Research Summaries

🔬 Science Experiments: Research Summaries

Part 1 of 7 — Experimental Design: The Anatomy of an Experiment

On the Enhanced ACT, Science is an optional section: 40 questions in 40 minutes, every question with 4 answer choices. It earns its own score and is not part of the composite (which comes from English, Math, and Reading). Science questions come in three passage formats:

This lesson is about Research Summaries — the passages that describe experiments. You do not need outside science knowledge for most of these questions. You need to know how an experiment is built, because almost every question asks what was changed, what was measured, what was kept the same, and what the results allow you to say.

The six parts of every experiment

The quick test for IV vs. DV: ask "Did the experimenter pick this value, or record it?" Picked values are the IV. Recorded values are the DV. If a researcher set the temperatures at 20, 40, 60, and 80 °C and then timed how long sugar took to dissolve, temperature was picked (IV) and time was recorded (DV).

Constants are not the same thing as the control group

This is the most common mix-up on the test, so keep the two ideas separate.

• A constant is a variable that never changes from trial to trial: the same 200 mL of water, the same 6-volt battery, the same room temperature.
• The control group is a group or trial — the one that gets no treatment (or the standard treatment) so the others can be compared with it.

In a fertilizer study with plants given 0 g, 5 g, and 10 g each week, the 0 g group is the control group. Water, soil, and light are constants. The 5 g group is not a control just because it has the smallest dose; it is still being treated.

Kinds of control groups

A medical result such as "40 of 50 patients felt better after taking Drug X" means little by itself. If headaches often fade within an hour anyway, you need a placebo group of similar patients. Only if the Drug X group improves more than the placebo group is the drug doing something.

Hypotheses and predictions

A hypothesis names a relationship between the IV and the DV, and it predicts a direction:

• Hypothesis: "Increasing the angle of a ramp increases how far a toy car rolls past the bottom."
• Prediction for the study: as the angle goes from 10° to 40°, the rolling distance should get longer.

When the ACT asks which hypothesis a study was designed to test, match the hypothesis to what was varied and what was measured. A hypothesis about ramp surface cannot be what a study tested if every trial used the same surface.

Reading a results table

• IV: ramp angle (left column, evenly spaced values someone chose)
• DV: distance rolled (right column, measured values)
• Trend: distance increases as angle increases

If the procedure says the same car and the same ramp surface were used each time, the car's mass and the surface are constants.

Worked Examples

Procedure: A student placed 2.0 g of yeast in each of four flasks containing 100 mL of water. The water in the flasks contained 0%, 2%, 4%, or 6% sugar. All flasks were kept at 30 °C. She collected the gas released by each flask for 20 minutes and recorded its volume in milliliters.

Solution:

1. What did she pick? The sugar percentage (0, 2, 4, 6) → IV.
2. What did she record? The volume of gas → DV.
3. What stayed the same? Yeast mass (2.0 g), water volume (100 mL), temperature (30 °C), collection time (20 min) → constants.
4. Which flask is the baseline? The 0% sugar flask → control. It shows how much gas the yeast releases with no added sugar.

ACT skill: Underline numbers that repeat in every trial (constants) and numbers that change (the IV).

Question: Using the yeast study above, which hypothesis was the study designed to test?

• Yeast releases more gas at higher temperatures.
• Yeast releases more gas when more sugar is available.
• Larger amounts of yeast release more gas.
• Yeast releases more gas in larger volumes of water.

Solution: The study varied only sugar percentage and measured gas, so the hypothesis must link sugar to gas: "Yeast releases more gas when more sugar is available." Temperature, yeast mass, and water volume were all held constant, so no hypothesis about them could be tested here.

Practice Set A — The Yeast Study 🎯

A student placed 2.0 g of yeast in each of four flasks of 100 mL of water containing different amounts of sugar. Every flask was kept at 30 °C, and gas was collected for 20 minutes.

Label the Ramp Study 🔍

A student released the same toy car from a wooden ramp tilted at 10°, 20°, 30°, and 40° and measured how far the car rolled past the bottom.

Practice Set B — New Studies 📋

ACT-Style Practice

Mini-passage: To study how water temperature affects the time needed for a tablet to dissolve, a student dropped identical 1 g tablets into 250 mL of water at 10, 25, 40, and 55 °C. The water was not stirred. She recorded the time until no solid remained: 96 s, 61 s, 38 s, and 24 s.

Try each question in under 30 seconds.

Stirring was the same in every trial (none), so it is a constant. A variable can be the IV only if it changes from trial to trial.

ACT Tip: Before reading the questions, spend about 20 seconds labeling the IV, DV, and constants in the margin of each experiment. Many questions are answered by those labels alone.

