Inferences

CHAPTER EIGHT · INFORMATION AND IDEAS

A friend tells you: “The last time Marcus canceled plans this last-minute, it turned out he’d lost his job the week before and didn’t want to say so yet.” You don’t need to be told what to think the next time Marcus cancels last-minute. That is an inference: moving from a stated pattern to an unstated but supported conclusion. You also know, just as instinctively, the difference between that reasonable bridge and an unsupported one: concluding Marcus has lost his job specifically because he canceled once is a much bigger leap than concluding something might be wrong.

College Board calls this skill Inferences. A question gives you a passage that stops short of saying something outright, sometimes literally breaking off mid-sentence at a blank, and asks what follows most logically from what the passage actually gives you. The material tested can be stated outright or implied, and the correct answer has to be the choice the text supports most strongly while adding the fewest assumptions of the four. A common Inferences format presents a passage whose final thought is completed by the answer choice, but the skill is drawing the best-supported inference, not filling a blank as such.

Choose the Conclusion the Evidence Supports

The answer to an inference question may paraphrase language already in the passage, and it must express a conclusion the evidence actually supports rather than simply repeat a detail you were asked to locate. What the answer cannot do is rest on a fact the passage never supplied. Read the passage as a small argument with its final step missing, and that missing step is often visible before you ever reach the answer choices.

When the direction of the argument is clear, predict the conclusion before you compare choices. When it is not clear, treat each choice as a hypothesis and test it against the passage directly. Several choices may be true in the real world and completely unsupported by this specific passage; reading the choices cold, before you have a prediction of your own, invites you to build a case out of assumptions you supplied yourself rather than ones the passage gave you.

Use Connectives as Logical Clues

Many of these passages place a connective word directly in front of the blank, and that word narrows what can logically follow it. Read it closely when it is there. When no connective appears, take the relationship from the passage itself, since a real share of these items give you no signal word at all.

When the connective directly governs the blank in the way shown below, it narrows the field as follows:

Therefore, thus, and so require a consequence of what the passage just established; they cannot introduce a choice that adds nothing to the sentence before it. However, but, and yet require something running against the expectation just raised; they cannot be followed by anything that simply continues in the same direction. Because and since require the reason for the statement in front of them; they cannot be followed by a result, which would run the causal arrow backward. This suggests that requires a cautious conclusion, hedged exactly as the phrase itself is hedged; it cannot be followed by a claim of proof. In other words requires the same content restated more plainly, sometimes drawing out what it already implies; it cannot introduce a new fact the passage never gave. It follows that requires a conclusion strongly supported by what precedes it; it cannot introduce a guess that needs a fact from outside the passage.

Reading the stub aloud with your own prediction attached. Take the words that survive right before the blank and finish the sentence yourself, in your own words. The completion has to satisfy both the logic of the passage and the grammar of the sentence it completes, so check the two separately before you commit to an answer.

Recognize Common Inference Patterns

Many of these passages hand you two facts and leave the connection between them unstated. Some turn on a single decisive fact, and some need three facts held together. Read the patterns below as shapes that recur often, not as a fixed architecture every single item follows; each one is valid only under the specific condition named alongside it.

A passage giving a necessary condition, then a case where that condition is absent, supports the conclusion that the outcome requiring it cannot occur, but only when the passage actually states the requirement, in wording such as only when or is required for. A passage giving an expectation, then a measurement running against it, supports the conclusion that the expectation needs revising, in the specific respect measured. A passage giving a requirement, then a supply falling short of it, supports the conclusion that the requirement goes unmet. A passage giving two groups alike in one way, then a difference in outcome between them, supports the conclusion that the shared feature cannot by itself explain the difference, though it may still contribute to the outcome in both groups. A passage giving a timing, then a second timing that no longer matches it, supports the conclusion that if two processes must overlap for an outcome, and their timing no longer overlaps, that outcome may be disrupted.

Match Inference Strength to the Evidence

A common distractor keeps the correct topic and the correct direction and overstates the certainty or the scope without announcing it. Match the strength of your own language to the strength of the passage’s language.

If the passage says a factor may contribute to an outcome, a conclusion may say it is one factor associated with that outcome; it may not say it causes or is responsible for the outcome. If the passage says something held in the plots studied, a conclusion may say it held in these plots; it may not say in all soils or generally. If the passage says researchers suggest, a conclusion may say researchers propose that; it may not say researchers have shown that. If the passage says no difference was found, a conclusion may say the study did not detect a difference; it may not say the factor has no effect, which a single null result cannot establish on its own. If the passage says an effect held for most of the sample, a conclusion may say a majority; it may not say every case or without exception.

Separate Supported Conclusions From Possible Ones

Two statements can both sit comfortably beside a passage, and one of them rests entirely on the passage while the other rests on something you brought to it yourself. Distractors of the second kind are the ones most often chosen.

The mill closed in 1971, and the town lost four hundred jobs within a year. Census records show the population falling steadily from 1972 onward.

A statement the passage actually supports: the population decline began after the mill closure and around the period in which the town lost four hundred jobs. A statement that is merely possible: families left because they could not find other work. The second reading feels entirely natural, and it rests on a premise about motive that the passage never actually supplies, since census records count departures, not the reasons behind them.

To separate a supported inference from a merely possible one, ask what unstated premise each choice requires. The best answer requires no unsupported factual premise beyond what the passage already gives, while a merely possible choice needs one you supplied without quite noticing you’d done it. A merely possible choice can still be a reasonable conclusion drawn from the facts on the page; reasonable is not the same test as best supported. Rank the survivors by how much each one assumes, and prefer whichever one is supported most directly.

