Evaluating conclusions with evidence

Compare the claim with the data. The claim is that "Brand C absorbs the most water of the brands we tested." The Mean (mL) column shows A = 23, B = 30 and C = 35.

Step by step:

1. Brand C has the highest mean (35 mL), higher than B (30 mL) and A (23 mL).
2. Check the trials too: Brand C's lowest trial (34 mL) is still higher than Brand B's highest trial (31 mL). So C absorbed the most in every trial.
3. Choice B's reason is false, since C was highest in every trial. Choice C misreads the claim: it is only about "the brands we tested", so testing three brands is enough. Choice D's reason (strength) was never measured.

The correct answer is A.

Tip: A claim is supported when the data directly match it. Read the exact wording of the claim. Its limits (here, "of the brands we tested") tell you how much evidence it needs.

Check each statement against the table. Only one needs information that is not there.

Step by step:

1. A: Brand A's mean is 23 mL, the lowest. Supported.
2. B: Brand B's mean (30) minus Brand A's mean (23) = 7 mL. Supported.
3. C: Each brand's trials are close together: A 22–24, B 29–31, C 34–36, all within 2 mL. Supported.
4. D: "Best value for the money" requires prices and possibly other qualities, and the test measured neither. Not supported.

The correct answer is D.

Tip: On "NOT supported" questions, the right answer usually brings in a new idea (price, strength, safety, taste) that was never measured. Supported statements can be checked with numbers you can point to.

Test the claim against each row of the table. The ad claims FrostKeep keeps drinks colder than any other bottle tested, so FrostKeep would need the lowest temperature.

Evidence: After 6 hours, FrostKeep 7 °C vs. Brand X 6 °C. After 12 hours, FrostKeep 13 °C vs. Brand X 11 °C. Brand Y was warmest both times (10 °C and 18 °C).

Step by step:

1. FrostKeep beat Brand Y, but that is not enough for "any other bottle."
2. Brand X kept the water colder than FrostKeep at both measurement times.
3. One counterexample is enough to show the claim is not supported.

The correct answer is No, because Brand X kept the water colder than FrostKeep at both 6 and 12 hours.

Check: Remember the passage note: "A lower temperature means the bottle kept the water colder." Brand X's numbers are lower.

Tip: For a claim with any, all or best, look for a single case that breaks it. One exception defeats the claim.

Check each choice against the two tables.

Evidence: Earth is 147.1 million km from the Sun in early January (closest) and 152.1 million km in early July (farthest). Northville: −3 °C in January, 23 °C in July. Southport: 24 °C in January, 10 °C in July.

Step by step:

1. "Earth is closer to the Sun in January than in July" — 147.1 < 152.1. Supported.
2. "Northville is warmer in July than in January" — 23 > −3. Supported.
3. "When Northville has its warm season, Southport has its cool season" — in July, Northville is 23 °C while Southport is 10 °C. Supported.
4. "Earth's closest approach to the Sun causes summer everywhere on Earth" — in January, when Earth is closest, Northville averages −3 °C. Not supported.

The correct answer is Earth's closest approach to the Sun causes summer everywhere on Earth.

Check: If closeness caused summer for everyone, both cities would be warm in January. Only Southport is.

Tip: A claim about everywhere or all fails with a single counterexample — here, one cold city during closest approach.

Look at how the two groups were treated. The passage says the music plants were in her "sunny kitchen" while the no-music plants were on her "shaded back porch."

Step by step:

1. The kitchen plants grew more (mean \dfrac{13 + 11 + 12}{3} = 12 cm) than the porch plants (mean \dfrac{9 + 7 + 8}{3} = 8 cm).
2. But music was not the only difference. Light, and probably temperature, also differed. Plants need light to grow, so more sunlight alone could explain the extra growth.
3. Because two variables changed together, the data cannot show which one caused the difference.
4. Choice A is false: the music plants grew more. Choice C uses outside opinion instead of the evidence. Choice D (units) is not a weakness.

The correct answer is B.

Tip: A conclusion about a cause is only as strong as the design. If anything other than the tested variable differs between groups, the evidence is weak.

To find out whether music matters, the new evidence must remove the problems in the gardener's test. Her music plants were in a "sunny kitchen" and her other plants on a "shaded back porch", and she used only three plants per group.

