Textual Evidence for the Exact Claim

CHAPTER SIX · INFORMATION AND IDEAS

If a friend tells you a new restaurant is better than the old one, and then tells you the new place has valet parking, you would not consider the question settled. Valet parking might be true, and it might even matter to some other question, but it does nothing to prove one restaurant is better than another. Three separate diners returning the same dish bears more directly on food quality than valet parking does, though it still would not settle the broad claim that one restaurant is better overall.

College Board calls this skill Command of Evidence, and this chapter covers the textual form of it: a passage gives you a claim, an interpretation, or a conclusion, and asks which piece of evidence bears on it most directly. The specified point usually sits in the question’s own stem, and sometimes sits inside the passage itself. The correct choice has to do more than discuss the same subject. It has to support, weaken, illustrate, or otherwise address the exact point named, not just the general topic around it. A companion chapter covers the same skill when the evidence arrives as a table or graph instead of a sentence.

Identify the Exact Claim

The stem usually hands you a specific claim or point to work with. Summarizing the passage is not the task here, and choosing the evidence that bears most directly on that one target is. Fix the target claim before evaluating the choices.

Before comparing choices, identify the claim’s owner, assertion, scope, and the requested evidence direction: whether you are being asked to strengthen it, weaken it, or simply identify what bears on it.

Asking which finding bears most directly on the claim. Ask whether a candidate finding would be more expected if the claim were correct than if a relevant alternative were correct instead. The larger that difference, the more strongly the finding discriminates between the two possibilities. Not every support question involves an explicit rival explanation, and when one is genuinely in play, compare the predictions the two accounts actually make.

Treating if true as an instruction to accept it. When a stem asks which finding, if true, would support or weaken a claim, grant each proposed finding as true for purposes of evaluating its effect on the claim. You are not being asked whether the finding is plausible, or whether such a study could realistically be run. Grant the finding exactly as given and ask what it does to the claim once granted. A student who starts doubting the finding itself has begun answering a question the exam never asked.

Match the Evidence to the Claim Type

A claim makes a specific demand on evidence, and the shape of the claim tells you exactly what will satisfy that demand. Support raises the plausibility of a claim rather than proving it outright. The task is to identify which choice bears most directly on the claim, not which would settle the issue beyond doubt.

A claim’s shape determines exactly what evidence can satisfy it.

Repeating the claim is not the same as supplying evidence for it. One choice on many of these questions simply repeats the claim in fresh words. It reads as agreement, and agreement is not evidence. Ask what the choice actually contributes that the claim did not already assert on its own: a new observation, a comparison, a measurement, or a quotation. A choice can restate part of a claim and still earn its place, if it adds an observation that bears on a different part of that same claim, so look for the addition rather than stopping the moment you notice an echo.

Two choices can both support the claim, and one can still support it more directly. These questions rarely divide neatly into one supporting choice and three irrelevant ones. When several choices genuinely support the claim, compare directness, relevance, scope, and, whenever rival explanations are actually in play, how well each piece of evidence distinguishes between them.

Association and causation demand different levels of trust. A finding that two things simply occur together supports a causal claim only weakly, because a third factor, or the relationship running in the reverse direction, can produce the exact same pattern. A comparison in which the proposed cause was deliberately assigned or manipulated supports a causal claim far more strongly. Neither kind of evidence is disqualified outright; state what the evidence actually does in words the case allows: it supports, it is consistent with, it weakens, or it does not establish.

The right outcome can arrive by the wrong route. A finding can produce the exact outcome named in a claim and still support the wrong explanation for why that outcome occurred. If a claim says X works because of mechanism A, evidence that X works through mechanism B supports the result itself while leaving mechanism A untested, or even making it less likely to be the real explanation. On any question concerning mechanism, separate the outcome from the route: ask not only whether the effect happened, but whether it happened for the specific reason the claim actually names.

Match Weakening Evidence to the Claim

A weaken question asks for the finding that makes a claim harder to hold, and exactly what counts as weakening depends entirely on what the claim itself asserts.

If the claim says X is necessary for Y, evidence weakens it by showing a clear case of Y occurring without X, since that single case makes the stated requirement fail. If the claim says X is sufficient for Y, evidence weakens it by showing a clear case of X occurring without Y, since that case makes the stated guarantee fail. If the claim says X contributes to or increases Y, evidence weakens it by showing that the association disappears under a better comparison, that a rival explanation fits the pattern as well or better, or that the proposed mechanism fails to operate where it should; an ordinary causal claim of this kind permits Y to occur without X and X to occur without Y sometimes, so neither of those facts alone actually touches it. If the claim says something is universally true, one valid counterexample weakens it completely, since all, always, and none admit no exceptions whatsoever.

Relevance decides everything on these questions. A finding about cost or unpopularity leaves an empirical claim about what actually happened exactly where it stood, and that same finding bears directly on a separate claim about whether a program can be adopted or sustained going forward. Read exactly what the claim asserts, then ask whether the candidate choice actually touches that specific assertion. Necessity and sufficiency arguments only apply when the claim itself genuinely asserts one of them; most claims assert neither.

Match Quotations to the Whole Claim

The literary form of this question describes a character or a work and asks which quotation illustrates that description. Two checks resolve many of these literary quotation questions.

Match the form of the quotation to the form of the claim. If the claim concerns behavior, choose a quotation that shows the behavior in action. If the claim concerns what the narrator or another character explicitly states, a direct statement may be exactly the evidence the claim calls for; direct narration is not automatically weaker than a scene showing behavior, and neither form beats the other outside the context of what a specific claim actually needs.

Match the source and the moment to the claim as well. A claim about how a character behaves in company is not carried by a line about her private thoughts alone in a room. Another character’s observation can serve when the claim concerns a trait the observation actually demonstrates, and it cannot serve when the claim concerns something only the narrator maintains.

The claim. In the novel, Adeline covers her nervousness with an excess of formality.

A quotation that only states it. “Adeline had never been comfortable in large rooms.”

A quotation that actually shows it. “‘I trust the journey was not disagreeable,’ she said, and set the cup down twice before it stayed.”