Comparison groups can rule explanations out. When a passage adds a second species, country, group, or time period, ask what the two cases share and what actually differs between them. A feature shared by both cases cannot, by itself, explain an outcome present in only one of them. A feature that genuinely differs between the cases becomes relevant, though difference alone still does not prove causation on its own.

Several changes at once create the opposite problem. If temperature rose, habitat changed, and predators declined before the same outcome appeared, the passage has supplied several candidate causes at once. Unless the passage’s own evidence separates them, an answer that selects just one of them as the cause goes further than the passage actually supports.

Worked Examples of Supported Inferences

Use Timing to Separate Events

The orchard’s apples are pollinated by a solitary bee that emerges in the second week of April and flies for about fourteen days. The old apple variety blossomed in that same second week of April. Growers have now replaced it with a variety that blossoms three weeks earlier, and records show the bees still emerging on their usual date. The change therefore ___

A. risks leaving the blossom open before its pollinators are flying

B. proves that the new variety produces less fruit than the old one

C. will draw bees into the orchard from neighboring valleys

D. shows that the bees have adjusted to the earlier blossom

The passage gives the bees’ flight period, the old blossom date that matched it, and a new blossom date three weeks earlier. Therefore asks for the direct consequence of that mismatch, which is A: the old blossom coincided with the flight period, the new blossom arrives three weeks earlier, and the bees keep their usual date, leaving a real gap. B invents a yield comparison that never appears; proves claims a certainty two dates alone cannot reach. C introduces bees from neighboring valleys the passage never mentions. D states the exact opposite of what the records show; the bees kept their usual date rather than adjusting to the new one.

Stop at the Evidence Limit

Volunteers who practiced a musical passage in the evening performed it more accurately the following morning than volunteers who practiced the same passage after waking. The researchers note that the evening group had also gone longer without a break before the test.

Based on the text, which conclusion is most logically supported?

A. Practicing before sleep consolidates skill more effectively than any other factor

B. The results fit a role for sleep in skill consolidation, and the difference in break timing leaves the matter open

C. Break timing rather than sleep produced the difference in performance

D. Practicing on waking prevents a musical passage from being learned

B keeps the actual finding and carries the researchers’ own caution alongside it, which preserves the study’s unresolved confounding. A ranks sleep against “any other factor,” a field of alternatives this two-group study never actually tested. C is the mirror image of A: break timing is a real rival explanation, and the study picks no winner between the two. D converts a smaller score into total failure; the morning group performed less accurately, and less accurate is not zero.

Apply the Rule to the Case

The fungus produces fruiting bodies only after a week in which night temperatures stay above twelve degrees. Weather records from the reserve show no such week during the 2023 season. It follows that in the reserve in 2023 the fungus ___

A. produced no fruiting bodies

B. fruited later in the season than usual

C. was killed by the cold nights

D. will not fruit in future seasons either

The first sentence sets a necessary condition for fruiting. The second sentence reports that condition was absent all season. The conclusion, A, follows from those two facts alone: only after makes the warm week a genuine requirement, and the weather record shows it never occurred. B still describes fruiting happening, which the missing condition rules out entirely, whether early or late. C adds a fate, being killed, that the passage never reports; absence of fruiting says nothing about the fungus’s survival. D reaches past the evidence entirely; one season sits on the page, and future seasons do not.

Use a Negative Finding to Narrow the Conclusion

Two districts adopted the same reading programme in the same year. Scores rose sharply in one district and stayed flat in the other. A review found that teachers in both districts had received the same curriculum, the same number of training hours, and the same instructional support. The review concludes that differences in teacher training ___

A. do not account for the gap between the two districts

B. have no bearing on reading scores

C. explain why scores rose in the first district

D. were greater than the review had expected

A stays exactly inside the two districts the review actually compared: training was held equal across both, so it cannot by itself explain why their outcomes diverged. B overgeneralizes from two districts in one review into a general law about reading scores everywhere. C asks an equal factor to somehow produce unequal results, the opposite of what the review found. D directly contradicts the review’s own finding that training was equal, leaving nothing to be “greater” than expected.

Combine Stated Facts Carefully

A correct inference can state something no single sentence in the passage says outright, as long as it only combines facts the passage actually supplied. New wording is allowed. A new factual premise is not.

Cold tolerance becomes more costly to maintain as temperature falls, and higher elevations are colder. Researchers surveying three mountain ranges found progressively fewer species at each successive elevation band.

Which choice most logically completes the text: “The pattern the researchers observed ___”

A. is consistent with fewer species being able to tolerate the increasing physiological cost of cold tolerance at higher elevations

B. proves that cold tolerance evolved specifically as a response to elevation

C. shows that every species has a fixed elevation limit it cannot cross

D. will reverse if average temperatures rise in the future

A combines the two given facts, rising physiological cost and colder elevations, into a conclusion neither sentence states alone, without adding anything beyond them or claiming the passage has ruled out other explanations. B converts an explanation into an evolutionary origin story the passage never supports. C converts a general pattern into a fixed, universal rule about “every species.” D predicts a future reversal on a topic the passage never addresses.

A choice can also fail by explaining why when the passage only shows what. Participants in a study might justify choices they never actually made, which supports a mismatch between the choice and the later explanation; it does not by itself identify the unconscious process that caused that mismatch. A passage reporting only the mismatch cannot support a completion naming its underlying cause.

