Opening subject page...
Loading your content
Master the skill of using experimental evidence to support, refute, or refine scientific claims on the ACT.
Science is not just about collecting data — it is about interpreting that data to arrive at meaningful conclusions. Since the earliest days of systematic observation, scientists have wrestled with a fundamental question: How do we know when the evidence actually supports a claim? The ability to draw conclusions from experiments and to evaluate claims made by other researchers is one of the most critical thinking skills tested on the ACT Science section. Roughly 20–25% of ACT Science questions fall under the category of Evaluation of Models, Inferences, and Experimental Results, making this skill essential for a competitive score.
On the ACT, you won't need to memorize the history of scientific reasoning. However, understanding that conclusions must be grounded in evidence — and that claims can be supported, weakened, or left unresolved by data — is exactly the mindset the test rewards. The core question the ACT keeps asking, in many different forms, is this: Does the evidence actually justify this conclusion?
Before diving into ACT-specific strategies, you need to understand five foundational ideas that underpin every question about conclusions and claims. These principles appear again and again across data representation passages, research summaries, and conflicting viewpoints passages.
The diagram below illustrates the decision-making process you should follow every time the ACT asks you to draw a conclusion or evaluate a claim. Starting from the experimental data, you work through a series of logical checkpoints before arriving at a justified conclusion. This framework applies to all three ACT Science passage types: Data Representation, Research Summaries, and Conflicting Viewpoints.
Notice that the middle branch — Claim Needs Modification — is especially important on the ACT. Many questions present claims that are mostly right but slightly too broad or slightly too narrow. The test loves to include answer choices that overstate the evidence. When you see a claim that is close but not exactly right, look for an answer that narrows or refines it rather than accepting or rejecting it entirely.
Although the ACT Science section does not test math heavily, there is a logical framework you can apply systematically. Think of it as a set of reasoning rules rather than formulas. Each rule helps you avoid a specific type of error that the test writers build into incorrect answer choices.
When a question asks whether data supports a conclusion, apply the Support Test: If the conclusion predicts a specific pattern, does the data show that pattern? For example, if the claim states "increasing temperature increases reaction rate," the data should show higher rates at higher temperatures. If even one data point clearly violates this pattern without explanation, the claim is weakened.
The Scope Test asks: Does the conclusion stay within the boundaries of the experiment? A study that only tested plants in a greenhouse cannot conclude anything about plants in the wild. Watch for words like "all," "always," "never," and "every" — these absolute terms often signal an overextension. ACT Science frequently uses these words in wrong answers.
The Alternative Explanation Test checks whether the conclusion is the only reasonable explanation for the data. If the experiment did not control for a confounding variable, then the observed pattern could have another cause. This test is particularly important in Conflicting Viewpoints passages, where two scientists offer different explanations for the same observations.
The Consistency Test is for questions that present new information and ask how it affects an existing conclusion. If the new data is consistent with the prediction made by the claim, the claim is strengthened. If the new data contradicts the prediction, the claim is weakened. If the new data is irrelevant to the claim, the claim is neither strengthened nor weakened.
The ACT Science section presents conclusion and claim questions in several predictable formats. Recognizing the question type immediately tells you which reasoning rule to apply. The diagram below categorizes these question types and maps them to the strategies discussed earlier.
| Question Type | Typical Wording | What to Do |
|---|---|---|
| Support / Refute | "Do the results of Experiment 2 support the hypothesis that...?" | Compare the specific data trend to the specific prediction in the hypothesis. Look for direct matches or contradictions. |
| Strengthen / Weaken | "Which of the following findings would most weaken Scientist 1's claim?" | Identify the key assumption of the claim. The correct answer will directly challenge that assumption with new evidence. |
| Best Conclusion | "Based on the results, which of the following conclusions is best supported?" | Eliminate answers that overextend, understate, or contradict the data. The correct answer is the most precise match. |
| Predict Outcome | "Based on Scientist 2's model, what would happen if...?" | Follow the model's logic to its natural extension. The correct answer is consistent with the model's reasoning, not necessarily with the data. |
Let's walk through a realistic ACT-style problem step by step. Imagine a passage describes two experiments on plant growth. In Experiment 1, researchers grew tomato plants at five different light intensities (1000, 2000, 3000, 4000, and 5000 lux) and measured average plant height after 30 days. In Experiment 2, they repeated the procedure but added 50 mg/L of fertilizer to each pot. The data tables show that plant height increased with light intensity in both experiments, but the plants with fertilizer were consistently taller at every light level.
