Original public samples
Read the reasoning.
These original questions were prepared for Logic Refined. They are not official LSAT questions.
Question 1
A bookseller has two formulas for forecasting sales of new titles. The older formula uses a book’s genre and advance orders. The new formula also uses details of a book’s cover and description. Its designer chose the details and their weights after examining the recorded sales of every title released last spring, adjusting the formula to make its forecasts fit those records. An independent clerk then checked the arithmetic and confirmed that, on those same titles, the new formula had a lower average forecasting error than the old one. The designer argues: “The clerk’s independence answers the worry that I favor my own work. We have now demonstrated that, for next season’s new titles, the new formula will have a lower average forecasting error than the old one.”
The designer’s reasoning is most vulnerable to criticism on the grounds that it
- takes an association between features of books and their sales as establishing that those features cause people to buy the books.
The designer claims forecasting accuracy, not that covers or descriptions cause sales. Associations can help prediction without representing causal influences. The problem concerns how accuracy was tested, not a causal interpretation of the selected features.
- assumes that adding information to a forecasting method necessarily improves the accuracy of the forecasts it produces.
The designer offers measured performance on last spring’s titles, not simply the presence of more information. That performance is inadequate as an independent test, but the argument does not claim that every additional detail must improve a forecast.
- treats independent checking of a calculation as making the observations used in that calculation an independent test of the formula.
The clerk independently verifies the error calculation, but the formula was adjusted using the very sales being checked. Accurate arithmetic does not turn the records used to fit the formula into a separate test of its forecasts for new titles.
- rejects a simpler forecasting method without establishing that complicated methods generally outperform simpler ones.
A particular method could outperform another without complicated methods being better in general. A general rule about complexity is unnecessary. What is missing is a test on titles that were not used to build the formula.
- infers that no particular forecast will be badly mistaken from a reduction in the average error of a group of forecasts.
The comparison concerns average forecasting error, and the designer’s conclusion claims only a lower average error. It does not promise a small error for every title. Large individual mistakes can coexist with a lower average, and the designer never says otherwise.
Correct answer: C. The clerk independently verifies the error calculation, but the formula was adjusted using the very sales being checked. Accurate arithmetic does not turn the records used to fit the formula into a separate test of its forecasts for new titles.
Question 2
Viewers were randomly assigned to read the same accurate synopsis either just before or just after watching an ambiguous scene from a film. The synopsis explained a motive that the film revealed only near its end. Before the second group read the synopsis, all viewers rated the protagonist; those who had already read it rated the protagonist as more calculating. Once everyone had read the synopsis, everyone saw the complete film. Afterward, the two groups rated the protagonist equally calculating. Nevertheless, when the scene was shown again, the viewers who had read the synopsis first said that the scene itself provided stronger evidence of calculation than the other viewers said it did. The groups were equally accurate in recalling the scene’s events.
Which one of the following is most strongly supported by the information above?
- Learning a character’s motive earlier can change the evidence viewers attribute to a scene without improving their recall of the scene’s events.
This matches the persistent evidence-attribution difference and the absence of a recall advantage. Random assignment strengthens the inference that earlier access to the motive can produce the difference.
- Viewers who learn a character’s motive before seeing a scene can mistake the explanation of that motive for an event shown in the scene.
The experiment does not show viewers remembering an explanatory statement as an event in the scene. Different evidential readings of correctly remembered events are sufficient to explain the result.
- The viewers who read the synopsis first were more confident in their eventual interpretation of the protagonist than were the other viewers.
The final rating measures how calculating the character seems, not confidence in that judgment. Equal or unequal confidence is compatible with the data.
- The complete film had a greater influence than the synopsis on the viewers’ eventual interpretation of the protagonist’s motives.
Both groups receive both sources before the final rating. Their convergence does not rank the influence of the film against that of the synopsis.
- Reading the synopsis before the scene gave viewers a more accurate understanding of the protagonist’s motives than reading it afterward did.
The study supplies no independent accuracy criterion for the viewers’ understanding of the character. An accurate synopsis can influence a rating without making that rating more accurate than a later reader’s.
Correct answer: A. This matches the persistent evidence-attribution difference and the absence of a recall advantage. Random assignment strengthens the inference that earlier access to the motive can produce the difference.