The second version carries both halves of the claim at once. Her spoken sentence is stiff and ceremonious, which is the formality; the hand that cannot place a cup on the first try supplies the nervousness underneath it. A quotation illustrating only the formality half would leave the other half of the claim completely unsupported.

Worked Examples by Claim Type

Support the Proposed Mechanism

Leafcutter ants carry cut leaves along cleared trails to the nest, and a much smaller worker often rides on the fragment. A researcher proposes that these riders guard the carrier against parasitic flies, which lay eggs on the heads of ants whose jaws are full.

Which finding, if true, would most directly support this proposal?

A. Carriers travel more slowly when a rider is present

B. Carriers without riders are struck by flies several times more often than carriers with them

C. Riders are on average smaller than the ants that carry the leaves

D. Colonies with more foragers expand their nests faster

The researcher proposes that riders reduce fly attack on carriers, so the supporting finding has to show a lower attack rate specifically when riders are present, which makes B correct. A reports a cost of carrying a rider that says nothing about flies at all. C only restates a fact the passage already reports about size; a restatement adds no new evidence. D concerns colony growth, a different question entirely from the ants actually under attack.

Use a Finding That Challenges the Conclusion

A city built protected bicycle lanes along six corridors, and cyclist injuries on those corridors fell by 40 percent over the following two years. Planners concluded that the lanes had made the corridors safer.

Which finding, if true, would most directly weaken the planners’ conclusion?

A. Cyclist injuries fell by a similar amount on comparable corridors that received no lanes

B. The lanes cost substantially more than the city had budgeted

C. The number of cyclists using the six corridors rose during the same period

D. Some drivers reported confusion at the new intersections

A weakens the inference that the lanes themselves caused the reduction, because the same outcome occurred on comparable corridors that received no intervention at all, supplying the comparison the original conclusion lacked. B bears on whether the lanes were worth building and leaves their actual safety effect untouched. C does not weaken the conclusion; more riders alongside fewer total injuries is compatible with improved safety, although the passage does not provide exposure-adjusted injury rates. D reports confusion, not injury, and a handful of driver reports is thin evidence against a measured 40 percent change.

Choose a Quotation That Shows the Claim

In Marisol Vega’s novel The Long Terrace, the narrator claims that her uncle’s generosity was a way of keeping people obliged to him.

Which quotation from The Long Terrace most effectively illustrates the claim?

A. “My uncle gave more freely than any man in the district, and everyone said so.”

B. “He paid for my cousin’s schooling and mentioned it whenever she disagreed with him.”

C. “He kept his accounts in a green ledger that he never let anyone open.”

D. “My aunt believed he was the kindest of her brothers.”

B shows both halves of the claim: a gift, and then that same gift produced again whenever she steps out of line, which is obligation demonstrated rather than named. A shows the generosity and supplies no evidence that it was ever used to create leverage over anyone, which is the whole of the claim. C shows secrecy about money, a different trait entirely. D shows kindness, which does not by itself show generosity being used to create obligation.

Support Both Sides of a Comparison

Bakers have long held that loaves leavened with a sourdough culture stale more slowly than loaves raised with commercial yeast. Bakery laboratories measure staling as the firmness of the crumb, so a firmer loaf is a staler one.

Which finding, if true, would most strongly support that claim?

A. Sourdough loaves stayed soft for five days in a test kitchen

B. Sourdough loaves measured firmer on day five than on the day they were baked

C. Loaves of both kinds measured equally firm on the day they were baked, and by day three the yeast loaves had gained twice as much firmness as the sourdough loaves

D. Bakers surveyed in four countries reported that sourdough keeps well

C supplies a shared starting point and a measured change over the same period for both kinds of loaf, which is exactly what a claim about a slower rate requires. A reports one side of a two-sided claim, with nothing to compare it against. B is also one-sided; it shows sourdough staling at all, and says nothing about whether it stales more slowly than the alternative. D reports impressions from the very people who already hold the belief, not a measurement.

Find the Passage’s Own Evidence

Mapmakers in the eighteenth century copied one another freely, and a single error could travel through a dozen atlases before anyone checked it. One range, first drawn across a stretch of West Africa in the 1790s, appeared on dozens of published maps over the following century. Surveys found no such mountains, and cartographers dropped the range only in the 1880s.

Which detail from the text most directly supports the point that a single error could persist through many atlases?

A. Mapmakers in the eighteenth century copied one another freely

B. A range first drawn in the 1790s appeared on dozens of maps and was dropped only in the 1880s

C. Surveys found no such mountains

D. The range was drawn across a stretch of West Africa

The claim is already in the passage; the question asks which detail most directly supports it. B gives both the number of maps and the span of decades together, which is the persistence itself, stated directly. A explains how an error could spread in general, and would be equally true even if no particular error had ever actually persisted. C establishes that the range was an error and says nothing about how long that error lasted. D gives a location with no bearing on persistence at all.

Support the Claim Directly

A finding can be entirely true and squarely about the same subject as a claim, and still fail to show that the claim itself is correct.

Claim: a particular roof coating reduces heat loss in buildings.

Which finding, if true, would most directly support the claim?

A. Coated roofs are common on buildings in cold regions

B. Buildings with the coating lose heat more slowly than matched buildings without it

C. The coating is more expensive than uncoated roofing material

D. Roofing contractors recommend the coating for older buildings

B directly tests the claimed effect: a comparison showing slower heat loss where the coating is present. A is relevant to the topic and shows only where the coating is popular, not that it changes anything. C bears on cost, a different question from performance. D reports a recommendation, not a measured effect.

Support Every Part of a Two-Part Claim

Claim: a bird species moved its range north because winters in its historical range became milder, which reduced the availability of its preferred cold-adapted prey.

Which finding, if true, would most directly support the claim?