Preserve Causal Limits

Dense colonies of a species might show high carbon dioxide levels and a particular trait more often than sparse colonies. The association may be entirely real, and “the trait evolved in response to carbon dioxide” needs considerably more than an association to support it.

A city’s population grew sharply over a decade. During the same period, average temperatures rose, a new highway opened, and a major employer relocated to the area.

Which choice most logically completes the text: “The population growth ___”

A. cannot be attributed to any single one of these changes based on the passage alone

B. was caused primarily by the new highway

C. occurred because of the rise in average temperatures

D. would not have happened without the new employer

A is the only choice that respects what the passage actually establishes: three changes occurred alongside the growth, and nothing in the passage separates their individual contributions. B, C, and D each select one candidate cause from the three and assign it sole or primary credit the passage never actually assigns.

A choice can also fail by treating one of two possible explanations as settled. If a passage says either A or B could explain a result, and an answer choice picks B, that choice has added a certainty the passage itself never reached.

Know What Actually Justifies a Causal Claim

A correlation can arise because a third factor influences both variables; observational co-occurrence alone does not eliminate that possibility. One specific design feature helps close that gap. When participants are placed into groups by random assignment, treatment assignment becomes independent of participants’ preexisting characteristics in expectation. Random assignment reduces systematic confounding and tends to balance known and unknown factors across groups, especially in larger samples, but it does not guarantee exact balance on every factor in a finite study. With that caveat, a later difference between groups can typically be attributed to the treatment itself, not to a preexisting group difference.

Researchers wanted to know whether a new fertilizer increases crop yield. Two hundred plots were assigned to receive the fertilizer or a placebo treatment using a random number generator, with neither the researchers scoring the plots nor the workers tending them told which plots received which treatment. Plots receiving the fertilizer produced 18 percent more grain on average.

Which choice most logically completes the text: “The study’s design ___”

A. supports a causal conclusion, because random assignment reduces systematic pre-treatment differences between groups and supports causal attribution of the average outcome difference to the assigned treatment, subject to the study being carried out as described

B. supports a causal conclusion, because the yield difference was large

C. cannot support a causal conclusion, because a single growing season is too short to draw any conclusion

D. cannot support a causal conclusion, because field studies are inherently less reliable than laboratory studies

A names the actual feature doing the work: random assignment, not the size of the effect, is what licenses the causal conclusion. B mistakes a large effect for a well-designed one; an equally large difference from a poorly designed study would prove nothing. C and D both reject the causal conclusion for reasons the passage never raises, season length and setting, instead of engaging with the actual design feature described.

An observational study finding the same 18 percent difference, with farmers choosing for themselves whether to use the fertilizer, would support only an association: farmers who chose the fertilizer might also have better soil, more experience, or more resources, and any of those could explain the yield difference on its own.

Recognize When a Sample Silently Excludes Itself

Two related traps hide inside how a sample was gathered, not in what it measured. Selection bias shows up when the people or cases in a study got there through some process connected to the very trait being studied, so the sample no longer represents the group the conclusion is about. A survey of gym members about exercise habits says something about people who already join gyms, not about people generally. Survivorship bias is a narrower selection problem: only cases that remain observable are included, so whatever caused the others to drop out never appears in the data at all.

A researcher studies successful startup founders and finds that most describe having taken large financial risks early in their careers. The researcher concludes that taking large financial risks tends to lead to startup success.

Which choice most directly weakens this conclusion?

A. The founders were interviewed several years after their companies became successful

B. Many founders who took similarly large financial risks failed and were never included in the study

C. Some founders described their early risks as stressful at the time

D. The study included founders from several different industries

B identifies the missing group whose absence makes the conclusion unreliable: if failed risk-takers were never studied, there is no way to know whether large risks mostly led to success or mostly led to failure that simply never appeared in this sample. A concerns timing of the interviews, not who was included. C describes an emotional reaction, not a flaw in the sample. D describes variety within the sample without addressing what the sample is missing.

Keep Possibility, Probability, and Certainty Separate

Could, may, and is consistent with leave real alternatives open. Likely or probably requires evidence that actually favors one possibility over the others, not just evidence that is compatible with it. Separately, likely does not become must; strong evidence can still leave a passage short of inevitability.

A marine survey found that reef fish populations near a protected zone were larger than those near unprotected areas of similar depth and habitat.

Which choice most logically completes the text: “The comparison ___”

A. suggests that protection may be a contributing factor, since matching the protected and unprotected areas on depth and habitat reduces those factors as alternative explanations

B. proves that protection causes fish populations to grow

C. is consistent with several possible explanations equally, since fish populations can vary for many reasons

D. shows that unprotected areas will never support large fish populations

Matching depth and habitat genuinely narrows the field, reducing two obvious measured differences as explanations, which is real information a comparison without that matching would not supply. A reflects exactly that: a hedge, may, paired with the specific reason the field has narrowed, not a claim that protection is confirmed. B claims proves and causes, a certainty no single observational comparison can reach, matched or not. C treats every possible explanation as equally live, which ignores the two factors the matching has already reduced. D turns a comparison into a permanent, universal claim about unprotected areas generally.

Keep Population, Time, and Setting Fixed

A result in one species, age group, language, city, or sample cannot automatically become a claim about every member of a larger category, the distant past or future, or a different setting entirely.

A twelve-week trial tested a new fertilizer on wheat grown in sandy soil.