Step by step:

1. Choice C puts all plants in the same room with the same light, temperature and water, so music is the only difference. It also uses many plants, reducing the effect of one unusual plant.
2. Choice A repeats the same flawed setup, so it would repeat the same confound.
3. Choice B is opinion from other people, not measured data.
4. Choice D adds another kind of music but still has no proper comparison group under identical conditions.

The correct answer is C.

Tip: The best "additional evidence" comes from a controlled test that changes only the variable in the claim, with a larger sample.

Find each peak using the correct axis. Deer (blue, left axis): 400, 520, 650, 700, 600 … highest in Year 4. Wolves (red, right axis): 10, 12, 16, 22, 27, 25 … highest in Year 5.

Step by step:

1. Wolves peaked one year after deer peaked. This fits the predator–prey idea in the passage: more deer (food) let more wolves survive, then more wolves eat more deer (choice C).
2. Choice A confuses the two peaks: Year 5 is the wolf peak; deer were already falling (700 → 600).
3. Choice B is false: wolves declined from Year 5 to Year 9 (27 → 13).
4. Choice D claims a single cause and says wolves ate "all" the extra deer. The graph shows the populations change together, but other factors (food, weather) could also matter, and the deer did not disappear.

The correct answer is C.

Tip: In predator–prey graphs, the predator's peak usually comes a little after the prey's peak. Check each line on its own axis.

The claim says wolves alone caused the decline. Evidence that challenges it would show another cause was also at work.

Step by step:

1. Choice B gives a second cause: deer starving in a harsh winter. If many deer died from cold and hunger in Year 5, then wolves were not the only reason the deer population fell.
2. Choice A is consistent with the claim. It already matches the graph, since wolves rose to 27 in Year 5 while deer fell.
3. Choice C (wolves eating deer) supports the idea that wolves had an effect. It does not show another cause.
4. Choice D is about Years 8–10, when deer were increasing again. It says nothing about the cause of the drop after Year 4.

The correct answer is B.

Tip: Evidence supports a claim when it agrees with it and challenges it when it points to a different explanation or contradicts it. Words like alone and only make a claim easier to challenge.

Check each claim against the evidence. Only claims that the table can confirm are supported.

Evidence: 6 h — FrostKeep 7, Brand X 6, Brand Y 10 °C. 12 h — FrostKeep 13, Brand X 11, Brand Y 18 °C.

Step by step:

1. FrostKeep kept water colder than Brand Y — 7 < 10 and 13 < 18. Supported.
2. All three let the water warm up — every bottle started at 4 °C and ended higher. Supported.
3. Brand X kept the water coldest — 6 and 11 are the lowest values. Supported.
4. FrostKeep keeps water below 10 °C for 24 hours — the test stopped at 12 hours, and FrostKeep was already at 13 °C by then. Not supported (the data even point the other way).

The correct answer is FrostKeep keeps water below 10 °C for at least 24 hours.

Check: A claim about a time beyond the test (24 hours) can never be supported by a 12-hour test.

Tip: In NOT supported questions, three choices will match the data. Hunt for the one that goes beyond the data or contradicts it.

Read both lines and keep to what they show.

Evidence: The retreat line climbs from 0 m (1990) to 610 m (2020). The temperature line rises from 8.1 °C to 10.3 °C, with one small dip in 2005.

Step by step:

1. Over 1990–2020, summers generally got warmer and the glacier retreated about 610 m. This describes the data accurately.
2. Temperature is the only cause goes beyond the data; a graph of two variables cannot rule out other causes (such as less snowfall).
3. Will disappear by 2040 is a prediction about the future that the graph does not give enough information to make.
4. Grew longer between 2000 and 2005 misreads the graph: the retreat total still increased (150 → 230 m).

The correct answer is from 1990 to 2020, average summer temperatures generally rose while the glacier retreated a total of about 610 meters.

Check: Every number in the correct answer can be read directly from the graph.

Tip: The best-supported conclusion is often the most modest one. Be suspicious of only, proves and long-range predictions.

Read the SkeeterShield row and compare it with the plain-lotion row.