A. The species has been recorded farther north than in any prior decade

B. Winters in the historical range grew measurably milder over the same period, and populations of the species’ cold-adapted prey declined there during those years

C. The species’ northern range now overlaps with a related species

D. Birdwatchers reported more northern sightings on citizen-science platforms

B is the only choice that supports both predicted components of the proposed mechanism at once: winters growing milder, and the cold-adapted prey declining over that same period. It does not by itself prove that the milder winters caused the prey decline or that the prey decline caused the range shift; it shows the two trends the mechanism predicts actually co-occurring, which is stronger support than either trend alone but still short of establishing each causal link directly. A supports only that the species moved, which the claim already states as a premise; it says nothing about why. C and D both concern the movement itself again, not the proposed reason behind it.

Use Evidence That Distinguishes Rival Explanations

Theory A holds that birds leave a coastal roost before a storm because they sense the drop in air pressure directly. Theory B holds that they leave because falling temperatures disrupt their feeding.

Which finding, if true, would most strongly support Theory A over Theory B?

A. Birds left the roost several hours before the storm arrived

B. Birds left the roost during a sharp pressure drop that occurred while temperatures stayed constant

C. The storm caused significant damage to the roost itself

D. Birds returned to the roost within a day of the storm passing

Both theories predict that birds leave before a storm, so the departure itself, choice A, cannot separate them. B isolates exactly the variable that distinguishes the two theories: pressure fell while temperature held steady, and the birds still left, which is far more expected under Theory A than Theory B. C concerns damage to the roost, not the reason birds left it. D concerns their return, a question neither theory addresses.

Separate Weakening a Rival From Supporting the Main Explanation

A historian argues that a shift in trade routes, not a change in climate, explains why a coastal settlement was abandoned. Climate records show no significant change in the region during the relevant period.

Does the climate record support the historian’s trade-route explanation?

A. Yes, because it proves the settlement was abandoned for economic reasons

B. It weakens the climate explanation without directly showing that trade routes were the cause

C. No, because it shows the settlement was never abandoned at all

D. Yes, because trade records and climate records always move together

B states the relationship precisely: weakening a named rival can increase the relative plausibility of the trade-route explanation within the alternatives under discussion, but the climate record is not direct evidence that trade routes caused abandonment. A overclaims certainty the weakened rival alone cannot supply. C contradicts the passage’s own premise. D invents a general rule about trade and climate that nothing here supports.

Check for Confounding in Causal Comparisons

Farms using a new fertilizer produced larger tomatoes than farms that did not. The farms using the new fertilizer also received significantly more irrigation than the comparison farms.

Which finding, if true, would most directly weaken the claim that the fertilizer caused the larger tomatoes?

A. The fertilizer is more expensive than older alternatives

B. Farms that received the same extra irrigation without the new fertilizer produced tomatoes just as large

C. Farms using the new fertilizer are located in a warmer region

D. Tomato size varies naturally from plant to plant

B directly isolates the confound: if irrigation alone produces the same result, the fertilizer’s own contribution is no longer established by this comparison. A concerns cost, not size. C introduces a new, unexamined difference rather than testing the one already identified. D is true of tomatoes in general and touches nothing about this specific comparison.

Measure Both Sides of the Comparison

Claim: Method A removes rust more effectively than Method B.

Which finding, if true, would most strongly support the claim?

A. Surfaces treated with Method A showed reduced rust within a week

B. Surfaces treated with Method A showed significantly less rust than a matched set treated with Method B

C. Method A has a lower material cost than Method B

D. Method A is used more often than Method B

B is the only choice that actually compares the two methods against each other. A reports a result for Method A alone, with nothing to measure it against; a strong-sounding figure for one side does not by itself establish “more effective than.” C concerns cost, a different question from effectiveness. D reports usage habits, not outcomes.

Keep the Evidence Direction Straight

Claim: an increase in a region’s average income (X) leads to an increase in local small-business formation (Y).

Which finding, if true, would most directly support the claim?

A. Regions with high rates of new small-business formation later saw rising average incomes

B. Regions where average income rose saw new small-business formation rise in the years that followed

C. Regions with many small businesses tend to have diverse local economies

D. Small-business owners report high job satisfaction

B preserves the claimed direction: income first, business formation following it. A involves the same two variables and reverses which one comes first, which does not support a claim about income leading to business formation. This supports the proposed direction more than the alternatives do, but the order alone does not by itself establish that the income increase actually caused the business formation that followed it; a third factor could still drive both. C and D both concern small businesses without addressing the income relationship named in the claim at all.

Use Timing to Test a Causal Claim

A city credits a new streetlight program, installed in 2018, with a decline in nighttime crime in the district. Crime data show the decline actually began in 2014 and continued at a steady rate afterward.

Which finding, if true, would most directly weaken the city’s conclusion?

A. The decline in crime began in 2014, four years before the streetlight program started

B. The streetlights cost more to install than originally budgeted

C. Some residents did not notice the new streetlights right away

D. Crime rates vary from year to year in most cities

A rules out the 2018 program as the cause of the decline’s 2014 onset, though the program could still have affected the trend after installation; a cause cannot explain a trend that began before it existed. B concerns cost, not crime. C reports awareness, not outcomes. D is a generic fact about crime data that touches nothing specific to this city’s trend.

Match Evidence to the Named Group

Claim: a new teaching method improves test scores specifically among students who enter the year below grade level.

Which finding, if true, would most directly support the claim?

A. Test scores across the entire school rose after the method was introduced

B. Among students who entered the year below grade level, scores rose more than among their peers who entered at or above grade level

C. Teachers reported the method was easy to implement

D. The method has been adopted in schools across the district

B directly addresses the named subgroup and shows a larger gain there than among peers. It supports subgroup-specific benefit more directly than the schoolwide result, though a causal claim would still depend on the study design. A reports a broad, schoolwide result that may be true while telling you little about the below-grade-level students specifically; a wider population is not a substitute for the one named in the claim. C concerns ease of use, not outcomes. D concerns adoption, not effectiveness.

Match Evidence to the Claim’s Reach

Claim: a new irrigation method works effectively across a wide range of climates.

Which finding, if true, would most directly support the claim?

A. The method worked well in a single, unusually humid test region

B. The method produced strong results across trial sites in five climate zones ranging from arid to temperate

C. Farmers in one region reported the method was easy to install

D. The method uses less water than older irrigation systems

B matches the claimed reach: multiple, genuinely different climate zones. A supports only one case, and a claim about a wide range needs more than a single favorable environment. C concerns ease of installation, not performance across climates. D concerns water use, a different claim entirely.