Which choice most logically completes the text: “The trial’s results ___”

A. apply specifically to wheat grown in sandy soil under the tested conditions

B. apply to all crops grown in any soil type

C. establish how the fertilizer will perform in future decades

D. show the fertilizer works equally well in clay soil

A keeps the conclusion inside the exact population, crop, soil type, and time window the trial actually tested. B enlarges the population from one crop and one soil type to all crops and all soils. C enlarges the time period from twelve weeks to future decades. D enlarges the setting from sandy soil to a soil type the trial never tested at all.

A choice can also confuse what was tested with what exists. Researchers failing to detect a difference under the specific conditions they tested does not prove no difference exists under any condition whatsoever; the trial speaks only to the conditions it actually used.

Ground Motive and Attitude in the Passage

A person delaying, returning, hiding, repeating, or refusing something can support a limited inference about the behavior itself; a specific feeling behind it, such as fear, jealousy, guilt, generosity, or strategy, needs its own separate evidence.

A city council debated a new fee for three sessions before adopting it. Council members made no public statement about their reasoning during any of the sessions.

Which choice most logically completes the text: “The council’s decision process ___”

A. involved extended deliberation before the fee was adopted

B. reflected members’ anxiety about public backlash

C. was intended to minimize opposition from residents

D. revealed the council’s true priorities

A stays with the visible behavior: three sessions of debate before adoption. B invents a specific emotion, anxiety, that the silence alone cannot establish. C invents an intention, minimizing opposition, that an outcome alone does not automatically reveal; a policy’s effect does not by itself disclose the purpose behind adopting it. D infers a general attitude, “true priorities,” from limited behavior that supports something much narrower.

A choice can also infer attitude from mere attention. An author spending several sentences discussing a theory has not thereby revealed approval, skepticism, or admiration; length of discussion alone does not settle which attitude, if any, the author actually holds.

Keep Direction and Strength Aligned

Evidence may support can, may, or is associated with, while a tempting answer choice upgrades that same evidence to will, must, or is caused by without announcing the upgrade. Separately, an answer can feel obviously true in real life and still fail, if it depends on a bridge the passage itself never supplies.

Facts: a manuscript uses paper made after 1812.

Which conclusion follows validly from this fact alone?

A. The manuscript cannot have been written before 1812

B. The manuscript was written soon after 1812

C. The manuscript was written no later than 1815

D. The manuscript’s author had access to paper mills

A states exactly what the single fact establishes: a floor, not a ceiling. B adds “soon after,” a specific window the paper’s date alone cannot supply; that narrower window is a bridge invented rather than given. C invents a latest possible date the fact never establishes. D infers something about the author’s access to mills that the paper’s existence alone does not confirm; someone else could have supplied it.

Respect Limits and Complete Comparisons

Evidence may establish not before 1812 without establishing 1813, 1820, or any other exact date; a boundary can be the strongest conclusion the evidence licenses even when it does not supply an exact date. Separately, a comparison is only possible when both sides of it are actually supplied.

One dating method is accurate 90 percent of the time under laboratory conditions.

Which conclusion follows validly from this fact alone?

A. This method’s accuracy cannot be determined to be higher or lower than any other method not described here

B. This method is more accurate than radiocarbon dating

C. This method is the most reliable dating method available

D. This method is accurate exactly 90 percent of the time in every setting

A correctly declines to rank this method against methods the passage never describes. B invents a specific comparison to radiocarbon dating, a method never mentioned. C ranks this method as “most reliable” against an unstated field of competitors. D converts a laboratory figure into a claim about every setting, when the passage specifies laboratory conditions only.

A choice can also infer a ranking from incomplete information. Knowing that A exceeds B, and that B exceeds C, supports the conclusion that A exceeds C. Knowing only that A exceeds B says nothing at all about C.

Track Conditional and Causal Direction

If X, then Y does not mean that observing Y proves X; the same outcome may have other causes entirely. Separately, if low temperature raises metabolic cost, observing high metabolic cost does not by itself prove low temperature caused it. And separately again, failing to find fossils in one survey does not necessarily mean the species never lived in that location.

If a bridge’s expansion joints are properly maintained, the deck will not crack under normal temperature swings. A survey found the deck of one bridge free of cracks after a decade of normal use.

Which choice most logically completes the text: “This finding ___”

A. is consistent with proper maintenance, though the deck could also have avoided cracking for other reasons

B. proves the expansion joints were properly maintained

C. shows that improperly maintained joints always lead to cracking

D. means the joints were never inspected

A respects that the conditional runs in one direction only: proper maintenance predicts no cracking, and no cracking does not, by itself, prove proper maintenance, since other factors could also explain an intact deck. B reverses the conditional, treating the effect as proof of the specific cause. C reverses the original conditional entirely, turning a claim about proper maintenance into a claim about improper maintenance. D introduces inspection history the passage never mentions.

Use Counterexamples and Late Turns

One valid exception defeats a claim worded as all or always; that same exception can leave a claim worded as often, generally, or most completely intact. Separately, a passage can build one expectation and then use however, yet, despite, or regardless to introduce the fact a completion actually has to respect.

Predators generally avoid brightly colored insects, which usually signal toxicity. One brightly colored beetle species, however, is entirely edible and preyed upon regularly by local birds.