Question: A student claims: "Increasing light intensity always causes greater plant growth, and fertilizer has no effect unless light intensity is above 3000 lux." Based on the experimental results, is this claim supported?
Even students who understand the principles of drawing conclusions can lose points by falling into predictable traps. The ACT Science writers design wrong answers to exploit common reasoning errors. The table below compares these pitfalls with the correct approach.
| Common Pitfall | Why It's Wrong | Correct Approach |
|---|---|---|
| Choosing an answer that "sounds scientific" | Impressive-sounding terminology does not make an answer correct. The correct answer must match the data, not just sound plausible. | Always trace the answer back to specific data points in the passage, table, or graph. |
| Confusing correlation with causation | The data may show two variables rising together, but without a controlled experiment, you cannot claim one causes the other. | Unless the experiment specifically isolates the variable, choose language like "is associated with" over "causes." |
| Accepting extreme or absolute answers | Words like "always," "never," "all," and "none" require evidence covering every possible case — which experiments rarely provide. | Prefer moderate, qualified answer choices unless the data truly covers all conditions. |
| Extrapolating beyond the data range | If the experiment tested pH from 4 to 8, you cannot conclude what happens at pH 2 or pH 12. | Check the independent variable range. Valid conclusions apply only within the tested range. |
| Ignoring the control group | Without comparing to a control, you cannot attribute the result to the experimental variable. | Always identify the control condition and use it as your baseline for comparison. |
The skills you develop for the ACT Science section are the same skills used by professional scientists, doctors, engineers, and policy makers every day. Understanding how conclusions are drawn and claims are evaluated prepares you not only for a strong ACT score but for college-level science courses where you will be expected to read, critique, and design experiments.
| Skill on the ACT | Advanced Application |
|---|---|
| Evaluating whether data supports a conclusion | In college biology and chemistry, you will write lab reports where the discussion section requires you to explicitly state whether your data supports your hypothesis and explain why. |
| Identifying scope limitations | Research papers include a "Limitations" section. Understanding scope on the ACT is a direct preview of evaluating the generalizability of research findings. |
| Comparing competing claims | In AP and college courses, you will encounter scientific debates (e.g., nature vs. nurture, competing climate models) where you must evaluate evidence on both sides. |
| Recognizing correlation vs. causation | Statistics courses teach formal methods (randomized controlled trials, regression analysis) to distinguish correlation from causation — the concept you learn here is the conceptual foundation. |
If you continue in science, you will encounter concepts like statistical significance (using p-values to determine if results are meaningful), meta-analysis (combining data from multiple studies to draw stronger conclusions), and Bayesian reasoning (updating the probability of a claim as new evidence is gathered). The ACT does not require you to know any of these formally, but the logical foundation is identical: conclusions must be proportional to evidence.
Test your understanding with these five problems, which increase in difficulty. For each one, try applying the reasoning rules (Support Test, Scope Test, Alternative Explanation Test, Consistency Test) before reading the answer.
Drawing conclusions and evaluating claims are among the most heavily tested skills on the ACT Science section. Every valid conclusion must be directly supported by the data presented, must remain within the scope of the experiment, and must not confuse correlation with causation. When the ACT asks whether evidence supports, weakens, or is irrelevant to a claim, apply the four reasoning rules: the Support Test, the Scope Test, the Alternative Explanation Test, and the Consistency Test.
Remember the key strategies: break complex claims into individual testable parts, watch for absolute words ("always," "never," "all") that signal overextension, prefer the Goldilocks answer that fits the data precisely, and always trace your answer back to specific evidence in the passage. These reasoning skills will serve you well beyond the ACT — in college science courses, in evaluating news claims, and in making informed decisions throughout your life.