ACT Science • Research Summaries

Analyzing Experimental Design

Learn to identify variables, controls, and procedures so you can master the Research Summaries passages on the ACT Science test.

Why Experimental Design Matters

Modern science rests on the idea that claims about the natural world should be tested through carefully designed experiments. This isn't something humans have always done; for most of history, people relied on authority, tradition, or casual observation to explain phenomena. The development of the scientific method—and its emphasis on controlled experimentation—was a revolution that unfolded over centuries.

Understanding this history helps you see why ACT Science passages describe experiments the way they do: every detail about variables, controls, and procedures traces back to hard-won lessons about how to separate real cause-and-effect from coincidence.

1620
Francis Bacon publishes Novum Organum, arguing that knowledge should come from systematic observation and experimentation rather than pure logic or tradition. He championed the idea that scientists must gather data before drawing conclusions.
1747
James Lind conducts one of the first controlled experiments in medical history. He divides sailors suffering from scurvy into groups, gives each group a different dietary supplement, and discovers that citrus fruit cures the disease. This early use of comparison groups set the stage for experimental controls.
1882
Robert Koch develops his famous postulates to prove that a specific microbe causes a specific disease. His work formalized the concept of isolating a single variable—in this case, one type of bacterium—while holding other conditions constant.
1935
Statistician Ronald Fisher publishes The Design of Experiments, introducing rigorous statistical frameworks for experimental design, including randomization and replication. His work made it possible to quantify whether results were due to the tested factor or to chance.
Present Day
The ACT Science test builds directly on these principles. Every Research Summaries passage describes an experiment with identifiable variables, controls, and procedures—and the questions ask you to analyze and evaluate the design, exactly as a working scientist would.

The core question that experimental design answers is deceptively simple: How can we be sure that X actually causes Y? Every feature of a good experiment—independent variables, dependent variables, controls, repeated trials—exists to answer that question with confidence. On the ACT, your job is to recognize these features within the passage and use them to answer questions.

Core Principles of Experimental Design

Before you can analyze an experiment on the ACT, you need a clear vocabulary for the parts that make up any well-designed study. These five concepts appear—sometimes explicitly, sometimes implicitly—in every Research Summaries passage you'll encounter.

1

Independent Variable

The factor the experimenter deliberately changes between trials or groups. On the ACT, look for what's different across experiments. It's the "input" being tested.
2

Dependent Variable

The outcome being measured or observed. It's the "output" that responds to changes in the independent variable. Look for data tables, graphs, or recorded measurements.
3

Controlled Variables (Constants)

All factors kept the same across trials so that any change in the dependent variable can be attributed to the independent variable. These are often listed in the experimental setup.
4

Control Group

A baseline group that receives no treatment or a standard treatment. Comparing experimental groups to the control tells you whether the independent variable had an effect.
5

Replication & Sample Size

Repeating trials and using adequate sample sizes increases confidence that results are reliable, not due to random variation. ACT questions may ask why an experiment used multiple trials.
Key Takeaway
Think of an experiment like a recipe test in a kitchen. If you want to know whether adding vanilla extract improves cookies, you bake one batch with vanilla (experimental group) and one batch without (control group), keeping everything else—oven temperature, baking time, flour amount—exactly the same (controlled variables). The vanilla is your independent variable, and the taste rating is your dependent variable. If you only change one thing at a time, you can trust that any taste difference is actually caused by the vanilla.

Anatomy of an Experiment

The diagram below maps out how a typical experiment flows from hypothesis to conclusion. When you read a Research Summaries passage, try to mentally slot each piece of information into this framework. Knowing where each detail fits helps you answer questions quickly and accurately.

Figure 1 — The flow of a controlled experiment, from hypothesis through data comparison to conclusion.

Notice how controlled variables sit between the two groups: they represent everything that stays the same. The only intended difference between the experimental and control groups is the independent variable. Both groups produce data for the dependent variable, and comparing those data sets is what allows the researcher—and you, on the ACT—to determine whether the independent variable had a real effect.

