Adult Literacy Advanced Quiz: Using Multiple Sources
10 questions · exam conditions
0:00
Using Multiple SourcesQuestion 1 of 10

A distribution center tested a new system that allowed employees to select open evening shifts through an online portal.

Source 1 — Staffing records: During the eight weeks before the test, 12 percent of evening shifts remained unfilled 48 hours before they began. During the eight-week test, 5 percent remained unfilled at that point. The holiday shipping season began during the test.

Source 2 — Anonymous employee comments: Several employees wrote that choosing shifts through the portal was more convenient. Others said popular shifts disappeared too quickly.

Source 3 — Operations manager's memo: "The portal clearly caused the staffing improvement and will eliminate scheduling problems if adopted permanently."

Which note most accurately combines the sources without presenting an opinion or an uncertain cause as an established fact?

The portal reduced unfilled shifts from 12 percent to 5 percent, proving that employee choice eliminates scheduling problems even during unusually busy periods.
Unfilled shifts were less common during the test, although the records do not isolate the portal from seasonal changes; employees expressed mixed opinions about convenience.
Employees preferred the portal because staffing records improved, but the manager believed that popular evening shifts continued to disappear too quickly.
The holiday shipping season caused employees to use the portal more often, while anonymous comments established that the system was convenient overall.
← Back to quizzes

Adult Literacy Advanced Quiz

Adult Literacy Advanced Quiz: Using Multiple Sources

Practice Using Multiple Sources in Adult Literacy Advanced with focused quiz questions that help you check what you know, review explanations, and build confidence with test-style prompts.

What this quiz covers

This quiz focuses on Using Multiple Sources, giving you a quick way to practice the rules, question types, and explanations that matter most for Adult Literacy Advanced.

How to use this quiz

Try each quiz question before looking at the correct answer. Use the explanations to review missed ideas, then come back to similar questions until the pattern feels familiar.

All questions

Question 1

A distribution center tested a new system that allowed employees to select open evening shifts through an online portal.

Source 1 — Staffing records: During the eight weeks before the test, 12 percent of evening shifts remained unfilled 48 hours before they began. During the eight-week test, 5 percent remained unfilled at that point. The holiday shipping season began during the test.

Source 2 — Anonymous employee comments: Several employees wrote that choosing shifts through the portal was more convenient. Others said popular shifts disappeared too quickly.

Source 3 — Operations manager's memo: "The portal clearly caused the staffing improvement and will eliminate scheduling problems if adopted permanently."

Which note most accurately combines the sources without presenting an opinion or an uncertain cause as an established fact?

  1. The portal reduced unfilled shifts from 12 percent to 5 percent, proving that employee choice eliminates scheduling problems even during unusually busy periods.
  2. Unfilled shifts were less common during the test, although the records do not isolate the portal from seasonal changes; employees expressed mixed opinions about convenience. (correct answer)
  3. Employees preferred the portal because staffing records improved, but the manager believed that popular evening shifts continued to disappear too quickly.
  4. The holiday shipping season caused employees to use the portal more often, while anonymous comments established that the system was convenient overall.
Explanation: When combining multiple sources, your job is to report only what the evidence actually shows — not to assign causes, declare winners, or smooth over contradictions. Ask yourself: Does this claim go beyond what the data can prove? Answer B earns its accuracy by honoring the limits of each source. It acknowledges the staffing improvement (unfilled shifts dropped from 12% to 5%) without crediting the portal as the definite cause — because the holiday season was also underway, making it impossible to isolate which factor drove the change. It also captures the genuine split in employee opinion rather than flattening it into a single verdict. Nothing in B is stated with more confidence than the sources allow. Answer A commits the clearest error by calling the drop in unfilled shifts proof that employee choice eliminates scheduling problems. That conclusion overstates the data significantly — no source proves causation, and the manager's claim in Source 3 is an opinion, not an established fact. "Even during unusually busy periods" also smuggles in a favorable interpretation. Answer C scrambles the sources badly. It claims employees preferred the portal because staffing improved — reversing the logical relationship — and incorrectly attributes the "shifts disappearing quickly" complaint to the manager rather than to the employees who actually said it. Answer D invents a causal claim that no source makes: that the holiday season caused employees to use the portal more often. It also overstates employee opinion by saying comments "established" the system was convenient overall, when reactions were mixed. Strategy tip: When a question asks you to combine sources neutrally, flag any word that implies certainty — proved, caused, established, clearly — as a red-flag signal that an answer is overreaching.