Key Takeaways

• Enhanced ACT Science is optional: 40 questions, 40 minutes, 4 choices, scored separately from the composite.
• IV = picked by the experimenter (often the left column or x-axis). DV = recorded as the outcome (often the right column or y-axis).
• Constants are variables kept the same in every trial; the control group is the untreated or standard group used as a baseline. Do not confuse them.
• A placebo group is the control when people (or conditions) might improve on their own.
• A study can only test a hypothesis that links the variable it changed to the variable it measured.

🧪 Variables & Controls

Part 2 of 7 — Fair Tests, Confounding Variables, Sample Size, Replication & Bias

Part 1 named the pieces of an experiment. This part is about whether those pieces were put together fairly. Research Summaries questions often ask "Which change would improve the experiment?" or "Why can't the student conclude…?" The answer is almost always one of the design principles below.

Principle 1: Change one variable at a time

If two things change between trials, you cannot tell which one caused the difference. A variable that changes along with the IV is called a confounding variable.

The period was longer in Trial 2, but both the string length and the bob mass changed. The longer period could come from either one (or both). The fix: keep the mass at 100 g in both trials and change only the length.

How confounds hide in passages: look for differences in time of day, day of the week, location, equipment, or the people involved. "Rock music on Monday morning, classical music on Friday evening" changes the music and the time.

Principle 2: Hold everything else constant

Constants are what make the comparison fair. When the ACT asks why a researcher kept something the same, the answer is nearly always: so that any difference in the DV can be attributed to the IV.

Principle 3: Compare against a control

A control group (untreated, placebo, or standard condition) shows what happens without the treatment. Without it, a change might have happened anyway.

• Blinding: in a single-blind study the participants do not know whether they got the treatment or the placebo. In a double-blind study neither the participants nor the people measuring the results know. Blinding prevents expectations from shaping the results.
• Random assignment: placing subjects into groups by chance (for example, by drawing names) keeps the groups similar in every way except the treatment.

Principle 4: Use enough subjects and repeat trials

Two teams that get the same average difference are not equally trustworthy if one tested 8 people and the other tested 400. The larger sample deserves more confidence.

Replication means repeating the same test on the same kind of material. Testing a different material, changing the procedure, or measuring a different property is a new experiment, not a replication.

Principle 5: Watch for bias

The ACT "improve the experiment" checklist

When a question asks which change would most improve a study, check in this order:

1. Did more than one thing change? → hold the extra variable constant.
2. Is there no baseline? → add a control (or placebo) group.
3. Only one trial or a tiny sample? → repeat trials or test more subjects.
4. Is the sample unrepresentative? → sample randomly from the whole population.
5. Could the measurement be inconsistent or biased? → use one calibrated instrument and blind the observer.

Wrong choices usually add a new variable ("also test a third music genre on Wednesday"), change the DV ("measure accuracy instead of speed"), or make the conditions more different ("play the rock music louder"). None of these fixes the original flaw.

Worked Examples

Study: A student wants to know whether music genre affects typing speed. She types for 5 minutes with rock music on Monday morning and 5 minutes with classical music on Friday evening, one trial each.

Solution:

1. Variables that differ between the two sessions: genre (intended IV), day, and time of day → confounds.
2. Only one trial per genre → no way to tell a real difference from a lucky or unlucky session.
3. Best fix: test both genres at the same time of day, several times each, and compare the averages.

ACT skill: The best improvement removes the extra variable and adds repetition; a choice that does only one of these is weaker if the other option exists.

Study: Lab A tests a new hand lotion on 6 volunteers and finds skin moisture rises by 12%. Lab B uses the same procedure on 300 volunteers and finds a 4% rise.

Solution: Lab B's result is more reliable. With only 6 people, one or two unusual volunteers can push the average far from the true effect. A large sample averages out individual differences, so 4% is the better estimate. A bigger effect in a smaller study does not make that study more convincing.

Practice Set A — Spot the Flaw 🎯

Match the Flaw to the Fix 🔧

Practice Set B — Improve the Experiment 📋

ACT-Style Practice

Mini-passage: A student tested whether a new brand of battery lasts longer than an old brand. She ran the new battery in a flashlight with an LED bulb and the old battery in a flashlight with an incandescent bulb, once each. The new battery lasted 31 hours and the old one lasted 9 hours.

1. What confounds the comparison?
2. What two changes would make the test fair and reliable?

1. The bulb type changed along with the battery brand. LED bulbs use less power, so the bulb alone could explain the longer time.
2. Use identical flashlights and bulbs for both brands, and test several batteries of each brand, comparing the averages.

ACT Tip: If a "Why can't the student conclude…" question appears, list every difference between the conditions. The answer is the difference that is not the IV.