Which choice most logically completes the text: “This beetle species ___”

A. shows that bright coloration does not guarantee toxicity in every case

B. proves that bright coloration never signals toxicity

C. shows that all local birds have evolved immunity to insect toxins

D. suggests predators generally ignore coloration when hunting

A matches the passage exactly: one documented exception to a generally holding pattern, which weakens the universal version of the claim without erasing the general pattern itself. B overreaches from one exception into a blanket denial the word generally in the passage already rules out. C invents an evolutionary claim about “all local birds” the passage never makes. D contradicts the passage’s own opening claim that predators generally do respond to coloration.

Let the Final Sentence Set the Limit

A study can report a real effect, and then a final sentence can state that the effect disappears under one specific condition; the correct completion has to include that limit rather than simply repeat the original, more sweeping-sounding result.

A new coating reduced corrosion on steel samples by half in humid conditions. Under freezing conditions, however, the coating provided no measurable protection.

Which choice most logically completes the text: “The coating’s protective effect ___”

A. depends on the conditions it is used in, and does not extend to freezing temperatures

B. reduces corrosion on steel samples in general

C. is the most effective corrosion treatment available

D. will improve if applied in a thicker layer

A preserves the freezing-condition limit introduced by the final sentence. It does not repeat only the earlier, broader-sounding headline result. B restates the first sentence and drops the freezing-condition finding entirely. C invents a ranking against other treatments the passage never mentions. D invents an untested modification, a thicker layer, that the passage never describes.

A passage full of names, or full of technical vocabulary, does not change the underlying task. The names can be mentally simplified to A, B, group 1, and group 2; what matters is who found what and which fact belongs to which group. Unfamiliar biological or archaeological terms can sit inside a passage whose actual inference depends on a simple relationship: earlier or later, more or less, same or different, or one explanation against another.

Find What Rival Explanations Share

A passage can offer two different explanations for the same finding without committing to either one. The question that follows is not always which explanation wins. Sometimes it asks what remains true no matter which explanation turns out to be correct, and the answer is the intersection of the two, not a choice between them.

Bacteria of the same strain turn up in both the leaves and the roots of a plant. One explanation holds that soil bacteria splashed upward onto the leaves during rain. A second explanation holds that airborne bacteria settled on the leaves first and were later carried down into the soil.

Which statement is best supported by the passage?

A. Soil bacteria are the more likely source, since splashing during rain is a common transfer mechanism

B. The bacteria found in the leaves and the roots may share a common source

C. Airborne transfer is confirmed as the mechanism connecting the two locations

D. The presence of bacteria in both locations cannot be explained by any known process

Both explanations disagree about direction, splashing upward against settling downward, but they agree on something neither one individually proves alone: both explanations imply movement or a shared transmission history connecting the same strain across leaves and roots. That shared implication is what the passage actually supports, which is B. A and C each pick a specific winner the passage never confirms. D denies an explanation the passage offers two separate accounts of.

The general rule: when a passage presents two candidate explanations without settling between them, ask what the explanations still share, not which one sounds more convincing. If the data cannot separate the two accounts, the best-supported inference is often the narrow claim both accounts would force to be true, not a decision between them.

Combine Partial Explanations Carefully

A different inference pattern shows up when a passage describes two components, each handling only part of a task, whose combination performs better than either part alone. The supported inference is that the combined performance comes from covering more of what the task actually requires, not from either component individually improving.

One solar cell material efficiently captures a narrow band of the light spectrum. A second material captures a different, non-overlapping band. A cell built by layering both materials together outperforms a cell made from either material alone.

Which statement is best supported by the passage?

A. The combined cell is more efficient because it captures more of the total spectrum than either material alone

B. The first material is a flawed design compared with the second

C. Future solar cells should use only the second material

D. The combined cell’s advantage comes from a chemical reaction between the two materials

Each material is described as covering its own separate band, and the complementary non-overlapping bands make broader spectral capture the supported explanation offered by the passage. That is A, the inference the passage most directly supports. B misreads a difference in coverage as a flaw in one material. C draws a design recommendation the passage never makes. D invents a chemical interaction the passage never describes; the improvement comes from combined coverage, not a reaction between the materials.

Recognize When an Additional Factor Is Needed

A passage can describe an expected process, then report a finding that process alone would not predict. A supported inference is that the stated process is incomplete on its own; naming a specific additional cause would require more evidence.

Cooling magma is expected to produce mostly a particular kind of rock low in silica. Surveys of a planet’s crust instead find substantial amounts of a silica-rich rock that this process alone would be unlikely to generate in such quantity.

Which statement is best supported by the passage?

A. The cooling-magma process never occurred on this planet

B. Some additional geological process likely contributed to the crust’s formation

C. The measurements of silica content must be inaccurate

D. A specific volcanic event, not described in the passage, produced the silica-rich rock

The passage establishes an expected pattern and a finding that pattern alone does not account for. The narrowest supported conclusion is that something beyond the expected process was involved, without specifying what. That is B. A overreaches into denying the process happened at all, when the passage only shows it cannot be the whole story. C dismisses data the passage never questions. D invents a specific event the passage does not name, a level of detail the evidence does not support.

Keep Conditional Logic in the Correct Direction

A strict conditional “if X, then Y” supports two logically valid moves: X allows Y to be concluded (modus ponens), and not-Y allows not-X to be concluded (the contrapositive). It does not support reversing the statement from Y to X or from not-X to not-Y. Several distinct versions of that reversal mistake show up often enough to name individually.