How to Read an ACT Experiment

Research Summaries passages on the ACT typically describe two or three experiments performed by scientists investigating a question. Each experiment tweaks one aspect of the setup while keeping other aspects the same. Your job is to decode this structure quickly. Here is a step-by-step process you can follow every time.

Strategy Framework
1
Step 1 — Skim the introduction.Identify the overall research question. What phenomenon is being investigated?
2
Step 2 — For each experiment, identify the independent variable.Ask: "What did the scientists change or manipulate this time?"
3
Step 3 — Identify the dependent variable.Ask: "What did they measure or record?" Look for tables, graphs, and data descriptions.
4
Step 4 — Identify the controlled variables.Ask: "What stayed the same across trials?" This is often stated in the setup paragraph.
5
Step 5 — Note any control groups.Is there a trial with no treatment, or a baseline condition used for comparison?

ACT questions about experimental design fall into several recurring categories. Some ask you to identify the purpose of a specific procedure ("Why did the scientists use a water bath at constant temperature?"). Others ask what would happen if a variable changed, or how the experiment could be improved. A few ask you to compare the designs of Experiment 1 and Experiment 2 to identify what's different.

The key insight is that most of these questions boil down to understanding which variable is which. If you can clearly label the independent variable, dependent variable, and constants, the answer choices usually become straightforward. Don't rush past the experimental descriptions—those two or three paragraphs are the most important text in the entire passage.

Key Takeaway
Reading an ACT experiment is like being a detective at a crime scene: you're looking for what changed (the suspect), what resulted (the evidence), and what stayed the same (the background). The passage gives you all the clues—you just have to organize them. Train yourself to annotate the passage by circling or underlining the independent variable, the dependent variable, and any controls before you even look at the questions.

Common ACT Question Types for Experimental Design

ACT Science questions that test your understanding of experimental design can be grouped into five main types. Recognizing the type helps you know exactly where in the passage to look for the answer. The diagram below maps these question types to the parts of the experiment they address.

Figure 2 — Five common ACT question types and how they relate to the experiment described in the passage.
Question TypeWhat It AsksWhere to Look
Variable IdentificationWhich factor is the independent/dependent variable?Experiment descriptions and data tables
Purpose of ProcedureWhy was a specific step performed?Setup paragraphs—look for controlled variables
Predicting OutcomesWhat would happen if a variable were changed?Trends in data tables and graphs
Improving the DesignWhat additional trial or modification would strengthen the study?Gaps in the range of independent variable values tested
Comparing ExperimentsHow does one experiment differ from another?Compare the setup paragraphs of each experiment side-by-side

When you encounter a question, first classify it into one of these five types. This tells you exactly which part of the passage holds the answer. For variable identification, look at what's changing and what's being measured. For purpose questions, think about what would go wrong if that procedure were skipped—if the step controls a variable, its purpose is to keep conditions consistent. For prediction questions, extend the trend you see in the data. For improvement questions, look for missing data points or uncontrolled variables. For comparison questions, line up the two experiments and spot the one thing that differs.

Worked Example

Let's walk through a sample ACT-style Research Summaries passage and question. Read the passage carefully, then follow the step-by-step analysis.

Sample Passage (Excerpt)
A group of students investigated how the concentration of salt (NaCl) in water affects the boiling point of the solution. In each trial, they added a measured amount of NaCl to 500 mL of distilled water in an identical glass beaker and heated it on the same hot plate set to "High." They recorded the temperature at which the solution began to boil.
TrialNaCl Added (g)Boiling Point (°C)
10100.0
220100.5
340101.0
460101.5
580102.1

Question: The students used the same hot plate setting and the same volume of distilled water in each trial. The most likely reason for keeping these factors constant was to ensure that:

  • A. the boiling point would always be exactly 100°C.
  • B. any observed change in boiling point could be attributed to the amount of NaCl.
  • C. the NaCl would dissolve faster in each successive trial.
  • D. the experiment would take the same amount of time for each trial.
Step-by-Step Solution
1
Step 1 — Classify the QuestionThis is a Purpose of Procedure question. It asks why the scientists (students) kept certain factors constant. We need to think about controlled variables.
2
Step 2 — Identify the VariablesThe independent variable is the amount of NaCl added (it changes from 0 g to 80 g). The dependent variable is the boiling point (it's what they measured). The controlled variables include the hot plate setting, the volume of water, and the type of beaker.
3
Step 3 — Reason About the Purpose of ControlsControlled variables exist so that the experiment is a "fair test." If the students changed the water volume and the salt amount at the same time, they wouldn't know which change caused the boiling point to shift. By keeping everything else constant, the students isolate the effect of salt concentration on boiling point.
4
Step 4 — Evaluate the Answer ChoicesChoice A is wrong—the boiling point clearly changes (that's the whole point of the experiment). Choice C is irrelevant to the purpose of controls. Choice D is about time, which isn't the focus. Choice B correctly states that holding other factors constant allows you to attribute changes in the dependent variable (boiling point) to the independent variable (NaCl amount).
5
Final AnswerThe correct answer is B. Keeping the hot plate setting and water volume constant ensures that any observed change in boiling point can be confidently attributed to the amount of NaCl added.

Strengths & Limitations of Experimental Designs

On the ACT, you may be asked to identify flaws in an experiment or suggest improvements. Not every experiment described in a passage is perfect—sometimes the questions test whether you can recognize what's missing or what could be done better. Understanding common strengths and limitations prepares you for these questions.

Design FeatureStrengthLimitation
Single variable testedClear cause-and-effect relationship establishedMay oversimplify complex systems where multiple factors interact
Control group includedProvides a baseline for comparisonSome phenomena don't have an obvious "no treatment" condition
Multiple trials / replicationReduces impact of random error; increases reliabilityMore time-consuming and resource-intensive
Wide range of independent variable valuesReveals trends and patterns across conditionsMay miss effects at untested intermediate values
Small sample sizeFaster and cheaper to conductResults may not be representative; hard to detect small effects

A common ACT question format is: "Which of the following changes to the experiment would most likely improve the reliability of the results?" The answer usually involves increasing sample size, adding more trials, testing additional values of the independent variable, or adding a control group that was previously missing. When you see this type of question, think about what's not in the current design that a careful scientist would want.

Key Takeaway
Think of an experimental design like a chain—it's only as strong as its weakest link. A beautifully controlled experiment with only one trial is unreliable because that single result might be a fluke. An experiment with dozens of trials but no control group leaves you wondering whether the treatment actually did anything. Good design requires all the links: clear variables, proper controls, adequate replication, and a meaningful range of tested values. ACT questions often ask you to spot the weak link.

Beyond the ACT: How Scientists Really Design Studies

The experimental design principles you've learned for the ACT are the foundation for much more sophisticated methods used in real scientific research. If you continue into college-level science courses, you'll encounter these advanced approaches that build on the same ideas.

ACT-Level ConceptAdvanced Extension
One independent variable at a timeFactorial design — testing multiple independent variables simultaneously to study interactions between them
Control group for comparisonDouble-blind studies — neither the subjects nor the researchers know who is in the control group, eliminating bias
Multiple trials for reliabilityStatistical significance testing — using math (like p-values) to determine whether results are likely due to the independent variable or to chance
Identifying confounding variablesRandomized controlled trials (RCTs) — randomly assigning subjects to groups to eliminate hidden confounding variables
Comparing two experimentsMeta-analysis — combining data from many studies to draw broader conclusions and resolve conflicting results

You don't need to know these advanced methods for the ACT, but understanding that they exist helps you appreciate why the test emphasizes the basics. If you can identify variables, recognize controls, and evaluate an experimental setup on the ACT, you already have the core scientific literacy that more advanced coursework will build upon. These are the same thinking skills that medical researchers, engineers, ecologists, and social scientists use every day.