Key Takeaways

• Change only one variable at a time; anything that changes along with the IV is a confounding variable.
• Constants exist so that a change in the DV can be attributed to the IV.
• Placebo groups, blinding, and random assignment keep expectations and pre-existing differences out of the results.
• Larger samples and repeated trials make averages more reliable; replication repeats the same procedure on the same material.
• The best "improve the experiment" answer fixes the original flaw; it does not add a new variable or switch the DV.

📑 Research Summaries

Part 3 of 7 — Reading a Multi-Experiment Passage

A Research Summaries passage describes a short research project, usually as two or three experiments that share a goal. Each experiment has a short procedure and a table or graph of results. The experiments are built to differ in a specific way, and many questions test whether you noticed how.

The usual layout

The "except" sentence is the key to the passage

Later experiments are usually described by how they differ from Experiment 1:

"Experiment 2: The procedure of Experiment 1 was repeated, except that every tray was kept at 25 °C and the volume of water was varied."

That one sentence tells you three things:

1. The new IV is water volume.
2. Temperature, which was the IV in Experiment 1, is now a constant (25 °C).
3. Everything else from Experiment 1 (seed type, light, number of seeds) is still the same.

Write the change in the margin next to each experiment, for example "E2: water varies, T = 25". Questions such as "How did Experiment 2 differ from Experiment 1?" become one-glance answers.

Common Research Summaries question types

Use the right experiment

A hypothesis about water can only be judged with the experiment that varied water. Data from the temperature experiment says nothing about water, even if the numbers look relevant. Wrong answers often quote a true fact from the wrong experiment.

Also check the whole range. If germination rises from 5 mL to 10 mL of water and then falls from 10 mL to 20 mL, the hypothesis "germination increases steadily with more water" is not supported, even though part of the data goes up.

The shared condition links experiments

When Experiment 2 holds temperature at the value that worked best in Experiment 1, one trial in Experiment 2 repeats a trial from Experiment 1. The two should give about the same result, and they act as a built-in consistency check. They also let you connect the two tables: the 25 °C, 10 mL condition appears in both.

Preview: combining two experiments

Sometimes each experiment varies a different factor, and a question asks about a condition that combines them. Example:

Prediction for 2 bulbs at 2 m: Experiment 1 says moving from 1 m to 2 m divides the reading by 4 (400 → 100). Experiment 2 says doubling the bulbs doubles the reading. So 2 bulbs at 2 m ≈ 100 × 2 = 200. Start from a tested value that matches one factor, then apply the effect of the other factor. Part 6 practices this in depth.

Worked Examples

Passage summary: In Experiment 1, a student dropped one rubber ball from heights of 0.5, 1.0, and 1.5 m and measured how high it bounced. In Experiment 2, she dropped balls made of rubber, plastic, and foam, all with the same diameter, from 1.0 m.

Question: Which variable was tested in Experiment 2, and which variable from Experiment 1 became a constant?

Solution: Experiment 2 changed the ball material (new IV). Drop height, the IV of Experiment 1, was fixed at 1.0 m (now a constant). Bounce height was the DV in both experiments, and ball diameter was held the same.

Data: Experiment 1 (all trays given 10 mL of water per day): 15 °C → 40%, 20 °C → 65%, 25 °C → 85%, 30 °C → 70% germination. Experiment 2 (all trays at 25 °C): 5 mL → 50%, 10 mL → 85%, 15 mL → 80%, 20 mL → 45%.

Question: Do the results support the hypothesis that germination increases steadily as more water is supplied?

Solution:

1. Water is varied only in Experiment 2, so use that table.
2. Germination rises from 50% to 85% (5 → 10 mL) but then falls to 80% and 45%.
3. A steady increase is not supported. The highest germination in either table, 85%, is the shared condition: 25 °C with 10 mL per day.

Practice Set A — The Enzyme Study 🎯

Catalase, an enzyme found in potatoes, breaks down hydrogen peroxide and releases oxygen gas. In every trial, potato extract was added to 10 mL of 3% hydrogen peroxide, and the oxygen released in 60 seconds was collected.

Experiment 1: 2 mL of extract was used at pH 7, and the temperature was varied.

Experiment 2: The procedure of Experiment 1 was repeated, except that the temperature was kept at 30 °C and the pH was varied.

Experiment 3: The temperature was kept at 30 °C and the pH at 7, and the volume of extract was varied.

Practice Set B — Using the Enzyme Study 📋

Use the three experiments from Practice Set A.

