ACT Science: Causation vs Correlation - Spot the Trap Every Time

Correlation means two variables happen together (if one goes up, the other goes up). Causation means one variable causes the other to change. Correlation does NOT imply causation. Example: ice cream sales correlate with drowning deaths (both rise in summer). But ice cream doesn't cause drowning; warm weather causes both. ACT Science tests this distinction to catch students who assume correlation is proof of causation. The trick: a study can show perfect correlation without proving causation, and the ACT will offer answer choices mixing causation language with correlation data.

Example experiment: "Students who drink coffee score higher on tests" shows correlation. But the data doesn't prove coffee causes higher scores; smarter students might drink more coffee, or students who study late drink coffee. The correlation exists, but causation is unproven. Any answer choice claiming "Coffee causes improved test scores" would be wrong; correct language: "Students who drink coffee tend to score higher" (correlation, not causation).

Trap 1: Confusing "happens after" with "causes." "In Trial 1, pH increased, and then reaction rate increased." This suggests pH change caused the reaction change. But the data only shows sequence, not causation. Trap 2: Ignoring alternative explanations. "As temperature rose, the enzyme's activity rose; therefore, temperature causes activity." But maybe the enzyme is more stable at higher temperatures, and stability (not temperature directly) causes activity. Trap 3: Assuming the controlled variable causes the measured variable without proof. A study changes light exposure and measures plant growth. This design shows correlation only; to prove causation, you'd need to rule out other factors (water, soil, genetics). ACT answer choices offer causation language even when the data only supports correlation.

Fix: when you see correlation data, watch for answer choices claiming "causes" or "results in." These are usually wrong. Correct language uses "correlates with," "tends to," or "is associated with." If the question directly asks about causation, check whether the study is designed to prove it (control of variables, elimination of alternatives). If not, correlation is all the data supports.

Study 1: "Students who exercise regularly have higher GPA." (Correlation only; we can't say exercise causes grades without ruling out confounding factors like motivation.) Study 2: "Experiment: Researchers held temperature constant and varied pH while measuring enzyme activity. Result: pH correlates with activity." (This design isolates pH, suggesting pH contributes to activity, but causation is still not proven without mechanism.) Study 3: "Survey: People who drink 2+ liters of water daily report better sleep." (Correlation only; self-reported data and no controlled variables mean causation is unproven.) In all three cases, the data shows correlation; none prove causation without additional experimental controls.

For each study, identify: What variable did the researchers intentionally change? What did they measure? Are other factors controlled? If all three are clear, causation is more likely. If factors are uncontrolled or confounded, it's correlation only.

Roughly 30% of ACT Science questions about experimental results ask about causation vs. correlation. The test deliberately mixes causation-sounding language with correlation-only data to trap students. If you understand the distinction, you'll avoid the trap and select correct answers that match the actual evidence. This is a reasoning skill, not a memorization skill; it transfers across all experiment questions on the ACT.

Before you answer any question asking about what caused a result, pause and ask: "Does the data prove causation, or only show correlation?" This habit, practiced once per science section, will boost your accuracy on inference questions by 2-3 points per test.