Question 2

A resident is researching the town's recycling performance in 2024.

Source 1 — Sanitation newsletter: Collection crews picked up 820 tons of material placed in recycling containers in 2024, compared with 760 tons in 2023. The newsletter calls the increase "an impressive community success."

Source 2 — Contractor's annual audit: Of the 820 tons collected in 2024, 123 tons were rejected as contamination. The remaining 697 tons were accepted and marketed to processing facilities.

Source 3 — Council meeting statement: A council member argued that residents need clearer instructions because too much nonrecyclable material is still being placed in containers.

Which note most accurately answers "How much collected material was actually sent for processing in 2024?" and separates that answer from judgment?

  1. Fact: 820 tons were sent for processing. Opinion: Rejecting 123 tons made the recycling program an impressive success.
  2. Fact: 697 tons were accepted for processing. Opinion: The collection increase was impressive, and residents need clearer instructions. (correct answer)
  3. Fact: 760 tons were accepted for processing. Opinion: The contractor rejected too much material as contamination in 2024.
  4. Fact: 123 tons were sent to processing facilities. Opinion: All remaining material should have been placed in ordinary trash.
Explanation: When a question asks you to separate fact from opinion, your job is two-fold: identify what the sources actually report as data, and strip away any evaluative language (words like "impressive," "too much," or "should"). Keep those two categories clean. Here, the question asks specifically how much collected material was sent for processing. Source 2 (the contractor's audit) gives you the clearest answer: of 820 tons collected, 123 tons were rejected as contamination, leaving 697 tons accepted and marketed to processing facilities. That 697 figure is a verifiable number from a third-party audit — it's a fact. The phrases "impressive success" (from the newsletter) and "residents need clearer instructions" (from the council member) are both judgments or recommendations, not data — they belong in the opinion column. Choice B handles this correctly: it states the accurate processing figure as fact and groups the evaluative language as opinion. Choice A is wrong because it labels 820 tons as sent for processing — but 820 was the collected total, not the processed total. It also misidentifies the opinion portion entirely. Choice C introduces 760 tons, which was the 2023 collection figure, not a 2024 processing figure — it confuses years and categories. Choice D claims 123 tons went to processing facilities, but those 123 tons were the rejected material, the exact opposite of what was processed. A reliable strategy: when sources give you a chain of numbers (collected → rejected → accepted), map each number to its specific meaning before answering. The question's wording is your guide — match the right number to the right step in the chain.

Question 3

A factory introduced a safety-training program and is reviewing injury data.

Source 1 — Safety logs: In the year before training, employees recorded 14 reportable injuries during 200,000 hours worked. In the year after training, they recorded 8 injuries during 160,000 hours worked.

Source 2 — Consultant's report: "The training caused a major safety improvement because the number of injuries fell by six."

Source 3 — Employee committee note: Workers generally supported the training, but the committee observed that production volume and total hours also declined during the second year.

Which conclusion best uses the sources and accounts for the difference in hours worked?