ACT-Style Practice

Passage summary: Experiment 1 measured the resistance of 1 m copper wires of diameter 0.5, 1.0, and 2.0 mm: 0.088, 0.022, and 0.0055 ohms. Experiment 2 measured 1.0 mm wires of length 1, 2, and 4 m: 0.022, 0.044, and 0.088 ohms.

Question: Estimate the resistance of a 2 m wire with a 2.0 mm diameter.

Start from Experiment 1: a 1 m, 2.0 mm wire has 0.0055 ohms. Experiment 2 shows that doubling the length doubles the resistance. So a 2 m, 2.0 mm wire has about 0.011 ohms. Using 0.044 ohms would be a mistake: that value is for a 2 m wire of the thinner 1.0 mm diameter.

ACT Tip: When a stem names a condition, find every table that contains part of it. The answer often requires one number from each.

Key Takeaways

• Research Summaries passages describe 2–3 related experiments; mark the IV, DV, and constants for each.
• The "repeated except…" sentence tells you the new IV and which old variable became a constant.
• Judge a hypothesis with the experiment that varied that variable, and check the entire trend, not just part of it.
• A condition repeated across experiments is a consistency check and the bridge between tables.
• To combine experiments, start from a tested value for one factor and apply the effect of the other.

⚖️ Conflicting Viewpoints

Part 4 of 7 — Comparing Competing Explanations

A Conflicting Viewpoints passage gives a short background paragraph, then two or more explanations of the same observation from "Scientist 1 and Scientist 2", "Student 1, 2, and 3", or "Hypothesis 1 and Hypothesis 2". There may be little or no data. The questions test whether you can keep each viewpoint's claims straight and judge new evidence against them.

Research Summaries often use the same reasoning: two hypotheses are proposed, and an experiment is run to decide between them. So this skill shows up in both formats.

Rule 1: Judge by the passage, not by what you think is true

The ACT does not ask which viewpoint is scientifically correct. It asks what each viewpoint claims and how evidence relates to those claims. A viewpoint you know to be outdated can still be "supported" by a particular finding in the question.

Rule 2: Build a claim table as you read

After each viewpoint, write a few words in the margin. Then compare.

Background: A lake's water became much cloudier after 2018.

Rows where the columns differ are disagreements. Facts both accept (the lake got cloudier after 2018) are agreements.

Rule 3: Support and weaken questions

The "but not" trap: an observation both viewpoints already accept cannot separate them. If both scientists agree the lake is cloudier, "the lake is cloudier" supports neither one over the other. Look for the finding that touches a row where the table differs.

Example: "Fertilizer levels in the lake did not change after 2018." This contradicts Scientist 1's source and leaves Scientist 2 untouched, so it weakens Scientist 1 but not Scientist 2.

Rule 4: Three or more viewpoints

With three viewpoints, a single finding can support two of them and weaken the third. Group the viewpoints by their key claim.

Background: A sealed, inflated balloon shrinks after an hour in a freezer.

If the balloon returns to its full size when warmed without being refilled, the air must still be inside, which weakens Student 1. Students 2 and 3 both survive, because both say no air escaped.

Rule 5: "How would Scientist X explain…?"

These questions ask you to apply a viewpoint to a new situation. Use only that scientist's mechanism. If Scientist 2 says carp stir up sediment, then Scientist 2 would explain "the water cleared after the carp were removed" by saying less sediment was being stirred up.

Reading strategy

1. Read the background for the shared observation.
2. Read Viewpoint 1 and write its claim in a few words; repeat for each viewpoint.
3. Note one key difference per pair of viewpoints.
4. For each question, find the row of your claim table that the finding touches.

Worked Examples

Viewpoints: Student 1 says a wool sweater keeps you warm because the wool itself produces heat. Student 2 says the sweater produces no heat; it slows the loss of heat your body produces.

Question: Which finding would support Student 2 but not Student 1?

• A person wearing the sweater feels warmer than without it.
• A thermometer wrapped in the sweater, with no person inside, stays at room temperature.
• The sweater feels warm right after it is taken off.
• Wool sweaters are thicker than cotton shirts.

Solution: Both students agree that the sweater makes a person feel warmer, so that finding cannot separate them. If wool produced heat, a thermometer wrapped in it should warm up. The thermometer staying at room temperature contradicts Student 1 and fits Student 2. A sweater feeling warm after use fits both (it held body heat, or it made heat), and thickness is not part of either claim.

Viewpoints: Three hypotheses explain why a lizard species is darker in the mountains than in the lowlands. H1: dark color absorbs more sunlight in cold air. H2: dark color hides lizards from predators on dark mountain rock. H3: color is set by diet, and mountain insects contain a darkening pigment.

Finding: Lowland lizards raised in the lab on mountain insects stayed light-colored.

