme raised test scores.
Match the pairs
Why: Five distinct failure modes appear here, and they are the complete set: wrong variable, wrong direction, cause rather than outcome, too weak, and simply neutral. Naming which one a choice commits is faster and more reliable than arguing with the choice on its own terms.
Sorting
The claim: a city's air quality has improved over the past decade.
Sort into buckets
What does each finding do?
Two supports, two causes, one contradiction and one neutral. That distribution is close to what a real question offers, which is why naming the category is faster than weighing each choice on its merits.
Comparison
Fill the blanks from memory.
Comparison matrix
| textual evidence | quantitative evidence | |
|---|---|---|
| Where evidence comes from | a quotation from the passage | a table, chart or graph |
| The extra first step | none beyond stating the claim | read the axis labels and units |
| The signature distractor | a true quotation about a different property | a real number from the wrong series or unit |
| What decides it | whether the quotation asserts the claim | whether the number measures the claimed quantity |
The two columns differ only in the second row. Everything else is the same discipline: state the claim, then measure each candidate against it.
Trade off
Fill in what each property buys.
Comparison matrix
| property of a choice | makes it stronger because | the weaker alternative |
|---|---|---|
| It contains a measurement | a number is hard to explain away | a description or an impression |
| It is comparative | it shows a difference rather than a level | a single figure with nothing to compare it to |
| It reports an outcome | it addresses the claimed fact directly | a cause, which only explains |
| It aggregates many cases | one exceptional case cannot drive it | a single anecdote |
When two choices are both relevant, these four properties nearly always identify the intended answer without any further reasoning about the passage.
Real world
One minute on why this is the most transferable question type on the SAT.
Discussion prompt
Evaluating whether evidence actually supports a claim is a daily skill. Where does the true-but-irrelevant pattern appear in real life?
Answer:
In advertising, constantly. A product claim about effectiveness is supported by evidence about popularity, or customer satisfaction, or how long the company has traded — all true, none bearing on whether it works.
In political argument, where a claim about a policy's effect is answered with evidence about its cost, its popularity, or its author's motives.
In news reporting, where a headline claims one thing and the study underneath measured something adjacent — a common pattern with health research.
The skill is identical to the SAT's: state precisely what is being claimed, then ask whether the evidence offered addresses that specific thing.
And the failure mode is identical too: being persuaded by a statement that is entirely true and answers a different question. That is why this is the most common question type in the section.
Ranking
The claim: a tutoring programme improved mathematics scores. Order from weakest support to strongest.
Put in order
Why: The duration is neutral and supports nothing. Self-reported confidence is relevant but weak, since feeling more confident is compatible with unchanged scores. A single student's improvement is a real measurement and could be exceptional. The group comparison is strongest: it is measured, aggregated and comparative, so it is hardest to explain away if the claim were false. Those three properties are what the ranking is tracking.
Concept
This type is 15.3 per cent of the section — about 8 questions — so it deserves the largest single block of your Reading and Writing preparation.
| session | what you do | why |
|---|---|---|
| 1 | Twenty textual questions, writing the claim in your own words before reading any choice. | The habit that fixes the whole type, practised in isolation. |
| 2 | Fifteen quantitative questions, reading axes and units aloud before the choices. | The extra step the data version needs, and the source of its signature distractor. |
| 3 | For fifteen questions, name the failure mode of each wrong choice. | Categorising is faster than arguing, and it builds recognition of the four patterns. |
| 4 | Mixed set under time, applying the sceptic test to the final two choices. | Trains the strength comparison that decides when two choices are both relevant. |
Naruhodo Tutoring — SAT Reading and Writing question bank export (sat-english-question-index.json) — filter the bank to this skill tag — 258 questions, roughly a third of each difficulty
Explain it to yourself
Close the deck.
Discussion prompt
Without looking, name the four ways a wrong answer fails on this type, with an example of each.
Hint: All four are true statements.
Answer:
Wrong variable: the claim is about yield and the choice is about cost. True, and about something else.
Wrong direction: the choice contradicts the claim rather than supporting it, while being squarely on topic.
A cause rather than an outcome: the choice explains why the claim might hold without showing that it does.
Too weak: the choice is relevant but vague, self-reported or anecdotal, where another choice is measured and comparative.
If you noted that all four are TRUE statements, you have the observation that changes how you read the choices.
Explain it
Two minutes, out loud.
Discussion prompt
A friend reads all four choices carefully, checks each against the passage, and still gets these wrong about half the time. What is going wrong?
Answer:
Diagnose it: they are checking accuracy, and every choice is accurate. Careful reading verifies truth, and truth is not what separates these choices.
Show them with a made-up example. Claim: a species nests earlier in warm springs. Choice: the species has expanded northward. Ask whether it is true — yes. Ask whether it shows earlier nesting — no.
Name what the question is really asking: not is this true, but does this prove THAT.
Give the physical habit: write the claim as one sentence, naming the specific variable, before uncovering the choices.
And the closing test: for each choice ask whether it would convince someone who doubted the claim. A true but off-target statement would not.
Commit first
Commit before you check.
Predict first
A claim states that a bakery's new recipe made its bread stay fresh longer. Which supports it?
Correct: Loaves made with the new recipe remained soft for four days, against two days previously
Why: The claim is specifically about how long the bread stays FRESH, so only a measurement of freshness over time can support it. Taste, ingredient count and sales are three different outcomes — and sales is the most tempting, because rising sales feel like proof the recipe is better. Better and longer-lasting are different claims, and only one of them was made.
Connect it up
Blank paper.
Draw it
Draw a ruler across the top of the page and label it THE CLAIM, with a note underneath: write it in your own words, naming the specific variable, before reading any choice. Below it draw four boxes for the four choices, and label each with one of the failure modes: wrong variable, wrong direction, a cause rather than an outcome, and too weak. Beside each write a one-line example. On the right, add a branch for the quantitative version with its extra first step: read the axis labels, units and legend. At the bottom write the sceptic test in large letters: would this convince someone who doubted the claim?
Exit ticket
One question before you close the deck.
Predict first
You are choosing between two evidence choices and both are clearly relevant to the claim. What decides it?
Correct: Which would be harder to explain away if the claim were false
Why: The question asks which supports the claim MOST strongly, which is a comparison of evidential strength. The sceptic test captures it: measured beats reported, comparative beats absolute, and aggregate beats anecdote. Position in the passage is irrelevant, length is not strength, and matching the passage's wording is exactly how a distractor is made to feel right without being relevant.
Recap
One type, one discipline: state the claim, then measure every choice against it.
| never do this | do this instead |
|---|---|
| Check whether a choice is true | Check whether it proves THIS claim |
| Read the choices before fixing the claim | Write the claim first, then uncover them |
| Accept a plausible cause as evidence | Ask whether it measures the claimed outcome |
| Take a number from a graph without reading the axis | Read the labels, units and legend first |
| Assume on-topic means supporting | Check the direction explicitly |
| Prefer the longest or most detailed choice | Prefer the one hardest to explain away |
Naruhodo Tutoring — SAT Reading and Writing question bank export (sat-english-question-index.json) — and the site's SAT pages to drill this type in isolation, then mixed
Want this taught 1-on-1? Alexander tutors SAT Prep — $55/session, free consultation.