  1. The injury rate fell from 7 to 5 injuries per 100,000 hours, but the sources do not prove that training alone caused the decline. (correct answer)
  2. The injury rate fell from 14 to 8 injuries per 100,000 hours, proving that employee support made the training effective.
  3. The injury rate remained unchanged because both injuries and hours worked decreased during the year after training.
  4. The injury rate increased from 5 to 7 injuries per 100,000 hours, although the total number of injuries declined.
Explanation: When a question gives you raw counts alongside different time periods or workloads, your first move should be to convert to a rate so you're comparing apples to apples. Raw numbers alone can mislead. The standard workplace injury rate is calculated per 100,000 hours worked: Rate=injurieshours worked×100,000\text{Rate} = \frac{\text{injuries}}{\text{hours worked}} \times 100{,}000 Before training: 14200,000×100,000=7 per 100,000 hours\frac{14}{200{,}000} \times 100{,}000 = 7 \text{ per 100,000 hours} After training: 8160,000×100,000=5 per 100,000 hours\frac{8}{160{,}000} \times 100{,}000 = 5 \text{ per 100,000 hours} So the rate did fall — from 7 to 5. But Source 3 flags that total hours and production volume both dropped, meaning other factors changed alongside the training. You cannot isolate training as the sole cause. That's exactly what answer A captures: a real rate improvement that cannot be definitively attributed to training alone. A is correct. B is wrong on two counts: the rates it states (14 and 8) are the raw injury counts, not rates per 100,000 hours. It also claims employee support proved effectiveness — a logical leap no source supports. C is wrong because it claims the rate was unchanged. The rate clearly dropped from 7 to 5; this answer confuses "both numbers fell" with "the ratio stayed the same." D reverses the direction entirely, stating the rate increased from 5 to 7. That gets the before/after values backwards. Strategy tip: Whenever you see data with different totals across time periods, always ask "rate or raw count?" Raw counts without context are almost never the full story on this exam.

Question 4

A community college is considering a one-year trial of evening certification courses.

Source 1 — Registration office: Last year, 74 applicants who met admission requirements declined enrollment. In a follow-up questionnaire completed by 41 of them, 25 selected "course times conflict with work" as one reason.

Source 2 — Regional employer survey: Nine of 14 responding employers said they would consider sharing information about evening courses with workers. The college contacted 60 employers in total.

Source 3 — Faculty senate statement: "An evening option is worth testing, but current evidence does not guarantee enough enrollment to sustain a permanent program."

Which recommendation is best supported by all three sources?

  1. Establish a permanent evening program because most qualified applicants and most regional employers have already committed to participating.
  2. Reject evening courses because fewer than half of the employers contacted promised to pay their workers' tuition.
  3. Conduct a limited evening-course trial while tracking enrollment, since some applicants reported scheduling conflicts but broader demand remains uncertain. (correct answer)
  4. Move all certification courses to evenings because work conflicts were the only reason qualified applicants declined enrollment.
Explanation: When a question asks which recommendation is "best supported by all three sources," your job is to find the option that accurately reflects every source — not just one or two. Think of it as a logic check: if any source contradicts or fails to support a choice, that choice fails. Here, the three sources paint a consistent but cautious picture. Source 1 shows some demand — 25 out of 41 survey respondents cited scheduling conflicts — but that's a partial sample of a partial group, leaving real uncertainty about total demand. Source 2 shows possible employer interest (9 of 14 responding employers said they'd consider sharing information), but only 14 of 60 employers even responded, so actual commitment is unknown. Source 3 explicitly calls for testing the idea while acknowledging the evidence is insufficient to guarantee a permanent program. Together, all three sources justify a cautious trial with continued monitoring — which is exactly what C recommends. A fails because neither applicants nor employers have "committed" to anything. Declining to enroll doesn't equal commitment, and "would consider sharing information" is far weaker than a promise to participate. B distorts Source 2 entirely — the survey never mentioned tuition reimbursement, and employers not responding is not the same as refusing to pay tuition. D misreads Source 1; the questionnaire listed scheduling conflicts as one reason among others, and only 25 of 41 partial respondents selected it — far from proof it was the only reason. A useful strategy: watch for answer choices that use absolute language like "only," "all," or "committed." Real-world data almost never supports such strong conclusions, and those words are usually traps.