Solution: H3 predicts that eating mountain insects darkens the lizards, so this finding weakens H3. H1 and H2 say nothing about diet, so the finding neither supports nor weakens them.

Practice Set A — The Cloudy Lake 🎯

The water in a lake became much cloudier after 2018.

Scientist 1: New farms upstream began draining fertilizer into the lake. The fertilizer fed rapid algae growth, and the algae clouded the water.

Scientist 2: An invasive carp species was introduced to the lake in 2018. The carp feed by rooting through the lake bottom, stirring up sediment that clouds the water.

Practice Set B — The Candle in the Jar 📋

A lit candle is placed in a dish, and a glass jar is lowered over it. After several seconds the flame goes out. Air normally contains about 21% oxygen.

Student 1: The flame burns until it has used up all of the oxygen in the jar.

Student 2: The flame uses up only part of the oxygen. It goes out once the oxygen level falls below the minimum a flame needs.

Student 3: The amount of oxygen does not matter. Carbon dioxide from the flame builds up and smothers it.

ACT-Style Practice

Background: A metal spoon and a wooden spoon have both sat in a 20 °C room for hours, yet the metal spoon feels colder.

• Student 1: The metal spoon is actually at a lower temperature than the wooden spoon.
• Student 2: Both spoons are at 20 °C; metal carries heat away from the hand faster, so it feels colder.

1. On what point do the students agree?
2. Which measurement would support Student 2 but not Student 1?

1. They agree that the metal spoon feels colder; they disagree about why.
2. A thermometer reading 20 °C on both spoons. Student 1 claims the temperatures differ, so equal readings contradict Student 1 and match Student 2.

ACT Tip: In Conflicting Viewpoints, wrong answers often restate the shared observation. If both viewpoints already accept a fact, it cannot support one over the other.

Key Takeaways

• Judge viewpoints by what they claim, not by which one you believe is true.
• Build a claim table: agreements are shared facts; disagreements are where the evidence can decide.
• "Supports X but not Y" needs a finding X predicts that Y cannot explain; a shared observation never works.
• A finding that contradicts one claim weakens only the viewpoints that make that claim.
• With three viewpoints, group them by their key claim; one finding can weaken one viewpoint and leave two standing.

🧐 Evaluating Conclusions

Part 5 of 7 — Does the Conclusion Follow from the Evidence?

Many Research Summaries questions end with a claim: "A student concluded that…. Is this conclusion supported?" or "Which conclusion is best supported by the results?" To answer, compare the claim with exactly what was tested. A conclusion is supported only if the data show it, for the subjects and conditions that were actually studied.

Five ways a conclusion goes wrong

Read the whole trend

Data on the ACT often rise and then fall (or the reverse). Describe the pattern in pieces:

• Supported: "Height increased up to 10 g, then decreased."
• Not supported: "Height increased with every added 5 g" (it fell after 10 g).
• Not supported: "More than 20 g would produce the tallest plants" (the trend is heading down).
• Check the extremes too: unfertilized plants (12 cm) were shorter than 20 g plants (17 cm).

Stay inside the tested scope

A study of one bacterial strain says nothing certain about other strains. A shoe tested on 20 professional male marathoners aged 25 to 30 says nothing certain about every runner. The safest supported conclusion names what was actually tested: "The fertilizer increased yield in the corn plants tested."

Watch for the absolute words that signal overreach:

Correlation, causation, and alternative explanations

When two variables change together in observational data (nobody assigned the conditions), ask: Is there a third factor that could drive both?

• Umbrella sales and traffic accidents both rise on certain days → rain causes both.
• Towns with more trees have lower summer temperatures → but if the tree-rich towns are also at higher elevations, elevation alone could explain the cooler temperatures.

A fact that offers such a third factor weakens a causal conclusion. A fact that keeps conditions the same across the comparison (every temperature recorded at noon) strengthens it.

"Cannot be determined"

If a study held a variable constant, it cannot tell you how that variable affects the outcome. A plant study done entirely at 25 °C cannot show whether temperature matters. On the ACT, the answer to "Do the results show how temperature affects growth?" would be no, because temperature was not varied.

How ACT answer choices are built

Conclusion questions often use a "Yes, because… / No, because…" format. Two steps:

1. Decide yes or no from the data.
2. Among the choices with the right yes/no, pick the reason that is true and relevant. A wrong choice often pairs the correct "No" with a true fact from the wrong experiment, or with a reason that does not address the claim.

Worked Examples

Data: A reaction ran at 2 mmol/min at 10 °C, 4 mmol/min at 20 °C, and 8 mmol/min at 30 °C. A researcher concluded that the rate is certain to be 64 mmol/min at 60 °C.