Evidence: SkeeterShield: 0% bitten in hours 1–4, then 3% (hour 5), 10% (hour 6), 37% (hour 7) and 63% (hour 8). Plain lotion: 93–100% bitten every hour.

Step by step:

1. For the first 5–6 hours, SkeeterShield protected almost everyone (0–10% bitten vs. over 90% for plain lotion).
2. By hour 8, 63% of SkeeterShield users were bitten — most of them.
3. So the lotion works strongly for about 5–6 hours but does not prevent bites for 8 hours, as the company claims. The claim is only partly supported.

The correct answer is SkeeterShield protected nearly all volunteers for about 6 hours, but by hour 8 most users were being bitten, so the 8-hour claim is not fully supported.

Check: The other choices either ignore hours 7–8, ignore hours 1–6, or claim the lotion has no effect despite the huge gap with the plain lotion.

Tip: When a claim includes a number (8 hours, 99%, twice as fast), test that exact number against the data.

To test salt fairly, salt should be the only difference between the pots.

Evidence: The table shows Pot A on a small burner and Pot B on a large burner.

Step by step:

1. A large burner delivers more heat, so a pot on it can boil sooner with or without salt.
2. Pot B had both salt and the bigger burner, so we cannot tell which one caused the 50-second difference.
3. The other choices describe things that were kept the same (water volume, stopwatch) or a minor detail (measuring with a tablespoon), which cannot explain why B was faster.

The correct answer is Pot B was heated on a larger burner than Pot A.

Check: Swap the pots' burners in your mind. If the salted pot then took longer, the burner — not the salt — was the cause.

Tip: When a conclusion seems too easy, list every difference between the groups. An uncontrolled variable that pushes results in the same direction is the most serious flaw.

Judge each study's design and what it measured.

Study 1: 200 volunteers "randomly assigned", a control group that drank "the same decaffeinated coffee with nothing added," a blind design ("Neither the volunteers nor the researchers who tested them knew"), and reaction time measured directly with a computer test.

Study 2: 50 self-selected website visitors, no control group, and no reaction-time measurement at all. People simply described themselves as "very alert."

Step by step:

1. The claim is about reaction time. Only Study 1 measured it.
2. Study 1 controls other factors through random assignment and a blind design, so the difference (280 vs. 300 ms) can be linked to caffeine.
3. Study 2 shows only that coffee drinking and feeling alert go together. It is an observational correlation, based on opinions, from a small, self-chosen group.

The correct answer is D.

Tip: Strong evidence comes from a large, randomized, controlled, blinded study that measures exactly what the claim is about. Surveys of opinions are weak evidence for cause and effect.

Use the Study 1 table: caffeine group "280" ms, no-caffeine group "300" ms. Difference: 300 - 280 = 20 milliseconds faster on average.

Step by step:

1. Choice B states exactly what was found: on average, adults given 200 mg reacted about 20 ms faster.
2. Choice A says "every person." The results are group means, and some individuals may not have improved.
3. Choice C is about higher doses, but only one dose (200 mg) was tested, so this extrapolates beyond the data.
4. Choice D is about memory, which was never measured.

The correct answer is B.

Tip: The best-supported conclusion matches the study's conditions (who, what dose, what was measured) and uses "on average" for group data. Watch for "every", "always" and claims about untested conditions.

The nurse's evidence is weak because the two rooms differed in grade, size and illness exposure, and there was only one room per group. The best new evidence would fix these problems.

Step by step:

1. Use many classrooms, so one unusual class cannot decide the result.
2. Use classrooms of the same grades, and assign stations at random, so the groups are alike at the start.
3. Track absences for a whole school year, so a single outbreak cannot decide the result.
4. If the station classrooms still have clearly fewer absences, the claim is much stronger.

The correct answer is a school-year study in which 20 classrooms of the same grades are randomly assigned to get stations or not, and the classrooms with stations have fewer absences.

Check: The other choices measure opinions, repeat the same flaw, or are an advertisement rather than an absence study.

Tip: The strongest evidence usually comes from a controlled comparison: many subjects, random assignment, the same conditions, and only one difference.

The ranger links retreat to warming. A finding challenges this if the effect (retreat) happened just as strongly without the proposed cause (warming).