Question 5

A library board is reviewing a three-month extension of Saturday hours.

Source 1 — Library records: Saturday visits averaged 430 per month during the three-month trial, compared with 350 per month during the same months a year earlier. A new online-renewal help desk also opened on Saturdays during the trial.

Source 2 — Exit survey: Of 180 visitors invited to participate, 96 responded. Of those respondents, 71 said the later closing time made the library easier for them to use.

Source 3 — Friends of the Library statement: "The trial was unquestionably responsible for the increased visits and should continue without further review."

Which summary best integrates the sources while preserving important limitations?

  1. Saturday visits increased, and most invited visitors credited the later closing time; therefore, the extended hours unquestionably caused the increase.
  2. Saturday visits increased during the trial, and most survey respondents valued the later closing; however, the help desk and incomplete survey response limit causal claims. (correct answer)
  3. The trial did not affect library use because fewer than 100 visitors completed the survey and the help desk attracted all additional visitors.
  4. The later closing was easier for every Saturday visitor, although library records show that average monthly attendance remained nearly unchanged.
Explanation: When a question asks you to integrate multiple sources, your job is to combine what each source actually shows — without overstating the evidence or dismissing it entirely. Pay close attention to which sources support strong conclusions and which ones introduce uncertainty. The library records confirm a real increase in Saturday visits (430 vs. 350 per month), so something changed during the trial. The exit survey shows that 71 out of 96 respondents valued the later closing — that's meaningful, but only 96 of 180 invited visitors responded, meaning more than half didn't participate. Non-respondents might feel differently. On top of that, the new help desk also launched on Saturdays during the trial, so you can't be certain whether the extended hours, the help desk, or some combination drove the increase. Answer B accurately reflects all of this: visits rose, most respondents valued the later hours, but the incomplete survey response rate and the help desk as a competing explanation both limit how confidently you can assign cause. Answer A echoes the Friends of the Library's overconfident language ("unquestionably caused"), ignoring the help desk and survey limitations entirely. Answer C swings too far the other way — it dismisses any effect of the trial, which contradicts the real visit increase the records document; no source claims the help desk was solely responsible. Answer D is factually wrong on two counts: not every visitor was surveyed, and attendance did not remain "nearly unchanged" — it rose by 80 visits per month. A useful strategy: when sources conflict or overlap, look for the answer that acknowledges both the evidence and its limits — that balance is almost always what "integrating sources" means on this exam.

Question 6

A health educator is taking notes about a workplace walking program.

Source 1 — Program log: Among 86 employees who enrolled, 61 recorded steps during both the first and final weeks. For those 61 employees, the median daily step count increased from 5,900 to 7,100.

Source 2 — Participant interview: "The lunchtime walks made the workday much more enjoyable, and I am certain they improved everyone's health."

Source 3 — Benefits office notice: The company predicts that continued participation could reduce future health-insurance claims, although no claims data have yet been analyzed.

Which labeling of notes correctly distinguishes reported facts from opinions or predictions?