Solution:

1. Within the data, the rate doubles every 10 °C, so the pattern itself is real.
2. No trials were run above 30 °C. Many processes level off or reverse at high temperatures (an enzyme, for example, can stop working).
3. Evaluation: the conclusion goes beyond the data. A prediction at 60 °C could be an estimate, but not a certainty.

Data: Experiment 1 (magnet touching the clips): magnets of strength 1, 2, and 3 units lifted 5, 10, and 15 paper clips. Experiment 2 (strength 2 magnet): held 1, 2, and 3 cm from the clips, it lifted 10, 4, and 1 clips.

Claim: "The number of clips lifted depends only on magnet strength."

Solution: Experiment 2 held strength constant, yet the number of clips changed with distance. So the answer is No, because the clips lifted also changed with distance. "No, because the clips fell as strength rose" has the right yes/no but a false reason, since clips rose with strength.

Practice Set A — The Trout Study 🎯

A biologist raised young trout of one species in five tanks, each held at a different water temperature. All tanks received the same food ration, and each tank held 20 fish.

Practice Set B — Weigh the Claim 📋

ACT-Style Practice

Mini-passage: A microbiologist spread one bacterial strain on plates containing 0, 2, 4, or 8 mg/L of an antibiotic and counted colonies after 24 hours at 37 °C: 210, 150, 60, and 0 colonies.

Decide whether each statement is supported, goes beyond the data, or cannot be determined.

ACT Tip: The best-supported conclusion is usually the most modest one that still matches the data. Absolute words such as "all," "always," "proves," and "exactly" are warning signs.

Key Takeaways

• A conclusion must match the whole trend; rise-then-fall data do not support "always increases."
• Stay inside the tested scope: subjects, conditions, and range. Beyond that is overgeneralization or extrapolation.
• A sample that does not represent the population (a gym, a swim team, young professionals) biases the result.
• Two things rising together can share a third cause; a fact that supplies that cause weakens a causal claim.
• A variable held constant cannot be evaluated. For yes/no choices, get the direction right first, then the reason.

🔗 Combining Experiments & Predicting New Trials

Part 6 of 7 — Using Several Studies Together

The hardest Research Summaries questions ask you to go beyond a single table: combine results from two experiments, predict the outcome of a trial nobody ran, design the next experiment, or decide which results would support a claim. Each of these has a reliable method.

Skill 1: Interpolation — predicting inside the tested range

If the new condition falls between two tested values, the result should fall between their results.

A 7 cm wing lies halfway between 6 and 8 cm, so its fall time should be about halfway between 1.6 and 2.0 s: about 1.8 s. Here the time rises by 0.4 s for every 2 cm, a steady pattern. If the pattern is curved, the estimate is rougher, but it still belongs between the neighbors.

Skill 2: Extrapolation — predicting outside the range

For a condition beyond the tested values, continue the trend only as far as it reasonably goes. ACT answer choices for these questions are often ranges ("greater than 2.4 s") or values that continue the pattern. A 12 cm wing would most likely fall in more than 2.4 s (roughly 2.8 s if the pattern holds). Remember from Part 5: an extrapolated value is an estimate, never a certainty.

Skill 3: Combining two experiments

When Experiment 1 varies one factor and Experiment 2 varies another, find the shared condition that links them, then apply each effect in turn.

1. Start from the trial that matches the new condition in one factor.
2. Use the other experiment to see how changing the second factor scales the result (doubles it, halves it, adds a fixed amount).
3. Apply that change.

Shared condition: 8 cm wings, 1 clip, 2.0 m → 2.0 s in both tables.

Prediction for 10 cm wings dropped from 1.0 m: Experiment 1 gives 2.4 s for 10 cm at 2.0 m. Experiment 3 shows that halving the height halves the time (2.0 → 1.0 s). So 10 cm at 1.0 m ≈ 2.4 ÷ 2 = 1.2 s.

Skill 4: Designing the next trial

To test a new variable, the new experiment must:

• change only the new variable,
• hold every other variable at the values already used (often the shared condition), and
• measure the same DV so the results can be compared with earlier ones.

A choice that changes two things, drops the measurement, or tests a variable already studied is wrong.

Skill 5: Which results would support a claim?

Turn the claim into a prediction about the data, then find the data set that matches it.

Claim: "Increasing the salt concentration of water raises its boiling point." Prediction: as salt goes up, boiling point goes up at every step.

Skill 6: Comparing studies that disagree

When two studies of the same question get different results, look for a procedural difference: different subjects, different starting conditions, different amounts, different measurement times. The best explanation is a difference that would plausibly change the DV. For example, a fertilizer that helped plants in poor soil may show no effect in soil that was already rich in the same nutrients.