Knowing Y occurred does not let you conclude X caused it; other paths to Y may exist unless the passage says X is the only route. Knowing X did not occur does not let you conclude Y did not occur, for the same reason. “X only if Y” means Y is necessary for X, not that Y guarantees X. “X if Y” means Y is sufficient for X, not that Y is required for X to happen some other way. When a passage states that X occurs only when both Y and Z are true together, neither Y nor Z alone is enough; treating either one in isolation as sufficient is exactly the reversal this pattern is built to catch.

The same discipline applies when a passage supplies more than one bound and asks what can be concluded from combining them. A material confirmed no earlier than 1812 and confirmed to exist by 1840 was created somewhere within that span, not at any single point inside it that a choice might tempt you to name specifically. A value reported as at least 35 combined with a separate value reported as no greater than 60 supports a combined range from 35 through 60, and nothing narrower than that combined interval is actually justified by the two facts together.

For the related question of when a missing observation counts as real evidence, the same expected-observation test taught for evaluating written evidence applies here without modification: absence only weakens a claim when the finding was actually predicted, actively sought, and detectable by the method used.

Use This Method

1. Read to the blank and note whatever word stands directly in front of it.

2. Name the two facts, or the two or three facts, the passage has actually put in play.

3. Finish the sentence in your own words before comparing the choices.

4. Match your own ending against the four choices, looking for the closest fit rather than an exact word match.

5. Test the survivors by asking how much each one assumes beyond what the passage actually gives.

Compare Close Conclusions

The strongest answer on this question type is not automatically the narrowest one. It is the most complete conclusion the supplied facts actually justify. The pairs below show how one word can alter scope, probability, or logical force.

Practice

Answers and Explanations

This last item is the clearest case of a conclusion that stops exactly where its evidence stops. The watermark establishes an earliest possible date, not an exact date and not a latest possible date.

Final Checks Before You Answer

A hard inference answer can look less obvious simply because it combines several pieces of the passage at once. That is allowed. The test is not whether the answer appears word for word. Ask whether every important part of the answer follows from the passage without adding an unsupported factual premise. Each check below comes with its own worked item.

An answer not stated word for word is normal for an inference, and an answer combining two separate sentences can be exactly what the question requires; neither of these makes an answer wrong. Sounding obvious is also not a problem, as long as the obvious-sounding conclusion actually follows from the text.

A vineyard’s yield fell by a third the year after a new irrigation system was installed. Soil records from that year show unusually low rainfall across the entire region.

Which choice most logically completes the text: “The lower yield ___”

A. cannot be attributed to the irrigation system alone, given the regional drought recorded the same year

B. proves the new irrigation system was defective

C. shows irrigation systems always reduce yield in dry years

D. means the vineyard should return to its old irrigation method

A combines two separate sentences, the yield drop and the regional drought, into a conclusion neither sentence states alone, and it sounds like a fairly obvious point once both facts are held together; obvious is fine here because the text actually supports it. B blames a single cause the regional drought calls into question. C converts one case into a universal rule about irrigation systems. D recommends an action the passage never proposes.

A sophisticated-sounding answer still fails if it depends on a factual premise the passage never supplies. Separately, an answer that supplies a cause has to be checked against what the passage actually established, not against how reasonable a cause sounds.

A species of moth in one forest shows a color pattern closely matching the bark of the tree it rests on. Researchers have observed the moth only on this one tree species.

Which choice most logically completes the text: “The color pattern ___”

A. matches the bark of the tree species where the moth has been observed

B. evolved through a multi-generational selective process driven by predator vision

C. demonstrates a well-documented evolutionary mechanism at work

D. developed because the moth’s ancestors also rested on this tree

A stays with the observed match itself, adding nothing beyond it. B sounds sophisticated and supplies an entire evolutionary mechanism, predator vision and multiple generations, that this single observation cannot establish. C sounds scientifically plausible in the same way and adds nothing checkable; “well-documented mechanism” is asserted, not shown by this passage. D supplies a cause, ancestral resting habits, that the passage never actually reports.

An evolutionary explanation for a pattern needs more than the pattern itself, naming one cause among several needs to rule out or at least acknowledge the others, and the word could is often appropriately cautious rather than a weakness in an answer.

A desert lizard’s coloration closely matches the sand of its home range. Population density, predator abundance, and sand color all vary together across the range’s different regions.

Which choice most logically completes the text: “The coloration ___”

A. could reflect selection for camouflage, though the passage does not rule out other factors that vary across the same regions

B. evolved specifically because sand color created selective pressure for camouflage

C. is best explained by predator abundance rather than sand color

D. shows that population density has no effect on coloration

A uses could appropriately, given that population density, predator abundance, and sand color all vary together and the passage supplies no way to separate their individual contributions. B commits to camouflage as the specific cause, more certainty than three intertwined variables can support. C picks a single rival cause with no more basis than the one B picked. D denies a factor’s relevance that the passage never actually tested in isolation.

Check probability words. Likely should be supported by evidence favoring one conclusion over relevant alternatives. Must requires the passage to leave no relevant alternative consistent with the stated conditions. A choice claiming all or always is defeated by a single supported exception.

A field trial found that plots treated with a new fertilizer produced higher yields than untreated plots in every one of six farm sites.