  1. Fact: All 86 participants increased their daily steps. Opinion: The median rose by 1,200 steps among employees with complete records.
  2. Fact: The program improved everyone's health and made lunchtime more enjoyable. Opinion: Future insurance claims will decrease if employees continue participating.
  3. Fact: Sixty-one participants had first- and final-week step records. Opinion/Prediction: Lunchtime walks were more enjoyable, and future claims may decline with continued participation. (correct answer)
  4. Fact: Insurance claims have already declined since the program began. Opinion: Only employees with complete step records benefited from the walking program.
Explanation: When reading informational sources, your job is to separate what was directly observed or measured (facts) from what someone believes, feels, or projects into the future (opinions and predictions). Ask yourself: "Could this be verified with data, or does it depend on someone's judgment?" Source 1 gives you measurable data: 61 out of 86 enrolled employees had complete records, and among those 61, the median daily step count rose from 5,900 to 7,100. These are documented, verifiable numbers. Source 2 is a personal interview — the participant uses phrases like "I am certain" and "more enjoyable," which signal subjective belief, not measured outcomes. Source 3 explicitly uses the word "predicts" and confirms "no claims data have yet been analyzed," making it clearly a forward-looking prediction. Answer C correctly labels the 61-participant data point as a fact (it comes directly from the program log) and correctly groups the enjoyment claim and the insurance prediction together as opinion/prediction — because one is a personal feeling and the other is an unverified forecast. Answer A gets things backwards, misidentifying the measurable median increase as an opinion and falsely claiming all 86 participants increased their steps — the log only tracked 61. Answer B treats a subjective interview quote about "everyone's health" as a fact, which it is not; personal certainty is not evidence. Answer D invents a fact that doesn't exist in any source — insurance claims have not already declined — and then mislabels an unsupported inference as an opinion. A useful habit: whenever a source includes words like predict, certain, enjoyable, or could, flag that statement as opinion or prediction, not fact.

Question 7

A researcher is evaluating a six-month bus-lane pilot.

Source 1 — City transportation report: GPS records show that the average end-to-end weekday trip on Route 8 decreased from 51 minutes before the pilot to 44 minutes during the pilot. On-time arrivals increased from 68 percent to 79 percent. The report does not include weekend trips.

Source 2 — Rider survey: Of 640 riders who voluntarily completed an online survey, 58 percent said service felt more predictable, while 27 percent said their total commute became shorter.

Source 3 — Newspaper editorial: "The dedicated lane has transformed travel for everyone who depends on Route 8 and should immediately become permanent."

Which set of notes best uses the sources to answer the question "Did the pilot improve travel on Route 8?" while clearly separating facts from opinions?

  1. Fact: Weekday trips averaged seven minutes less, and on-time arrivals rose. Opinion: Survey participants generally believed the route was more predictable, and the editorial supports permanence. (correct answer)
  2. Fact: Travel improved for every Route 8 passenger throughout the week. Opinion: The seven-minute reduction probably occurred because riders supported making the lane permanent.
  3. Fact: Most Route 8 riders experienced shorter commutes during the pilot. Opinion: The city's GPS records suggest that buses may have become more predictable.
  4. Fact: The pilot permanently transformed Route 8 service. Opinion: Weekend travel likely improved by approximately the same amount as weekday travel.
Explanation: When a question asks you to separate facts from opinions, your job is to check every claim against its source: does the source actually measure it, or does someone simply believe it? Facts come from objective data (GPS records, percentages); opinions come from personal judgments or interpretations (surveys, editorials). Answer A handles this cleanly. The GPS data directly measured a seven-minute average reduction in weekday trip time and an 11-point jump in on-time arrivals — those are verifiable facts from recorded data. The survey results and editorial claim are correctly labeled as opinions: survey responses reflect how riders felt, not what was objectively measured, and the editorial explicitly expresses a value judgment about making the lane permanent. B is wrong on two counts. Calling improved travel a fact "for every Route 8 passenger throughout the week" overstates the data — the report excludes weekends and doesn't cover all riders. B also invents a causal link between rider support and the seven-minute reduction, which no source supports. C reverses the fact/opinion labels entirely. The GPS data is objective evidence, not opinion, and calling "most riders experienced shorter commutes" a fact misrepresents the survey — only 27 percent of voluntary respondents said their commute got shorter, a self-selected group that can't represent all riders. D is built almost entirely on fabrications. "Permanently transformed" comes from an editorial opinion, not a measured outcome, and the claim about weekend travel is invented — no source addresses weekend data at all. Your strategy: before labeling anything a "fact," ask yourself, "Was this measured or recorded?" If it comes from someone's feeling, editorial stance, or speculation, it's an opinion regardless of how confidently it's stated.