Skill 7: Predicting under a hypothesis

"If Hypothesis 2 is correct, what would be the result of the new trial?" Apply the hypothesis's mechanism to the new setup, exactly as you did with "How would Scientist X explain…?" in Part 4. If a hypothesis says plants bend toward blue light but not red, a seedling lit with red from the left and blue from the right should bend right.

Worked Examples

Data: Experiment 1 (1 bulb): a light sensor read 400 units at 1 m, 100 at 2 m, and 25 at 4 m. Experiment 2 (sensor at 1 m): 1, 2, and 3 bulbs gave 400, 800, and 1,200 units.

Question: Predict the reading for 3 bulbs at 2 m.

Solution:

1. Shared condition: 1 bulb at 1 m → 400 in both tables.
2. From Experiment 1, moving from 1 m to 2 m divides the reading by 4.
3. From Experiment 2, 3 bulbs at 1 m gives 1,200.
4. 3 bulbs at 2 m ≈ 1,200 ÷ 4 = 300 units.

Studies: Study A found that Fertilizer N increased lettuce growth by 30%. Study B, using the same fertilizer dose, found no increase.

Procedural differences: Study A used sandy soil low in nitrogen; Study B used soil already rich in nitrogen. Study A measured growth after 30 days; Study B measured after 32 days.

Solution: A 2-day difference in measurement time is unlikely to erase a 30% effect. Soil that already contains plenty of nitrogen would leave little room for a nitrogen fertilizer to help, so the soil difference best explains the conflicting results.

Practice Set A — Paper Helicopters 🎯

Students built paper helicopters, added paper clips to the base for weight, and timed how long each took to fall to the floor.

Experiment 1: 1 paper clip, dropped from 2.0 m, wing length varied.

Experiment 2: 8 cm wings, dropped from 2.0 m, number of paper clips varied.

Experiment 3: 8 cm wings, 1 paper clip, drop height varied.

Predict the Fall Times 🧮

Use the helicopter experiments. Give each answer in seconds.

1. 10 cm wings, 1 clip, dropped from 1.0 m

2. 8 cm wings, 4 clips, dropped from 1.0 m

3. 5 cm wings, 1 clip, dropped from 2.0 m

Practice Set B — Design, Support, and Predict 📋

ACT-Style Practice

Mini-passage: Experiment 1 (all wires 1 m long) measured the resistance of copper wires with diameters 0.5, 1.0, and 2.0 mm: 0.088, 0.022, and 0.0055 ohms. Experiment 2 (all wires 1.0 mm in diameter) measured lengths of 1, 2, and 4 m: 0.022, 0.044, and 0.088 ohms.

1. Predict the resistance of a 4 m wire with a 0.5 mm diameter.
2. Which new trial would test whether copper and aluminum wires differ in resistance?

1. A 1 m, 0.5 mm wire has 0.088 ohms. Experiment 2 shows that 4 times the length gives 4 times the resistance, so 4 m gives about 0.35 ohms (0.088 × 4 = 0.352).
2. Measure a 1 m copper wire and a 1 m aluminum wire, both 1.0 mm in diameter, with the same equipment. Only the metal changes, and the copper result can be checked against the known 0.022 ohms.

ACT Tip: For a combined prediction, say the chain out loud: "Start at the tested value, then double (or halve) for the other factor." Wrong answers usually skip one link of the chain.

Key Takeaways

• Interpolate between neighboring tested values; extrapolate cautiously, following the trend's direction.
• To combine experiments, start from the shared condition or a matching trial and apply the second factor's effect.
• A good new trial changes only the new variable and keeps everything else at the earlier values, measuring the same DV.
• To find data that support a claim, turn the claim into a predicted pattern and match it at every step.
• Conflicting studies are best explained by a procedural difference that would plausibly change the DV.

⏱️ Timed Research Summary Passage

Part 7 of 7 — Putting It All Together Under Enhanced ACT Timing

This part pulls every skill from Parts 1–6 into one full Research Summaries passage, worked under real timing.

The timing you are working with

One minute per question includes the time spent reading the passage. A passage with 8 questions therefore gets about 8 minutes total, not 8 minutes plus reading time. That is why the reading routine below is short.

A 90-second passage routine

1. Introduction (15 seconds): what question are the researchers asking? Write it in a few words.
2. Each experiment (20–25 seconds): label the IV, DV, and constants in the margin. For "repeated except…" experiments, write only what changed.
3. Tables (10 seconds each): read the headings and units, and note the overall trend (up, down, or up-then-down). Do not study every number.
4. Go to the questions. Return to the passage only for the specific number or sentence a question needs.