Test Universal Claims With Counterexamples

Claim 1: this material never conducts electricity. Claim 2: this material usually does not conduct electricity.

Which single finding would weaken Claim 1 but leave Claim 2 largely intact?

A. A controlled sample of the material conducted a small electrical current under high humidity

B. Most samples of the material tested did not conduct electricity

C. The material is used in most modern insulation

D. The material was first identified over a century ago

A provides one genuine exception, enough to defeat never while remaining compatible with usually, which already allows for occasional cases like it. B supports both claims about equally and defeats neither. C and D touch neither claim’s actual assertion.

Interpret a Null Finding Carefully

A twelve-week trial found no statistically significant difference in soil moisture retention between plots treated with a new mulch product and plots that received a standard mulch.

Based on this finding, which conclusion is most directly supported?

A. The new mulch has no effect on soil moisture retention

B. This particular trial did not detect an effect of the new mulch on soil moisture retention

C. The new mulch is unsafe for long-term use

D. Standard mulch is more effective than any alternative

B stays exactly at the level the trial actually supports: no effect was detected in this trial, which is a fact about the trial’s result, not a settled fact about the mulch itself. A overclaims a conclusion the trial cannot deliver on its own; a trial can fail to detect a real effect because of its size, duration, or measurement limits. C and D both introduce claims this trial never addresses at all.

Check What the Study Measured

A survey asked shoppers how much they spent at a store last month and compared their answers to store receipts.

Which conclusion is most directly supported by this study design?

A. Shoppers accurately estimate their own emotional attachment to brands

B. Shoppers’ reported spending can be compared against their actual recorded spending

C. Shoppers’ satisfaction with the store predicts their future spending

D. Shoppers who spend more are more loyal to the store

B matches the study’s measured comparison: reported spending versus recorded spending. A introduces “emotional attachment,” a concept this study never measured at all. C introduces satisfaction and future spending, neither of which appears in this design. D introduces loyalty, another concept absent from what was actually asked or recorded.

Separate Post-Intervention Change From Causal Evidence

A company introduced a new onboarding program in January. By June, new-employee retention had risen by fifteen percent compared to the previous year.

Which finding, if true, would most directly strengthen the claim that the onboarding program caused the improved retention?

A. Retention rose by fifteen percent after the program began

B. A comparable department that did not adopt the new program saw no similar rise in retention over the same months

C. Employees who went through the new program reported enjoying it

D. The program took two weeks to complete

B supplies the comparison the timing alone cannot: a similar group without the program, tracked over the same period, that did not see the same change. A restates the timing itself, which is consistent with an effect and does not by itself establish one; a rise following a program is not automatically a rise because of it. C reports enjoyment, not retention. D reports duration, a detail about the program’s length rather than its effect.

Support Purpose With Evidence of the Stated Aim

Claim: a city redesigned its bus routes specifically in order to increase ridership among elderly residents.

Which finding, if true, would most strongly support the claim?

A. Ridership among elderly residents rose after the redesign

B. City planning documents from before the redesign state that increasing elderly ridership was the project’s primary goal

C. The new routes pass more frequently near senior centers than the old routes did

D. Elderly residents reported liking the new routes

A direct statement of intent, or context clearly fitting that intent better than the alternatives, is what a purpose claim actually needs. B is exactly that: a stated aim, on the record, before the outcome existed to be observed. A reports only the outcome, which does not by itself reveal why the redesign was undertaken; the increase could have been an unintended side effect of a redesign aimed at something else entirely. C is suggestive context and weaker than a direct statement, though it could support the claim in the absence of one. D reports a reaction, not a motive behind the original design.

Match the Quotation to the Claimed Trait

Claim: Mira is reluctant to leave her childhood home.

Which quotation most effectively illustrates the claim?

A. “Mira packed her suitcase the night before, folding each shirt twice.”

B. “Mira stood at the door for a long moment before finally turning the handle.”

C. “Mira had lived in the house for over twenty years.”

D. “Mira told her brother the house needed new paint.”

B shows hesitation directly: standing at the door, delaying the moment of leaving. A concerns Mira and leaving, and packing carefully is not itself reluctance; it could just as easily show care or organization. C establishes a fact about duration with no bearing on her feelings about leaving. D reports a practical comment about paint, unrelated to reluctance.

Support Every Part of a Compound Literary Claim

Claim: the abandoned house is both cramped and nearly silent.

Which quotation most effectively illustrates the claim?

A. “Not even the floorboards creaked as she turned sideways to pass between the two front rooms.”

B. “She had to turn sideways to pass between the two front rooms, though a radio still played faintly somewhere upstairs.”

C. “Dust had settled thick over every surface she touched.”

D. “The paint on the walls had long since faded to gray.”

Two qualities are claimed together, and the strongest quotation carries both at once. A does exactly that: turning sideways between rooms shows cramped space, and floorboards that do not even creak while she moves shows near silence, in the same single image. B shows the cramped space and directly contradicts the silence half with a radio playing; a quotation that undercuts half the claim is worse than one that simply omits it. C and D both concern neglect and appearance, neither cramped space nor silence specifically.

Match the Quotation to the Exact Claim

Claim: over the course of the novel, the character grows more confident in public settings.

Which quotation most effectively illustrates the claim?

A. “By the final chapter, she addressed the room without a single pause.”

B. “In the opening chapter, she stammered through her introduction and sat down early; by the final chapter, she addressed the room without a single pause.”

C. “She had always been confident among her closest friends.”

D. “Her father had been known for his commanding presence at gatherings.”

The claim asserts a change, which requires a contrast, not just an endpoint. B supplies both the earlier stammering and the later composure, showing the actual growth. A shows only the later moment and, without the earlier contrast, does not by itself demonstrate a change occurred. C describes confidence in a setting the claim does not concern, private friendship rather than public settings. D describes a different character entirely.

Keep Suggested Reasons at the Right Strength

A character repeatedly checks that the front door is locked before going to bed.