Question 8

A town expanded its seasonal farmers' market from one morning to two mornings per week.

Source 1 — Market administration records: The average number of vendors per week rose from 18 in the previous season to 27 after expansion. Total reported vendor sales rose 24 percent, while the number of market days doubled.

Source 2 — Resident survey: Of 300 randomly selected households, 162 responded. Among respondents, 59 percent said the additional market day was convenient; 21 percent said parking near the market became more difficult.

Source 3 — Vendor association statement: "The expansion has strengthened every participating business and benefited the entire neighborhood."

Which answer to "Did the expansion benefit vendors and residents?" is most appropriately supported by the combined sources?

  1. Yes. Sales rose 24 percent and most responding households found the schedule convenient, which proves that every vendor and every resident benefited from the expansion.
  2. No. Because market days doubled while sales rose only 24 percent, each vendor earned less per day, and the survey shows that residents opposed the parking situation.
  3. The evidence applies to residents only, because the vendor sales and participation figures come from the vendor association's statement rather than from the administrative records.
  4. The evidence is mixed: vendor participation and total sales rose, while surveyed residents reported both convenience and parking concerns; individual-level benefits for vendors or all residents are not established. (correct answer)
Explanation: When a question asks you to evaluate a claim using multiple sources, your job is to weigh what each source actually establishes — and notice when evidence is limited, mixed, or overstated. Here, the sources point in different directions. Administrative records show vendor count grew (18 to 27) and total sales rose 24%, which signals positive trends. The resident survey shows 59% of respondents found the extra day convenient, but 21% flagged parking difficulties. The vendor association calls the expansion a complete success for everyone. Taken together, the picture is genuinely mixed: some indicators improved, concerns exist, and broad claims about every vendor or all residents go beyond what the data can support. That makes D the strongest answer — it accurately reflects both the positive signals and the limitations without overstating either. A is tempting but commits a critical error: it leaps from group-level trends to individual-level guarantees. A 24% rise in total sales doesn't mean every vendor earned more, and 59% convenience among survey respondents doesn't mean every resident benefited. Watch for words like "every" and "proves" — they almost always overreach. B misreads the data twice. It assumes sales per day declined (possible, but not stated), and it exaggerates resident opposition — 21% reporting parking difficulty is not the same as residents opposing the expansion overall. C is factually wrong. The sales and vendor figures come from administrative records (Source 1), not the vendor association's statement (Source 3). Strategy tip: On source-evaluation questions, underline any absolute words ("every," "all," "proves"). If the evidence only supports a trend or majority, the correct answer will reflect that nuance rather than a sweeping conclusion.

Question 9

A neighborhood association is reviewing a tree-replacement ordinance adopted three years ago.

Source 1 — County environmental summary: Satellite estimates place tree-canopy coverage at 31 percent in the year the ordinance was adopted and 32 percent three years later. The estimates cover the entire county, including land outside the ordinance area, and severe storms occurred during the period.

Source 2 — Municipal permit records: Developers removed 310 regulated trees and documented 420 required replacement plantings. The records confirm planting but do not report how many replacement trees survived.

Source 3 — Business coalition letter: "The ordinance imposes an unreasonable expense while producing almost no environmental benefit."

Which conclusion best answers whether the ordinance increased the number of living trees in the regulated area?