The question types you will meet, and where you learned them

Triage: which questions to do first

• Fast (about 30 seconds): identify a variable, read a value, find the highest or lowest result. Do these immediately.
• Medium (about 1 minute): trends, "how did Experiment 2 differ," interpolation.
• Slow (up to 2 minutes): combining experiments, evaluating a conclusion with "Yes, because / No, because" choices, designing a new trial.

If a question is taking more than 2 minutes, eliminate what you can, choose an answer, mark it, and move on. Saved seconds from the fast questions pay for the slow ones.

Elimination habits that work on every question

• Cross out choices that quote a true fact from the wrong experiment.
• Cross out choices that use absolute words ("all," "proves," "exactly") about untested cases.
• In "Yes/No, because" questions, decide yes or no first, then compare only the two remaining reasons.
• In "supports X but not Y" questions, cross out any finding both viewpoints already accept.

Replication inside a passage

Researchers often run several dishes, plants, or trials per condition and report the mean. When individual trials are shown, look at how much they vary. If three dishes in one condition gave 83%, 86%, and 83%, the mean of 84% is trustworthy. If they gave 40%, 84%, and 128 eggs out of 200, something is wrong with the measurement. A single trial per condition is a weakness: the result could be a fluke, and repeating it would make it more reliable.

Now try the passage

Set a timer for 8 minutes and answer all 8 questions in the two practice sets below without stopping. Then read every explanation, including those for questions you got right, and note which Part each mistake came from.

Worked Examples

Passage: A student hypothesized that adding baking soda raises the pH of pond water. She added 0, 1, 2, or 3 g of baking soda to four 500 mL samples from the same pond and measured pH values of 7.0, 7.8, 8.3, and 8.6. She tested one sample at each amount.

Margin notes: IV = baking soda (g); DV = pH; constants = 500 mL, same pond; control = 0 g; trials = 1 per amount.

Likely question: "Which statement best evaluates the study?" The pH rose with every gram, so the results support the hypothesis. With only one sample per amount, repeated trials would make the result more reliable. The study cannot show the effect holds for every pond, because water came from only one pond.

Passage: Students found that classrooms with windows had higher average test scores than windowless classrooms. Every room with windows was in the school's new building, which also had newer computers and smaller classes.

Question: Why can't the students conclude that windows caused the higher scores?

Solution: Building age, computers, and class size all differ along with the windows, so each is an alternative explanation. Under time pressure, list every difference between the groups; any difference other than the one being tested is a confound.

Timed Passage — Brine Shrimp Hatching (Questions 1–4) 🎯

Brine shrimp eggs can stay dormant for years and hatch when placed in salt water. Biologists studied how conditions affect the percentage of eggs that hatch. In every trial, 200 eggs were placed in a dish of salt water, and the number hatched after 48 hours was counted. Each condition was tested in 3 separate dishes, and the mean percentage hatched is reported.

Experiment 1: Dishes were kept at 25 °C under continuous light. The salinity (grams of salt per liter of water) was varied.

Experiment 2: The procedure of Experiment 1 was repeated, except that salinity was kept at 30 g/L and the temperature was varied.

Experiment 3: Dishes were kept at 30 g/L and 25 °C. Half of the dishes were kept under continuous light, and the rest were kept in complete darkness.

Timed Passage — Brine Shrimp Hatching (Questions 5–8) 📋

Use the passage above. Two hypotheses had been proposed before Experiment 3:

Hypothesis 1: Light triggers hatching, so more eggs hatch in light than in darkness.

Hypothesis 2: Hatching depends only on salinity and temperature; light has no effect.

Diagnose Your Misses 🔍

Match each mistake to the Part that teaches the skill you need to review.

ACT-Style Practice: Timing Check

Score your timed attempt on the brine shrimp passage.

Extra question (1 minute): Based on Experiments 1 and 2, would you expect more eggs to hatch at 40 g/L and 30 °C than at 40 g/L and 25 °C?

Probably yes. At 30 g/L, raising the temperature from 25 °C to 30 °C raised hatching slightly (84% to 88%). If temperature has a similar effect at 40 g/L, hatching there should rise a little above 70%. This is a reasonable prediction, not a certainty, because no dish combined 40 g/L with 30 °C.

Key Takeaways

• Enhanced ACT Science: optional, 40 questions in 40 minutes, 4 choices, no penalty for guessing. Budget about 1 minute per question, reading included.
• Read a passage in about 90 seconds: goal, IV/DV/constants for each experiment, table headings and trends.
• Do fast questions first; cap any single question at about 2 minutes.
• Mean values from several trials are more reliable than single trials; a factor that changes along with the IV is a confound.
• After each timed set, trace every miss to its skill (variables, design, reading experiments, viewpoints, conclusions, predictions) and review that Part.