Practice Problems

Test your understanding with these five problems. They progress from straightforward to more challenging, mirroring the kinds of questions you'll see on the ACT Science section. Try to answer each one before revealing the explanation.

PROBLEM 1CONCEPTUAL
In an experiment studying the effect of light intensity on plant growth, a student placed plants at different distances from a lamp and measured their height after 2 weeks. All plants were given the same amount of water, planted in the same type of soil, and kept at the same temperature. In this experiment, the dependent variable is: A. the type of soil. B. the distance from the lamp. C. the height of the plants after 2 weeks. D. the amount of water given to each plant.
PROBLEM 2IDENTIFICATION
A researcher studied how different fertilizer concentrations affect tomato yield. She set up five groups of tomato plants, each receiving a different concentration of fertilizer (0%, 5%, 10%, 15%, and 20%). The group receiving 0% fertilizer served as the: F. independent variable. G. dependent variable. H. control group. J. experimental error.
PROBLEM 3INTERMEDIATE
Students conducted two experiments to test how temperature affects the rate at which sugar dissolves in water. In Experiment 1, they dissolved sugar in water at temperatures of 20°C, 40°C, 60°C, and 80°C, and measured how long it took for 10 g of sugar to fully dissolve. In Experiment 2, they repeated the same procedure but used salt instead of sugar. The primary reason the students performed Experiment 2 was most likely to determine whether: A. salt dissolves faster than sugar at all temperatures. B. the relationship between temperature and dissolving rate depends on the type of solute. C. water temperature has no effect on dissolving rate. D. the amount of solute used was sufficient for the experiment.
PROBLEM 4APPLIED
A biologist measured the heart rates of 10 mice before and after administering Drug X. She found that the average heart rate decreased from 650 beats per minute (bpm) to 590 bpm after the drug was administered. A classmate reviewed the experiment and argued that the results were inconclusive. Which of the following design improvements would most directly address the classmate's concern? F. Using mice of different ages in each trial. G. Measuring the heart rates at a different time of day. H. Including a second group of 10 mice that receives a placebo (inactive substance) instead of Drug X. J. Increasing the dose of Drug X administered to the mice.
PROBLEM 5SYNTHESIS
Consider the following two experiments from a Research Summaries passage: Experiment 1: Scientists measured the oxygen consumption of goldfish at water temperatures of 15°C, 20°C, 25°C, and 30°C. Each trial used 5 goldfish in a 10-liter tank with aerated, dechlorinated water. Experiment 2: The same scientists repeated the procedure from Experiment 1 but used guppies instead of goldfish. A student claims: "Experiment 2 proves that temperature affects all fish species the same way." Is this claim supported by the experimental design? A. Yes, because both experiments used the same temperatures and procedures. B. Yes, because the scientists controlled for water volume and aeration. C. No, because only two species were tested, which is insufficient to generalize to all fish species. D. No, because the experiments did not include a control group at room temperature.

Lesson Summary

Analyzing experimental design is one of the most valuable skills you can develop for the ACT Science section, and it applies to nearly every Research Summaries passage you'll encounter. The foundation rests on five core concepts: the independent variable (what the experimenter deliberately changes), the dependent variable (what gets measured), controlled variables (everything kept constant to ensure a fair test), the control group (the baseline for comparison), and replication (repeated trials to improve reliability). When reading a passage, your first task is to identify these elements—circle the independent variable, underline the dependent variable, and note the constants.

ACT questions about experimental design typically fall into five categories: variable identification, purpose of procedure, predicting outcomes, improving the design, and comparing experiments. Classifying the question type before looking at answer choices will help you focus on the right part of the passage. Remember that the purpose of any controlled variable is to isolate the effect of the independent variable, that a missing control group is a common design weakness, and that conclusions should never extend beyond what the data actually support. Master these principles, and experimental design questions become some of the most predictable—and scoreable—points on the entire ACT.

Varsity Tutors • ACT Science • Analyzing Experimental Design