Which choice most logically completes the text: “The consistency across all six sites ___”

A. makes it likely that the fertilizer itself, rather than a site-specific factor, produced the higher yield

B. means the fertilizer must work for every plot that receives it

C. proves the fertilizer always doubles crop yield

D. shows the untreated plots received no real soil at all

A uses likely exactly as the consistent, multi-site pattern actually earns it: real evidence favoring the fertilizer over a site-specific explanation, without claiming certainty. B claims must, ruling out every possible exception the trial cannot actually rule out. C claims always and doubles, both far beyond a higher yield. D misdescribes what an untreated plot is, inventing a claim the passage never makes.

Keep the population fixed. A choice claiming most has to be checked against whether the evidence really reaches a majority rather than merely some cases; a choice claiming the finding applies to everyone has to be checked against the actual population studied; and a choice claiming the finding applies elsewhere has to be checked against whether the setting was actually limited.

A survey of factory workers at one plant found that most respondents who used the new safety equipment reported fewer minor injuries.

Which choice most logically completes the text: “The survey’s finding ___”

A. applies to most respondents at this plant who used the equipment, based on the reported data

B. shows that all workers everywhere benefit from this equipment

C. applies equally to office workers who never used the equipment

D. demonstrates that minor injuries have been eliminated industry-wide

A keeps the claim at exactly the group and scope the survey actually measured: most respondents, at this plant. B expands “most” at one plant into “all” everywhere. C applies a factory-safety finding to a population, office workers, that was never studied. D expands a reported drop in minor injuries into full elimination across an entire industry.

Keep the time window fixed. A choice claiming a finding applies forever has to be checked against the actual observed time window; a choice giving an exact date has to be checked against whether the evidence supplies only a boundary; and a choice giving an exact amount has to be checked against whether the text supplies only a minimum, a maximum, a range, or a direction.

A lake’s water level has been recorded as higher every summer for the past nine years than it was the summer before construction of an upstream dam began.

Which choice most logically completes the text: “Compared with its level before the dam, the lake’s water level ___”

A. has been higher throughout the nine years recorded since construction began

B. will remain higher permanently

C. rose to exactly the level engineers originally projected

D. rose by precisely four feet since construction began

A stays inside the nine recorded years, the exact window the passage actually describes. B extends that pattern into a permanent, unbounded future the passage cannot speak to. C invents a specific engineering projection the passage never mentions. D invents an exact figure, four feet, when the passage supplies only a direction, higher, and no specific amount at all.

A choice claiming no effect has to be checked against whether the passage reports a measured zero, a failure to detect a difference, or simply the absence of any measurement at all. A choice claiming something is impossible makes a strong claim that failure to observe it usually cannot support. A choice naming a specific emotion needs behavior or narration that actually distinguishes that emotion from other plausible ones.

A twelve-month audit of a small clinic found no cases of a rare complication among two hundred procedures. The complication is known to occur in roughly one of every five hundred procedures nationally.

Which choice most logically completes the text: “The absence of cases at this clinic ___”

A. is consistent with the clinic’s low procedure count, given how rarely the complication occurs nationally

B. proves the complication cannot occur at this clinic

C. shows the clinic’s staff are more skilled than average

D. suggests the clinic’s patients feel anxious about the procedure

A correctly treats zero observed cases in two hundred procedures as unsurprising given a roughly one-in-five-hundred national rate, rather than as evidence of anything unusual. At a 1-in-500 rate, the expected number of cases in 200 procedures is 0.4, so observing zero cases is entirely plausible. B claims impossibility that a small sample and a low base rate cannot support; the complication is rare, not absent. C invents a skill comparison the audit never measures. D invents a specific emotion, anxiety, that a complication count cannot speak to at all.

A choice naming a specific motive or intention needs the reason for an action to be actually supplied, not inferred solely from what happened afterward; and a choice that reverses a conditional treats “if X then Y” as though it also meant “if Y then X,” which it does not.

A company reorganized its customer service department. Customer satisfaction scores rose in the following quarter.

Which choice most logically completes the text: “The reorganization ___”

A. was followed by a rise in customer satisfaction scores in the next quarter

B. was intended specifically to raise satisfaction scores

C. succeeded because management correctly diagnosed the department’s problems

D. shows that satisfaction always rises after a company reorganizes

A reports only the sequence the passage actually gives, without asserting why the reorganization happened or claiming certainty about its effect. B infers an intention from an outcome; a later rise does not by itself reveal what the reorganization was intended to do. C invents both a motive, correct diagnosis, and a causal claim the passage never establishes. D reverses the specific finding into a general rule, and does so by treating one outcome as if it guaranteed by itself that the reorganization caused it, which a single before-and-after sequence cannot establish.

A choice that reverses cause and result keeps the same two variables and simply swaps which one produced the other. A choice that gives a comparison needs both sides of that comparison actually supplied. A choice that gives a ranking needs every relevant competitor accounted for, not just the one or two mentioned.

Metabolic rate rises when an animal is exposed to cold temperatures. Field measurements found high metabolic rates in animals living at high elevation.

Which choice most logically completes the text: “The high metabolic rates observed at elevation ___”

A. are consistent with the cold temperatures typical at that elevation, though the measurements alone do not rule out other contributing factors

B. caused the cold temperatures recorded at that elevation

C. are higher than metabolic rates recorded at any other elevation studied

D. make this species the most cold-adapted animal ever measured

A keeps the direction the passage actually establishes, cold raising metabolic rate, without overclaiming that this single measurement rules out every other explanation. B reverses cause and result, having the metabolic rate produce the cold rather than the reverse. C invents a comparison to elevations the passage never mentions measuring. D invents a ranking against every animal ever measured, far beyond this one field study.