Question 3
A moving company now requires a crew member to repeat a customer's requested room aloud before a box is carried there. Wrong-room placements have declined. The operations manager argues that the repeated information helps expose misunderstandings and contributes to the decline. The foreman replies that customers sometimes change their minds during these exchanges, so a corrected assignment need not reflect an earlier misunderstanding. Moreover, the new procedure forces crews to pause; slower work, he argues, accounts for the entire decline without any benefit from repeating the information.
Which one of the following, if true, most strengthens the operations manager's argument?
- Even before the procedure was introduced, some customers' spontaneous repetitions of instructions were followed by crew members changing a box's assigned destination.
This resembles the proposed mechanism, but the original intended destination and the effect of the intervening delay are not established. The changed assignments could reflect changed requests or reflection during the interruption.
- In some matched trials with equal pauses, a recorded mismatch between the initial request and the crew's assignment was corrected after readback but survived a silent pause.
These cases preserve the requested destination and compare equal delays. The contrast connects repeated content with correction of a real misunderstanding while the pause-only account predicts no such advantage from the content. A limited but diagnostic observation materially supports a contribution without establishing its overall size.
- Nearly all wrong-room placements that still occur under the new procedure involve a readback that the customer either fails to hear or fails to answer.
The pattern suggests that effective engagement matters, but unheard or unanswered readbacks may also mark hurried work. It does not isolate repeated content from pace, and it conditions on the errors that remain.
- Among crews handling requests of comparable complexity, those that spend the longest time on readbacks tend to make the fewest wrong-room placements.
This association is consistent with a benefit from repeated information, but it is also what the foreman's slower-work explanation predicts. It leaves the competing causes entangled.
- On most moves under the new procedure, customers confirm the crew's intended destination before a box is carried, whereas confirmations previously occurred after the carrying had begun.
Earlier confirmation is compatible with the manager's mechanism, but it does not show that the spoken content, rather than the added pause, prevented errors. Nor does it distinguish corrected misunderstandings from revised customer preferences.
Correct answer: B. These cases preserve the requested destination and compare equal delays. The contrast connects repeated content with correction of a real misunderstanding while the pause-only account predicts no such advantage from the content. A limited but diagnostic observation materially supports a contribution without establishing its overall size.
Question 4
A piano maker tested new hammers by asking listeners to identify familiar melodies from short recorded performances played through background noise. Listeners identified more melodies in the new-hammer recordings than in the old-hammer recordings. The maker attributes the improvement to listeners being better able to hear the pitch relationships between successive notes. Musicians also reported that the new hammers made rapid passages less tiring to play.
Which one of the following, if true, most weakens the argument?
- In separate single-note trials, listeners were less accurate at naming a note’s pitch under the new hammers than under the old ones.
This gives a genuine reason to question a broad claim that the new hammers improve every kind of pitch perception. But naming an isolated pitch is not the same task as distinguishing relationships across successive notes. That difficulty can coexist with clearer relative patterns; the click finding reaches the actual melody-recognition advantage more directly.
- In a separate test, listeners could identify several of the familiar melodies from their rhythms without hearing their pitches.
This establishes that a nonpitch route to identifying some melodies is available. It does not show that the new-hammer advantage actually depends on that route. The click experiment goes further: it preserves the comparative advantage when pitch information is removed.
- Listeners also identified more melodies from the new-hammer recordings when those recordings were played without the background noise.
This extends the advantage to another listening condition. The maker need not claim that clearer pitch relationships help only in noise. Both noisy and quiet listening could benefit, so removing noise is not the same as removing the proposed pitch information.
- The musicians who found rapid passages less tiring under the new hammers did not identify more melodies than other listeners did.
This disconnects a reported playing benefit from listeners’ recognition scores. The maker attributes the recognition gain to the recorded pitch relationships, not to the recognizing listeners’ physical fatigue. The new hammers could affect playing comfort and audible information through different routes.
- The recognition advantage persisted after each recorded note was replaced by an unpitched click preserving its original timing and loudness.
The altered recordings retain timing and loudness patterns but not the pitch relationships the maker invokes. An advantage that persists there has a demonstrated route through nonpitch information. This directly weakens the attribution of the recognition advantage to clearer pitch relationships, without proving that pitch contributes nothing in the original recordings.