Which conclusion about the character is most directly supported by this detail alone?

A. The character is anxious about a specific, named threat

B. The character checks the lock repeatedly before bed

C. The character is haunted by a past break-in

D. The character plans to leave the house that night

B states only what the text actually shows: the repeated checking itself. A, C, and D each supply a specific motive, a threat, a past event, an escape plan, that the detail alone does not establish; repeated checking can show vigilance or unease without confirming any one of these particular explanations over the others.

Choose the Evidence That Reaches the Key Point

Claim: a bird’s crest helps it attract mates, rather than serving to regulate its body temperature.

Which finding, if true, would most strongly support the claim?

A. Crested birds mate more often than birds without a crest

B. Crest size predicts mating success even among birds kept at identical, controlled temperatures

C. The crest is brightly colored during breeding season

D. Larger birds tend to have larger crests

Both A and B support the general idea that crests matter to mating. B addresses the decisive part of this specific claim: it isolates mating success from temperature entirely, which addresses the claim’s decisive contrast between attracting mates and regulating heat. A is consistent with the claim and does not rule out temperature playing a role, since it says nothing about how temperature was controlled. C is suggestive and not a controlled test of either explanation. D concerns body size, not the mate-versus-temperature question the claim actually asks.

Match Certainty to the Evidence

A single field study found that a wetland restoration project was followed by an increase in the number of bird species observed at the site.

Which choice most accurately reflects what this finding shows?

A. The restoration proves that wetland projects increase bird diversity

B. Bird diversity increased at the restored site, but this single observational result does not by itself establish that restoration caused the increase

C. The restoration definitively caused every new species to appear

D. Wetland restoration always increases bird diversity

B preserves the observational result while withholding a causal conclusion the design does not establish: one observational field study at one site, an observed increase that the study cannot by itself trace to the restoration as its cause. A claims the finding proves a causal link a single study cannot establish on its own. C adds “definitively” and “every,” a level of certainty and completeness the finding does not support. D generalizes one site’s result into a universal rule.

Work Backward From the Claim in Weaken Questions

Claim: a new fertilizer increases crop yield.

Which finding, if true, would most directly weaken the claim?

A. Fields treated with the fertilizer yielded more than untreated fields under the same conditions

B. Fields treated with the fertilizer yielded about the same as untreated fields under the same conditions

C. The fertilizer is cheaper than competing products

D. Farmers reported the fertilizer was easy to apply

The stem asks for what weakens the claim, which means the choice that would otherwise look supportive, A, is wrong here precisely because it strengthens rather than weakens. B is the finding that actually undercuts the claim: no meaningful difference in yield despite the treatment. C and D both bear on cost and convenience, neither of which touches yield at all.

Identify the Hidden Comparison

A claim can carry a comparison without ever using the word than. Words like preserved, unusually, distinctive, disproportionately, more stable, and exceptional all carry a comparison against a reference point the sentence never names outright. Before testing any evidence against a claim like this, translate the claim into the comparison it is actually making.

A claim that a dialect preserves an old vowel sound is not really a claim about that dialect alone. It is a claim that the sound survives there and has changed or disappeared in the comparison dialects the claim is implicitly measuring against. Evidence that only shows the sound is present locally answers half the claim. Evidence has to address both sides of the hidden comparison before it can fully support or weaken a claim built this way.

Know When Absence Is Informative

The book already separates not measured from no effect. A sharper version of that same caution: absence of an observation only counts as real evidence when a hypothesis specifically predicted that observation should appear, and the method used was actually capable of detecting it. Meeting both conditions turns a missing result into a genuine finding. Missing either one leaves the absence uninformative.

A theory predicts a specific chemical signature should appear in a class of rock formations if a proposed early-ocean process actually occurred. A survey using instruments capable of detecting that signature at far lower concentrations than the theory requires finds no trace of it across every formation tested.

Which choice best describes what this survey result does to the proposed process?

A. It has no bearing on the process, since absence of evidence is never evidence of absence

B. It weakens the case for the process, since the predicted signature should have been detectable if the process had occurred

C. It proves the process never occurred anywhere

D. It confirms a competing theory that the survey never tested

The theory itself sets the expectation, and the instruments clear the detection bar the theory requires. Because the theory predicts the signature and the instruments could detect it at the required level, the null result is informative, which supports B. A applies a caution that belongs to weaker cases, not this one, where both conditions for informative absence are actually met. C overstates a weakening into a disproof. D credits a theory the passage never mentions being tested at all.

Check Whether the Measure Matches the Construct

A different problem shows up when a study’s claim and its evidence share the exact same operational definition. If a study operationally defines a construct using a specific measure, a change in that measure establishes a change in the operationalized measure. Whether that supports the broader construct depends on the validity of the operational definition.

A study claims a workplace program improved employee engagement. The offered evidence is a rise in optional meeting attendance.

Which choice best evaluates this evidence?

A. It directly measures the claimed outcome and fully supports the claim

B. It is unrelated to engagement and provides no support at all

C. It measures a plausible proxy for engagement, but it supports the broader claim only if the passage establishes that attendance validly represents engagement in this setting

D. It proves the workplace program had no effect on engagement

Attendance is not engagement itself, but it is a plausible behavioral indicator of it in many settings, which makes this a genuine proxy, not an irrelevant variable. Whether it actually supports the claim depends on a bridge the passage has to supply, not one a reader can assume. That is C. A skips past the bridge this evidence still needs. B dismisses a variable that is genuinely related, just not identical to the claim. D reverses the evidence into a conclusion it was never offered to support.

A separate study defines “civic engagement” as voter turnout, then reports that a new outreach program increased civic engagement, citing higher voter turnout among participants. The turnout figure is real evidence that civic engagement, as this study specifically defined it, increased. What it cannot establish is that civic engagement increased in every broader sense the phrase might suggest, community volunteering, local meeting attendance, or political knowledge, since the study measured none of those. Whenever a study operationalizes a broad construct as one narrow measure, ask what that measure actually captures and what it leaves out, not whether the finding itself is real.