  1. It did, because permit records show 110 more trees planted than removed, and the county's canopy estimate rose by one percentage point.
  2. It did not, because the small countywide canopy increase proves that nearly all 420 replacement trees failed to survive.
  3. The available sources do not establish the result: plantings exceeded removals, but survival was unreported and countywide canopy data do not isolate the regulated area. (correct answer)
  4. The ordinance produced almost no benefit because the business coalition evaluated its expense and the county experienced severe storms.
Explanation: When evaluating whether a policy achieved its goal, you must ask: do the sources actually measure the outcome, or only measure inputs and proxies? Here, the outcome you need is the net change in living trees inside the regulated area specifically. The permit records (Source 2) confirm that 420 trees were planted and 310 were removed — a net gain of 110 plantings on paper. But the records explicitly do not track survival. A planted tree is not the same as a living tree. Source 1 shows a 1-point countywide canopy increase, but it covers land outside the ordinance area and spans a period with severe storms, so it cannot isolate what happened inside the regulated zone. Because survival data are missing and the canopy data are too broad, no source closes the logical gap between "trees were planted" and "trees are alive and thriving." That gap makes C the only defensible conclusion. Answer A is tempting but overreaches — it treats unconfirmed plantings and an unrelated countywide metric as proof of success. Both pieces of evidence have documented limitations that A ignores. Answer B makes the opposite error: it treats a small canopy increase as proof of mass failure, which is a logical leap. The canopy data cannot prove that; many explanations exist for a modest countywide number. Answer D leans entirely on the business coalition's opinion (Source 3), which is advocacy, not evidence — and blaming storms doesn't answer whether the ordinance worked. The strategy to remember: gap-check every conclusion. Ask yourself whether the evidence actually measures what the conclusion claims, or whether a critical piece — here, survival data — is simply missing.

Question 10

Residents of an apartment building want to know whether recently installed window seals reduced heating use.

Source 1 — Building invoices: Weather-adjusted natural-gas use for November through February was 11 percent lower after installation than during the same four months one year earlier. Apartment occupancy was nearly unchanged.

Source 2 — Contractor's proposal, written before installation: The contractor estimated that the seals would reduce heating use by 18 percent if windows were the building's main source of heat loss.

Source 3 — Tenant message: "The hallways still feel cold, so the seals were a waste of money."

Which research note most accurately distinguishes observed evidence from estimates and personal judgment?

  1. Observed fact: Weather-adjusted gas use fell 11 percent. Estimate: Use might fall 18 percent under a stated condition. Opinion: The seals were a waste. (correct answer)
  2. Observed fact: The seals reduced gas use by exactly 18 percent. Estimate: Hallways may feel cold. Opinion: Occupancy was nearly unchanged.
  3. Observed fact: Cold hallways prevented any energy savings. Estimate: Gas use fell 11 percent. Opinion: Windows were the main source of heat loss.
  4. Observed fact: The seals alone caused the 11 percent decline. Estimate: Occupancy remained stable. Opinion: The invoices were adjusted for weather.
Explanation: When a question asks you to sort information by type — fact, estimate, or opinion — your job is to trace each claim back to its source and ask: Was this measured? Predicted? Or felt? Observed facts come from real measurements or records. Estimates are projections based on conditions or assumptions. Opinions are personal judgments that can't be verified objectively. Answer A handles all three categories correctly. The 11 percent gas reduction comes from actual building invoices adjusted for weather — that's a recorded, measurable outcome, making it an observed fact. The 18 percent figure came from the contractor's proposal, written before installation and only valid "if windows were the building's main source of heat loss" — a conditional prediction, which is exactly what an estimate is. The tenant's claim that the seals were "a waste of money" is a value judgment based on how the hallways feel, not on data — that's an opinion. Answer B misassigns everything. The 18 percent figure was never observed; it was a pre-installation estimate. Cold hallways are what the tenant experienced, not a prediction. And occupancy data came from building records, making it a fact, not an opinion. Answer C makes the most serious error: it treats the tenant's feelings as a proven fact ("cold hallways prevented any energy savings") and demotes the actual invoice data to an estimate. Feelings don't override measurements. Answer D falsely upgrades the 11 percent drop into a cause-and-effect proven fact — the invoices show correlation, not confirmed causation. Occupancy and weather adjustments are documented facts, not estimates or opinions. Your tip: when sorting evidence types, ask yourself, "Was this recorded, predicted, or felt?" That question will cut through most of these traps quickly.