Correct answer: E. The altered recordings retain timing and loudness patterns but not the pitch relationships the maker invokes. An advantage that persists there has a demonstrated route through nonpitch information. This directly weakens the attribution of the recognition advantage to clearer pitch relationships, without proving that pitch contributes nothing in the original recordings.
Question 5
Editor: Our Dutch edition of an English card-game rulebook needs checking, but our Dutch-speaking expert can inspect only a few passages. Different wordings can prescribe the same moves, so checking for matching words would be unhelpful. We will have a translator who has not seen the English original render the Dutch text back into English. Our editors can recognize any difference in rules between a returned passage and the original. They will list every Dutch passage whose return translation prescribes different rules, and no others. A colleague objects that a second translation can introduce its own mistakes. Nevertheless, a passage selected at random from the list will be more likely to misstate the original rules than one selected at random from the entire Dutch text.
Which one of the following is an assumption on which the editor’s argument depends?
- The return translator would preserve the Dutch meaning of at least one passage that misstated the English rules.
A faulty Dutch rule could become a different faulty rule on its return to English. That passage would still enter the review list. Faithfully preserving the Dutch mistake is therefore not required for the screening to help.
- A Dutch passage that states the original rules correctly would not acquire different rules in the return translation.
This rules out false alarms, but the editor predicts an improved chance of finding an error, not a flawless list. Some correct Dutch passages could be listed while the list still contains a higher proportion of faulty ones.
- Not every Dutch passage that misstates the rules would be translated back into a version matching the original rules.
If every faulty Dutch passage returned with the original rules restored, none would enter the list. The list could then contain only correct Dutch passages, or be empty, defeating the claimed improvement in finding actual mistakes.
- The return translation would depart from the Dutch rules in fewer passages than the Dutch translation departs from the English rules.
The second translator need not make fewer departures from the source. A new mistake can replace an earlier mistake while leaving a genuinely faulty Dutch passage detectable. Even if the return translation creates more false alarms than the Dutch text has errors, the list can still hold a higher share of faulty passages than the whole text does.
- Any returned passage that matches the original rules would have come from a Dutch passage that also states them correctly.
This would rule out every hidden Dutch error, but the proposed screening need not catch them all. One mistaken passage could return to the original rules and escape detection while other mistakes make the list more useful.
Correct answer: C. If every faulty Dutch passage returned with the original rules restored, none would enter the list. The list could then contain only correct Dutch passages, or be empty, defeating the claimed improvement in finding actual mistakes.
Question 6
After an agency introduced an online portal, it opened fewer new case files during the year than it had the year before, and its average time to a first substantive reply did not increase. Yet it ended the year with more open files than before. Follow-up messages on an existing case were not counted as new files. The portal made it easier for applicants to send such messages.
Which one of the following, if true, most helps to explain the apparent discrepancy?
- Under an unchanged rule, a file closed 90 days after its first substantive reply; portal users often sent follow-ups before that interval expired.
The described rule fixes closure to the first reply rather than the last applicant message. More follow-ups do not reset that clock, so this does not explain slower exits or a larger stock through the proposed mechanism.
- Under an unchanged rule, a file closed only after 90 days without an applicant message; portal users often sent follow-ups before that interval expired.
The stable rule resets the closing clock after each follow-up. Easier follow-ups can therefore keep files open longer, raising the pending stock despite fewer new files and unchanged speed of initial substantive replies.
- The portal sent immediate automatic acknowledgments, but the agency continued to count only substantive replies when calculating its reply-time average.
This explains why automatic speed did not lower the measured reply time. It does not explain why more files remained open despite reduced inflow.
- The portal discouraged applicants from sending follow-ups after they had received satisfactory substantive replies.
Fewer follow-ups could reduce continuing work, but no link to longer open-file duration is supplied. This does not account for the higher year-end stock.
- The portal made it easier to see which files were open, but the agency used the same definition of an open file in both years.
Greater visibility may change perceptions of a backlog, but the stated count is higher under an unchanged definition. Improved visibility alone does not establish a real increase in pending duration or files.
Correct answer: B. The stable rule resets the closing clock after each follow-up. Easier follow-ups can therefore keep files open longer, raising the pending stock despite fewer new files and unchanged speed of initial substantive replies.