Use the Right Rate, Denominator, and Dose

A count and a rate can move in opposite directions, and a claim that leans on one while evidence answers the other will look supported when it is not. A group can contribute more raw cases to a study purely because it is larger, entirely apart from any real difference in likelihood. Whenever a claim uses words like disproportionately, prevalence, likelihood, rate, share, or per capita, find the denominator before judging whether the evidence actually supports it.

Two hundred workplace injuries were reported at a factory employing twenty thousand people last year. A smaller factory employing five thousand people reported one hundred fifty injuries in the same period.

Which choice best compares injury risk at the two factories?

A. The larger factory is more dangerous, since it reported more total injuries

B. The smaller factory has a substantially higher injury rate per worker, despite reporting fewer total injuries

C. The two factories have identical injury rates

D. The data cannot be used to compare the two factories in any way

The larger factory’s rate is one percent of its workforce; the smaller factory’s rate is three percent, three times higher, even though its raw count is smaller. B is correct: the rates are 1% and 3%, respectively. A mistakes the larger raw count for a higher rate. C ignores a real and substantial difference the numbers reveal once each count is divided by its own denominator. D is too extreme; the data supports a specific, calculable comparison.

A related pattern worth recognizing: when a proposed cause increases in strength and a predicted effect increases alongside it in the direction a theory predicts, that graded relationship can strengthen a causal interpretation relative to a simple present-or-absent comparison. A treatment that reduces symptoms slightly at a low dose and substantially at a high dose, tracking the theory’s own prediction, is being supported by more than one data point pointing the same direction. A graded dose-response pattern matching a prior prediction can strengthen a causal interpretation relative to a single-level association, but its evidentiary value still depends on design and possible confounding.

Use This Method

1. Read the stem and find the claim. Note whose it is and whether you are supporting or weakening it.

2. Restate the claim in your own words, keeping its group, its condition, and its mechanism intact.

3. Name the demand. A comparison needs both sides measured. A causal claim needs evidence bearing on the proposed relationship, and where a rival explanation is genuinely in play, evidence that separates it from that rival. A universal claim falls to one counterexample.

4. Grant every choice as true and ask what each one actually does to the claim.

5. Choose the choice that acts on the claim itself, not on its surrounding subject matter.

Quick Review of Hard Cases

Each of these was taught earlier with a full worked example. Here, each gets its own compact illustration, so this list teaches on its own rather than just pointing back.

The evidence supports the topic, not the claim. Coated roofs being common in cold regions is a relevant fact about the topic, and it never shows that the coating itself changes heat loss.

The evidence supports only half of a two-part claim. Evidence that a bird species moved north supports the movement; the claim also needs the milder-winters explanation addressed, not just the fact of the move.

The finding would happen under both explanations. Birds leaving before a storm is predicted equally by a pressure theory and a temperature theory; only a finding that separates the two, like pressure falling while temperature holds steady, actually discriminates between them.

The rival explanation is weakened, but the proposed explanation is not directly shown. Weakening climate change as a cause of abandonment can increase the relative plausibility of a trade-route explanation, but the climate record is not itself direct evidence that trade routes were the actual cause.

The cause and result occur together, and a difference hides underneath. Farms with a new fertilizer and more irrigation both showing larger tomatoes cannot separate which factor mattered until a comparison isolates one from the other.

A comparison is claimed but only one side is informative. A strong result for Method A alone does not establish “more effective than Method B” without a matched result for B to measure it against.

The direction is reversed. A claim that income increases predicts business formation is not supported by a finding where business formation came first and income rose afterward, even though both findings involve the same two variables.

The timing makes the proposed cause impossible as a starting point. A decline that began four years before a program started cannot have been triggered by that program, whatever effect the program may have had on the rate of decline afterward.

The evidence is too broad. A schoolwide score increase may be true while telling you little about the specific below-grade-level students a claim actually names.

The evidence is too narrow. One favorable result in one climate cannot establish that a method works “across a wide range of climates,” however strong that single result looks.

A universal claim needs very different evidence from an ordinary one. One documented exception defeats never; that same exception barely touches usually, since usually already allows for cases like it.

No detected difference is treated as proof of no difference. A twelve-week trial finding no significant difference in soil moisture retention tells you about that trial’s result, not about whether the mulch truly has no effect at all.

The study did not measure the thing the answer claims. A survey comparing reported spending to receipts never measured emotional attachment to a brand, so no choice about attachment can be supported by it.

A result after an intervention is not automatically a result of the intervention. Retention rising in the months after a new onboarding program began is consistent with the program working, and it needs stronger causal evidence, such as an appropriate comparison group or randomized design, before the rise can be attributed to the program.

A purpose claim is supported by an outcome rather than an aim. Ridership rising after a bus route redesign shows an outcome. A contemporaneous stated aim is especially strong evidence of purpose; sufficiently diagnostic design or context can also support an intention when no direct statement is available.

The quotation mentions the right character but not the claimed trait. Mira packing her suitcase carefully concerns Mira and leaving; it does not by itself show reluctance, since careful packing is just as consistent with simple organization.

The quotation shows one part of a compound description. A claim that a room is both cramped and silent needs a quotation carrying both qualities at once; a line that shows one and contradicts the other is worse than a line that simply leaves one out.

The quotation names the trait but does not fit the exact claim. A claim about growing confidence needs the earlier contrast as well as the later moment; a quotation showing only the confident ending does not by itself demonstrate that anything changed.

The literary choice invites a motive the quotation never actually gives. A character checking a locked door repeatedly shows vigilance in action; naming the specific fear behind it, a past break-in, a named threat, goes beyond what the repeated checking alone establishes.

Two choices support the claim, and one addresses its decisive part. Evidence that crested birds mate more often supports the general idea; evidence that crest size predicts mating success even at a controlled temperature isolates the exact contrast, mates versus temperature, that this specific claim turns on.

The answer adds certainty the evidence does not have. One field study showing more bird species after a wetland restoration establishes that the increase happened; it does not establish that the restoration caused it, and a choice claiming proof of causation outruns what a single study can establish.

The stem asks for evidence against the claim. A finding that would normally look supportive, like a fertilizer producing a yield gain, is the wrong answer entirely when the question asks what would weaken the claim instead; reversing the direction of the question before evaluating the choices is the first move, not the last.

