ACT guideStudy & preparation
ACT® Science Practice: Read Data, Check Experiments and Compare Claims
Short answer
ACT Science practice should train three skills ACT identifies: interpreting data, understanding scientific investigations, and evaluating arguments or models with evidence. The standard section has 40 questions in 40 minutes, and calculators are not permitted. Work through a question without adding assumptions, explain why the alternatives fail, and use the mistake to choose your next practice task.
Which reasoning skills does ACT Science test?
Build ACT Science practice around the three skills in ACT's science description: interpreting data, understanding investigations, and evaluating explanations against evidence. ACT's 2026–2027 preparation booklet lists 40 questions in 40 standard minutes, and its science description prohibits calculators; both sources were checked November 4, 2026.
ACT gives these approximate content-category ranges:
| Skill category | Approximate share | What to practice |
|---|---|---|
| Interpretation of Data | 38–50% | Find the right values, units, comparisons, and relationships |
| Scientific Investigation | 18–32% | Identify what changes, what is measured, and what stays constant |
| Evaluating Scientific Arguments/Models with Evidence | 24–38% | Compare predictions with the observations supplied |
The ranges describe reasoning skills, not percentages assigned to biology, chemistry, earth and space science, or physics. ACT uses those subjects as contexts. Introductory science knowledge may be needed; advanced knowledge is not required. If unfamiliar terminology blocks your understanding during review, identify that gap alongside the reasoning error.
The three exercises below are original teaching examples, not official ACT questions. Their data and situations are invented. They illustrate methods without measuring readiness or covering every question type. Try each before reading its explanation.
If you are still deciding whether to select Science, resolve that first with the ACT optional science guide. The work here starts after that decision.
How do I read a data display without adding assumptions?
Start by naming what the question asks you to compare. Then locate the relevant row or axis, confirm the units, and perform only the requested comparison.
Original exercise 1: read an interval. An invented experiment records these measurements:
| Elapsed time (minutes) | Measured volume (mL) |
|---|---|
| 2 | 5 |
| 4 | 9 |
| 6 | 11 |
How much does the measured volume increase from 4 to 6 minutes?
- A. 2 mL
- B. 4 mL
- C. 6 mL
- D. 11 mL
Answer: A, 2 mL. The relevant measurements are 9 mL at 4 minutes and 11 mL at 6 minutes. Subtract the earlier volume from the later volume: 11 − 9 = 2 mL.
Each alternative points to a different reading mistake:
| Answer | Why it fails |
|---|---|
| 4 mL | This is the increase from 2 to 4 minutes, the wrong interval |
| 6 mL | This is the increase from 2 to 6 minutes, a wider interval |
| 11 mL | This is the final volume, not the increase |
The table does not give a measurement at 8 minutes. An answer about that time would need additional information or an explicitly supplied model. Do not silently extend the latest increase into a scientific rule.
For a follow-up task, use another table or graph and write the requested interval and units before calculating. On a graph, read the axis labels and scale first. If you selected D, practice distinguishing a final value from a change; if you selected B or C, practice finding the exact comparison.
When deciding whether you want another preparation resource, check the availability of full ACT prep, planned for launch, against your study needs. Use the error you identified to make that comparison specific.
How do I identify variables and experimental controls?
Separate three questions: What does the experiment change? What does it measure? What conditions are held constant? A condition kept equal is not the factor being tested.
Original exercise 2: inspect the design. Two groups of the same plant species receive equal water and soil and are kept at the same temperature for the same study duration. One group receives 4 hours of light per day; the other receives 8. The recorded outcome is average height change. No results are supplied.
Which factor is deliberately changed between the groups?
- A. Amount of water
- B. Daily light duration
- C. Average height change
- D. Plant species
Answer: B, daily light duration. Water and species stay the same. Average height change is the measured outcome; it is not the condition deliberately assigned to differ between groups.
Now suppose a second design gives the 8-hour group more water as well. That comparison changes both light and water, so it cannot isolate the effect of light. Calling water a “controlled condition” would be incorrect in that second design.
Notice the limit of the original setup: equal water helps describe the comparison, but it does not prove that more light caused greater growth. No growth measurements were given. The exercise concerns experimental reasoning, not a claim about how real plants grow.
For your next investigation question, write three short labels: changed factor, measured outcome, conditions held constant. If two conditions differ, mark both before choosing an answer about what the study can establish. If you confused the outcome with the changed factor, repeat that labeling task on a different experiment before adding time pressure.
How do I compare competing explanations?
Translate each explanation into a prediction, then ask whether the supplied observation distinguishes those predictions. A result that fits one model better in a stated comparison does not prove that model works in every situation.
Original exercise 3: compare models. Two invented models describe a sensor:
| Model | Prediction |
|---|---|
| Model A | The reading changes with light intensity, but not with temperature |
| Model B | The reading changes with temperature, but not with light intensity |
In the supplied comparison, temperature is held constant. Light intensity changes, and the sensor reading changes.
Which conclusion is supported?
- A. The observation supports Model A over Model B under their stated predictions.
- B. The observation supports Model B because the reading changes.
- C. Both models predict this result because both mention changes.
- D. Model A is proved correct for all temperatures and light intensities.
Answer: A. Model A predicts a response to changing light. Model B predicts no response to changing light when temperature stays fixed. The observation therefore favors A in this comparison.
B ignores which factor changed. C matches the word “changes” without comparing the actual predictions. D extends one observation into a universal conclusion. These are different errors, so record the one you made.
To investigate the temperature prediction next, you could propose holding light constant and changing temperature. That would be a new experiment, not an observation already provided. Keep proposed evidence separate from recorded evidence.
On the next competing-explanations question, make a compact note for each model: factor → predicted response. Then connect the observation to the relevant prediction. If you want additional preparation after identifying that need, review full ACT prep, planned for launch, when comparing your next study options.
What should my error log change before the next timed section?
Use each error to assign a concrete task. This is an editorial review method, not an ACT scoring rubric:
| Error you found | Record this detail | Next task |
|---|---|---|
| Wrong axis, units, or interval | The label or comparison you missed | Locate values and state units before calculating on a new display |
| Changed factor confused with a control | Which condition differs and which stays equal | Label the design before answering |
| Two changing conditions treated as one | The additional factor you overlooked | Explain why the comparison cannot isolate the proposed factor |
| Model chosen by a shared word | The model's actual prediction | Match each prediction to the observation |
| Conclusion goes beyond the evidence | The extra assumption your answer needs | State what the supplied evidence supports and what remains untested |
Also review correct guesses and unfinished questions. Record whether the obstacle was understanding the science context, locating evidence, making the comparison, or managing time. A label such as “bad at science” does not tell you what to practice next.
After targeted review, choose an unused Science section from ACT's free official practice resources, checked November 4, 2026. Use a current form with 40 questions and the standard 40-minute limit, without a calculator, and keep its matching key and scoring instructions. The ACT practice-test guide explains form selection and scoring in detail.
Complete the attempt before checking answers. Then explain the missed questions and correct guesses before opening another fresh section. Compare whether the same reasoning errors recur, not just whether the total changes. Repeating a remembered exercise can rehearse a method, but it is not a fresh performance check.
Your next step is to choose one error from this workshop, complete a different example of that skill, and reserve time for an official section plus review. The ACT study plan for 2027 shows how those sessions can fit into a longer schedule. If you are considering more preparation, check whether full ACT prep, planned for launch, is available for that schedule.
Trust the source
Official sources
Exam policies can change. Use these primary sources for the most current details.