Step by step:

1. Retreat at the same rate during 1900–1950, when summers were cooler and not warming, would show that something other than warming can produce this retreat. That weakens the claim.
2. A distant city warming too is not about this glacier.
3. Faster retreat after 2010, when temperatures rose fastest, supports the claim.
4. More tourists has no stated connection to the glacier's melting.

The correct answer is old records show the glacier retreated at the same rate from 1900 to 1950, when summers were cooler and not getting warmer.

Check: Ask of each choice: "If this were true, would I be less confident that warming is the cause?" Only the first choice does that.

Tip: To challenge a cause-and-effect claim, look for (a) the effect without the cause, or (b) the cause without the effect.

Compare how each source was produced.

Study 1: volunteers were "randomly assigned"; there was a comparison group using a look-alike lotion with no repellent; every volunteer had the same mosquito exposure every hour; and the researchers recording bites "did not know which lotion each volunteer had used."

Source 2: 30 reviews chosen and posted by the company itself; no comparison group; unknown conditions (how many mosquitoes, how long, how much lotion).

Step by step:

1. Random assignment and a look-alike lotion make the groups alike except for the repellent.
2. Equal, measured exposure removes differences in mosquito numbers.
3. Blinded recorders prevent bias in counting bites.
4. The website reviews have none of these controls, and the company could leave out bad reviews.

The correct answer is Study 1, because it compared randomly assigned groups under the same measured conditions, with recorders who did not know which lotion was used.

Check: 27 out of 30 sounds impressive (90%), but without a comparison group we cannot know whether those campers would have been bitten anyway.

Tip: Rank evidence by its controls: comparison group, random assignment, same conditions, blinding, and an independent source.

The paper test is new evidence. The same 5 g sheet fell in 1.9 s when flat but in 0.67 s when crumpled.

Step by step:

1. Mass did not change between the two drops, so mass cannot explain the difference.
2. What changed was the shape: a flat sheet pushes against much more air, so air resistance slows it down.
3. The crumpled 5 g paper (0.67 s) fell almost as fast as the 110 g steel ball (0.64 s).
4. Revised idea: objects fall at nearly the same rate unless air resistance (shape and size) slows some down more than others.

The correct answer is shape, not mass, caused the difference, because the same sheet fell much faster once crumpled; air resistance can slow objects with a large surface.

Check: This revision explains both experiments: the three same-shape balls fell together, and the flat paper fell slowly.

Tip: When new data seem to show a difference, ask what really changed between the trials. Good scientists revise a conclusion to fit all the evidence.

The student says summer happens when Earth is closest to the Sun (early January). Look for data that clash with that.

Step by step:

1. Northville averages −3 °C in January, far colder than its 23 °C in July — and January is exactly when Earth is closest. This directly contradicts the claim.
2. Southport is warmer in January (24 °C) than in July (10 °C). That matches the claim, so it cannot challenge it — it is the trap.
3. The 5-million-km change in distance is a description; by itself it does not show which season occurs when.
4. Similar distances from the equator is a detail that makes the two cities fair to compare; it is not evidence against the claim.

The correct answer is Northville is much colder in January, when Earth is closest to the Sun, than in July.

Check: Both cities are the same distance from the Sun on any given day, yet they have opposite seasons. Distance cannot explain that; the tilt of Earth's axis does.

Tip: Evidence that agrees with a claim cannot challenge it, even if it is true and interesting. Look for the data that the claim cannot explain.

Compare the two lines before and after the policy.

Evidence: Prescriptions (left axis) rose from 20 thousand (Year 1) to 44 thousand (Year 6), then fell to 30, 22, 18 and 16 thousand in Years 7–10. Resistance (right axis) rose from 5% to 19% (Years 1–6), reached 20% in Year 7, then fell to 18%, 15% and 13% in Years 8–10.

Step by step:

1. Before the policy, use and resistance rose together.
2. After use dropped, resistance rose only 1 point (19% → 20%), then began to fall about a year later.
3. When a proposed cause is reduced and the effect then shrinks, the cause-and-effect claim is strengthened (though not proven — other changes could also have happened).

The correct answer is they strengthen the claim, because resistance leveled off and then fell after prescriptions dropped.

Check: The small rise in Year 7 (19% → 20%) does not undo the pattern; resistant bacteria do not vanish instantly, so a short delay is reasonable.