A genuinely independent source or replication can add evidentiary value that a second citation to the same underlying dataset does not. A different research team, using a different method, reaching the same result adds real support; a second mention of the exact same original study or dataset does not, since it is not a genuinely separate check on the finding. How much it strengthens the claim depends on method, quality, and what was replicated.

Match Evidence to the Exact Claim

Different claims call for different kinds of evidence. The goal is never to memorize a scientific rulebook. It is to ask what observation would actually make this particular claim more believable, less believable, or directly visible in a quotation.

Practice

Answers and Explanations

This last item rewards the test introduced early in the chapter: the claim sets a reused timber against a freshly cut one, so useful evidence has to separate those two specific histories. Every choice except B is equally at home under either one.

Final Checks Before You Answer

When two choices still look useful, do not ask which one sounds more impressive. Ask what exact sentence in the claim each choice actually supports, then check what the other choice changes without announcing it, leaves out, or assumes. Each check below comes with its own worked item.

Truth is only the first test a choice has to pass. A choice can be completely true, can concern the same general subject as the claim, or can support only part of the claim, and still fail to do the one job that

matters: bearing on the claim’s exact assertion.

A linguist claims that the region’s dialect preserves a vowel sound lost elsewhere in the language two centuries ago.

Which finding, if true, would most directly support the claim?

A. The dialect is spoken by fewer people today than a century ago

B. Recordings of elderly speakers in the region show the vowel pronounced in a way matching descriptions from two centuries ago, while recordings from other regions do not

C. The vowel sound appears in the region’s dialect

D. The region has a well-documented linguistic history

B addresses the full claim: preserved locally, lost elsewhere. A is true and concerns the general topic, declining speakers, without touching preservation at all. C is true and covers only half the claim; the sound’s presence locally says nothing about its absence elsewhere. D is true and generic, and supports nothing specific the claim actually asserts.

A choice can give an outcome, give a cause, or describe a finding that two rival explanations would predict equally well, and only the choice that actually isolates the claimed mechanism does real work.

A company claims a new checkout design reduced customer wait times because it processes payments faster.

Which finding, if true, would most strongly support the claim that faster payment processing is the reason wait times fell?

A. Wait times fell by two minutes on average after the new design was installed

B. The new design processes payments in half the time of the old one, and wait times fell by roughly the same proportion

C. Customers also reported friendlier staff after the redesign

D. Both the new design and a drop in daily customer volume would predict shorter wait times, and volume also fell that quarter

B directly measures the proposed mechanism and shows both faster payment processing and shorter waits. Similar proportional changes are compatible with the mechanism but are not, by themselves, proof that no other factor contributed. A gives only the outcome, with no evidence about why it happened. C is irrelevant to processing speed entirely. D describes a finding that a rival explanation, lower volume, predicts just as well as the claimed mechanism does, which makes it useless for distinguishing between them.

A choice can weaken a genuine rival explanation, show a bare association, or describe a result that followed an intervention, and only evidence tied to a real alternative actually strengthens the claim it’s meant to support.

An engineer argues that a new coating, not increased cooling airflow, explains reduced surface pitting in a batch of test parts.

Which finding, if true, would most strongly support the engineer’s claim?

A. Airflow in the test batch did not differ from a comparison batch that also improved, while coating use did differ

B. Pitting reduction and coating use rose together over the trial period

C. Pitting improved in the weeks after the coating was introduced

D. The coating contains a corrosion-resistant compound according to laboratory tests

A directly weakens the genuine rival, airflow, by showing it cannot explain a difference that coating use can, which makes the engineer’s claim more plausible by elimination. B shows only a bare association, too weak on its own to establish which factor mattered. C shows a result following the intervention, consistent with an effect without ruling out other explanations. D is a fact about the coating in a lab setting, disconnected from this specific batch’s parts.

When a proposed cause arrives too late to explain the beginning of a trend, the timing alone is often decisive, regardless of how strong the cause’s later effect might be.

A bridge design has performed well under heavy snow load since it was first built in the 1970s. Engineers reinforced a key support in 2015.

Which finding, if true, would most directly weaken the claim that the 2015 reinforcement explains the bridge’s strong performance?

A. The bridge performed well under heavy snow load throughout the 1980s and 1990s, decades before the reinforcement was added

B. The reinforcement added significant weight to the bridge’s main span

C. Engineers used a new type of steel in the 2015 reinforcement

D. The reinforcement was completed in six months

A is decisive: a cause added in 2015 cannot explain decades of strong performance that came before it existed, whatever effect it may have had on performance afterward. B and C both describe features of the reinforcement itself without addressing timing at all. D reports a construction detail with no bearing on the explanation.

An absolute claim and a hedged claim demand very different evidence, even from the exact same finding.

Claim 1: this alloy never corrodes in marine environments. Claim 2: this alloy usually resists corrosion in marine environments.

A single documented case of the alloy corroding after five years in a marine test site defeats Claim 1 outright: never admits no exceptions, and one genuine exception is enough. That same case barely touches Claim 2: usually already allows for occasional failures, so one documented exception is compatible with the claim rather than fatal to it. The finding did not change between the two claims. The wording of the claim changed how much that one finding was able to do.

A study reporting no detectable difference, a study that never actually measured the outcome a choice claims, and a choice built around one dramatic-sounding number, all fail the same way: by claiming more than the study itself delivered.

A twelve-week trial found no statistically significant difference in fuel efficiency between a new engine additive group and a control group. A separate review of maintenance logs found no data on driving conditions during the trial.

Which choice most accurately reflects what this trial shows about the additive’s effect on fuel efficiency?

A. The trial found no detectable difference in fuel efficiency between groups, which does not by itself prove the additive has no effect

B. The additive reduced participants’ reported maintenance costs by a wide margin

C. Fuel efficiency differences reached nearly eight percent in some individual cases

D. The additive has no effect on fuel efficiency whatsoever

A stays exactly at the level the trial supports: no detected difference, not a settled absence of effect. B invents a cost measurement this trial never took. C reaches for a dramatic individual figure that says nothing about the actual group comparison the claim depends on. D converts an absence of detection into a certainty the trial cannot deliver.