A choice claiming one theory is correct has to be checked against whether the evidence merely weakens a rival or is simply compatible with the favored theory; and a choice has to respect however, yet, or despite sitting near the blank, since that contrast word usually changes what can logically follow it; and a choice that repeats an early finding has to be checked against whether a later sentence limits it.

Researchers long believed a particular fish species migrated south for warmer water. Tagged fish tracked over three years, however, stayed within the same ten-mile stretch of river year-round.

Which choice most logically completes the text: “The tracking data ___”

A. directly conflict with the claim that this population migrates south for warmer water during the observed years

B. confirm that this species migrates for warmer water, just over a shorter distance

C. prove no fish species ever migrates for temperature reasons

D. show the fish moved south by exactly ten miles each winter

A honors the contrast however introduces: three years of tagged fish staying in one short stretch directly weighs against the earlier migration belief for this population during these years, not toward confirming it. B keeps the old belief alive in weakened form when the data actually conflict with it outright. C generalizes from one species to every fish species that migrates for any reason. D invents a specific seasonal movement, exactly ten miles south each winter, that “stayed within the same stretch year-round” directly rules out.

When a passage is full of names, reduce them mentally to A, B, group 1, and group 2; when a passage is full of technical vocabulary, translate only the specific relationship the conclusion actually needs; and when two choices both seem to fit the general topic, identify the one exact, unresolved claim the blank has to complete.

A researcher’s study of Species A found elevated cortisol in specimens exposed to Compound X. A second researcher’s replication used a related species, Species B, and found no such elevation.

Which choice most logically completes the text: “The two studies together suggest that the cortisol effect ___”

A. may be specific to the original species tested rather than the compound generally, given the differing result in a related species

B. proves Compound X has no biological effect on any species

C. shows the second researcher’s methodology was flawed

D. confirms the first researcher’s original finding beyond doubt

Stripped of names, this reads as: Study 1 on Species A found effect X; Study 2 on a related Species B did not. A captures exactly that relationship. B overreaches into “no species,” when only two related species were tested at all. C invents a methodological flaw neither study’s description supports. D ignores the second study entirely, when the question specifically asks what the two studies together suggest.

One choice can need exactly one additional, unsupported fact to work, which is enough to eliminate it; two choices can both require real reasoning from the passage, which is perfectly fine; and a narrower-sounding answer is wrong exactly when it leaves out a conclusion the facts clearly support, not because narrow answers are inherently safer.

A bakery’s bread sales fell 20 percent the month a nearby competitor opened. The bakery’s own price per loaf had also risen 15 percent that same month.

Which choice most logically completes the text: “The drop in sales ___”

A. cannot be attributed solely to the new competitor, given the price increase in the same month

B. resulted entirely from the new competitor opening nearby

C. would have happened even without the price increase

D. shows customers strongly prefer the competitor’s bread

A needs no fact beyond what the passage already gives: two changes, one outcome, and no way to isolate either one’s individual contribution. B needs one additional, unsupported fact, that the competitor was the sole cause, which the same-month price increase directly complicates. C needs an unsupported counterfactual about what would have happened under different conditions the passage never tested. D needs an unsupported preference judgment neither figure can establish on its own.

A broader-sounding answer is not automatically wrong; it is wrong only when it extends past what the evidence actually supports, and it can be exactly right when it preserves every condition the passage actually stated rather than dropping one to sound more modest.

A new tutoring program improved test scores among students who attended at least eight sessions. Students who attended fewer than eight sessions showed no measurable improvement.

Which choice most logically completes the text: “The observed score improvement ___”

A. occurred only among students who attended at least eight sessions, and did not extend to students who attended fewer

B. applied only to a small number of exceptional students

C. was too small to matter for most students

D. occurred regardless of how many sessions a student attended

A is the broader, two-part answer, and it is correct because it preserves both conditions the passage actually states: improvement above the threshold, no improvement below it; broad is not the problem here, dropping either half would be. B invents a claim about “exceptional students” the passage never makes. C invents a judgment about size, “too small to matter,” that the passage never offers. D directly contradicts the stated difference between the two attendance groups.

When nothing seems to follow from a dense or unfamiliar passage, strip it down to two or three plain statements, especially the sentence immediately before the blank and any sentence it refers back to; and when the final choice still feels close, finish the sentence “for this answer to be true, I also have to assume ___” and prefer whichever choice needs the least filled in.

Isotopic analysis of tooth enamel from an excavated settlement shows an unusual ratio consistent with a diet dominated by marine resources, a pattern otherwise undocumented at inland sites of the same period. The settlement lies eighty miles from the nearest coastline.

Which choice most logically completes the text: “The isotopic pattern ___”

A. suggests these residents obtained marine resources despite the settlement’s inland location

B. proves the settlement engaged in a coastal trade network

C. shows the settlement was originally located on the coast and later moved inland

D. confirms these residents traveled to the coast personally to gather food

Stripped to plain statements: enamel shows a marine-heavy diet signature; the site is eighty miles inland; this pattern is not seen at other inland sites of the period. A states only what those three facts require, that marine resources reached these residents somehow, without specifying how. For B, finish the test sentence: “for this to be true, I also have to assume the resources arrived through organized trade specifically.” The passage supports resources arriving somehow, not trade specifically. C requires assuming an entire relocation history nothing in the passage suggests. D requires assuming personal travel to the coast, one specific mechanism among several the isotopic evidence alone cannot distinguish. A needs the least filled in, which is exactly why it is the answer.