Tip: When new data arrive, ask: "Do they fit the claim, contradict it, or say nothing about it?" Avoid words like prove — data strengthen or weaken claims.

The original test has two problems: the burners were different, and there was only one trial of each pot.

Step by step:

1. Use the same burner for every trial, so heat is not a difference.
2. Run several trials with and without salt, so one lucky or unlucky trial cannot decide the result.
3. Start with water at the same temperature each time.
4. If salted water consistently boiled sooner under these conditions, the salt would be the likely cause; if not, the claim fails.

The correct answer is times from five trials of each pot, all on the same burner and starting with water at the same temperature.

Check: The other choices collect opinions, re-use the uncontrolled setup, or measure something unrelated to boiling time.

Tip: Two features make evidence much stronger: controlling other variables and repeating trials.

The new claim is about dose: more caffeine leads to more improvement. Study 1 tested only one amount ("200 mg"), so it cannot show a dose pattern.

Step by step:

1. To test a dose claim, the amount of caffeine must be the independent variable, with several levels.
2. Choice A does this and keeps the strengths of Study 1 (random assignment, blinding, measured reaction time). If reaction time keeps getting shorter as the dose rises, the claim is supported. If it levels off or gets worse, the claim is challenged.
3. Choice B adds more volunteers but still uses one dose, so it cannot show a dose pattern.
4. Choice C repeats the weak survey approach: opinions, no measured reaction time.
5. Choice D tests time after drinking, not amount of caffeine.

The correct answer is A.

Tip: Match the additional evidence to the exact claim. A claim about "more X → more Y" requires data at several levels of X.

Match the investigation to the claim. The new claim is about the temperature at which water boils, so the dependent variable must be a temperature reading, and the amount of salt should be the only thing changed.

Step by step:

1. Independent variable: amount of salt (0, 1, 2, 3 tablespoons per liter) — several levels show whether more salt has a bigger effect.
2. Dependent variable: temperature measured with a thermometer when the water boils.
3. Controls: same volume of water, same pot, same burner.
4. Repeat each level 3 times for reliability.

The correct answer is heat equal volumes of water with 0, 1, 2 and 3 tablespoons of salt per liter on the same burner, measure the temperature of each when it boils, and repeat each 3 times.

Check: Measuring boiling time answers a different question. Using different burners repeats the original flaw. A single trial gives no repeat and no comparison.

Tip: Before choosing a design, name the variable the claim is about. The design must measure that exact variable.

Compare how well each study isolates the cover crop.

Study A: a before-and-after comparison on one field. The passage notes that "Year 2 had much less rain than Year 1." Less rain means less erosion, so the drop from 9 to 4 tons could come from the weather, the rye, or both.

Study B: 20 plots "with the same soil and slope," cover crops assigned randomly, both groups measured in the same years (so the same rain), and results averaged over 3 years.

Step by step:

1. Study B's groups differ only in the cover crop; rain, soil and slope are shared.
2. Random assignment prevents choosing "better" plots for rye.
3. Ten plots per group and three years reduce the effect of chance.
4. Study A cannot separate the rye from the drier weather.

The correct answer is Study B, because the cover-crop and bare plots had the same soil, slope and weather, and plots were assigned at random.

Check: Study B's difference (7.8 vs. 3.1 tons) appeared under the same rain, so weather cannot explain it.

Tip: A before-and-after study on one site is weak whenever something else (weather, season, equipment) also changed between the two times.

The question is whether the cover-crop effect holds on steep slopes too. A good design needs a fair cover-crop-versus-bare comparison at each slope.

Step by step:

1. Use plots on both steep and gentle slopes, with similar soil.
2. On each slope, randomly assign half the plots to a cover crop and half to bare soil.
3. Measure soil lost from every plot over several years (so weather is shared and averaged).
4. Compare the cover-crop effect on steep plots with the effect on gentle plots.

The correct answer is on both steep and gentle slopes, randomly assign half the plots to a cover crop and leave half bare, then measure soil loss for several years.

Check: In the comparison of steep with cover vs. gentle without, slope and cover crop change together, so the effect of each cannot be separated.

Tip: When a study adds a second variable, every combination needs to be tested (steep + cover, steep + bare, gentle + cover, gentle + bare).