A choice can use the correct group named in a claim, or describe a policy’s general result, and only the one that keeps both the group and the specific outcome intact actually answers a targeted claim.

A policy was adopted specifically to produce a steeper drop in dropout rates among first-generation college students than the university was already achieving for students generally. Two years later, overall university dropout rates fell, and first-generation dropout rates fell by roughly the same amount as the rest of the student body.

Which finding would most directly show that the targeted group improved more than the rest of the student body after the policy was adopted?

A. Overall university dropout rates fell after the policy was adopted

B. First-generation students’ dropout rates fell more sharply than those of other students after the policy was adopted

C. The policy was intended to help first-generation students

D. Faculty advisors praised the policy in a survey

B is the only choice keeping both the named group and a result specific to that group, showing a differential effect the general figure in A cannot show on its own. A reports the policy’s overall result, which the passage already tells you was matched by the rest of the student body, not the targeted, additional effect the policy was specifically meant to produce. C restates intent rather than reporting an outcome. D reports opinion, not a measured result for any group.

A quotation has to be about the right character and has to instantiate the exact trait, not just share a key word, name one adjective in passing, or seem too literary to bother checking against the claim precisely.

Claim: Elena is fiercely protective of her younger brother.

Which quotation most effectively illustrates the claim?

A. “Elena’s brother often teased her about her cooking.”

B. “Elena stepped between her brother and the older boys without a word.”

C. “Elena was known throughout the neighborhood as fierce.”

D. “Elena’s brother was the only person who could make her laugh.”

B is about the right character and shows the exact trait: physically placing herself between her brother and a threat. A is about the right character and shows teasing, not protection. C contains the passage’s own key word, fierce, without ever connecting it to her brother or to protection specifically; word overlap alone does not instantiate a claim. D shows closeness and humor, a real bond and a different quality from protectiveness.

A quotation does not need to be the most dramatic line on the page to be the strongest evidence; a quiet, precise line can outperform a vivid, ambiguous one, exactly as a vivid, unambiguous line can outperform a quiet, ambiguous one.

Claim: the old man was overwhelmed with grief at the funeral.

Which quotation most effectively illustrates the claim?

A. “He wept openly, collapsing against the casket as mourners looked on.”

B. “He nodded once when his name was called, and said nothing else all day.”

C. “The funeral was held on a gray Tuesday in October.”

D. “Many people attended the funeral to pay their respects.”

A is dramatic and also the most precise fit: open weeping is an unambiguous depiction of grief. B is quieter and genuinely more ambiguous here; silence at a funeral could show grief, shock, numbness, or simple reserve, and nothing in the line settles which one. Neither literary register wins by default; fit decides, and here the more dramatic line happens to be the more precise one. C and D describe the setting and attendance, neither bearing on grief at all.

A choice can show behavior without inventing a motive behind it, and separately, it can overclaim certainty with words like proves or definitively where supports or is consistent with is the honest description of what the evidence delivers.

A marine biologist observes a dolphin pod remaining near a deceased calf for several days.

Which choice best characterizes what this observation alone supports?

A. It proves the pod experienced grief

B. It is consistent with grief-like behavior, without establishing the pod’s actual emotional state

C. It definitively rules out any explanation other than grief

D. It shows the pod actively mourned the calf’s death

B is calibrated to what a behavioral observation alone can support: consistent with, not proof of, an internal-state interpretation. A and C both claim a certainty, proves and definitively, that a single observed behavior cannot deliver on its own. D asserts “mourned” as if it were an observed fact rather than an interpretation layered on top of the observed behavior, which is exactly the kind of invented motive this check warns against.

When a stem asks what would undermine or challenge a claim, reverse your evaluation before judging the choices, and when two choices both seem to weaken it, prefer the one that needs fewer added assumptions to actually do so.

Claim: a new fertilizer formula increases crop yield.

Which finding, if true, would most directly undermine this claim?

A. The fertilizer increases crop yield in most controlled field trials

B. A large controlled trial found no increased yield among fertilized plots compared to a matched untreated group

C. Farms that use the new fertilizer tend to have better irrigation systems overall, which could independently explain any observed yield increase

D. Fertilizer sales have increased over the past decade

A looks like it belongs on a support question, and the stem here asks for the opposite; reversing direction first rules it out immediately. Both B and C can seem to undermine the claim, and they are not equally strong: B shows the effect directly absent in a controlled comparison, while C requires the added assumption that irrigation, not fertilizer, is actually doing the work, an assumption the passage never confirms. B needs less to succeed, which makes it the better choice. D is irrelevant to crop yield.

When scientific language, literary wording, or simply the density of a passage makes nothing feel decisive, strip the claim down to its plainest form before comparing it against the choices.

A pharmacokinetic analysis indicates that the novel compound’s bioavailability, when administered via the oral route, is significantly attenuated relative to intravenous administration, a differential attributed to first-pass hepatic metabolism.

Stripped down, the claim says something simple: less of the compound reaches systemic circulation after oral dosing than after an equivalent intravenous dose, because first-pass hepatic metabolism removes part of the orally administered compound before it reaches the systemic circulation.

Which finding, if true, would most directly support this claim?

A. Blood levels of the compound were measured several hours after oral dosing

B. Blood levels of the compound were far lower after oral dosing than after an equivalent intravenous dose, and levels of a liver breakdown product rose sharply only after the oral dose

C. The compound was well tolerated by patients in both dosing groups

D. The oral dose was three times larger than the intravenous dose

Once the claim is stripped to plain language, B is the only choice that actually supplies both halves of it: the comparison between routes, and the liver mechanism behind the difference. A gives a measurement with nothing to compare it against. C concerns tolerability, not blood levels. D introduces an unequal dose that would confound the comparison rather than support it. The identical stripping technique works on dense literary prose: reduce a heavily worded character description to the one trait or condition the quotation actually has to show, and work from that plain version instead of the ornamented one. When the choices remain close, recheck the claim for limiting words such as more, only, because, all, intended, or changed.