01 / 21

Evaluating & Modeling
Experiments

Master the art of analyzing experimental design, identifying variables, and interpreting data the way the ACT Science section demands.

ACT Science

What Is a Scientific Experiment?

An experiment is a controlled procedure designed to test a hypothesis. On the ACT, you will read passages describing experiments and must evaluate how well the design supports the conclusions.

Hypothesis
A testable prediction about how one variable affects another.
Procedure
The step-by-step method used to collect data in the experiment.
Conclusion
The claim made based on the results, which you must evaluate.

The Three Types of Variables

Independent What you change (manipulate) Dependent What you measure (observe) Controlled What stays the same (constants) EXPERIMENTAL VARIABLES

Why This Skill Matters on the ACT

About 6-8 questions per ACT Science section directly test your ability to evaluate experimental design. These questions ask you to identify flaws, predict outcomes of modified experiments, and determine which conclusions are supported by the data.

15-20%
of ACT Science questions focus on evaluating experimental models and design.
Passages 4-6
Research Summaries passages are where these questions appear most often.

The Control Group Rule

Control Group No treatment applied Baseline for comparison No variable changed Experimental Group Treatment applied Results compared to control Variable changed vs

Correlation vs Causation

Correlation Causation
Definition Two variables change together One variable directly causes the other to change
Proof needed Only a pattern in data Controlled experiment with isolated variable
Example Ice cream sales and drowning both rise in summer Fertilizer X causes 20% more plant growth vs control
ACT phrasing "is associated with" "is the cause of" / "results in"

How to Read an Experiment Passage

1
Identify the hypothesis. What claim is being tested? Look for the purpose statement.
2
Find the independent variable. What did the scientists deliberately change between groups or trials?
3
Find the dependent variable. What was measured or observed as the outcome?
4
Check the controls. Was there a control group? Were other variables held constant?
5
Evaluate the conclusion. Does the data actually support the claim? Are there alternative explanations?

Worked Example: Plant Growth

A student tested whether light color affects plant height. She grew 3 groups of 10 plants each under red, blue, or white light for 30 days, measuring height weekly. All plants received the same water, soil, and temperature.

0 5 10 15 20 Height (cm) Week 1 Week 2 Week 3 Week 4 Red Blue White

Worked Example: Finding Design Flaws

A researcher tested whether a new drug lowers blood pressure. She gave the drug to 50 patients and measured their blood pressure after 4 weeks. All 50 patients showed lower blood pressure, so she concluded the drug works.

Missing Control
No placebo group to compare against the treatment group.
Fix
Add a group of 50 patients receiving a placebo under the same conditions.

Worked Example: Modifying the Design

In the plant growth experiment, a student wants to determine whether light intensity (not just color) affects growth. How should the experiment be modified?

1
Change the independent variable from light color to light intensity (e.g., low, medium, high brightness).
2
Keep the same dependent variable: plant height measured weekly over 30 days.
3
Add new controls: all groups must now receive the same color of light (e.g., all white) so only intensity varies.
4
Include a control group: plants grown in normal ambient light for comparison.

Sample Size & Repeatability

Two critical factors determine whether experimental results are reliable and valid:

Sample Size
Larger samples reduce the impact of random variation. Testing 5 plants is weaker evidence than testing 50.
Repeatability
If other researchers can repeat the experiment and get similar results, the findings are more credible.

Reading Tables & Graphs

1
Read the title and axis labels first. Know what is being measured before looking at numbers.
2
Identify trends. Is the data increasing, decreasing, or staying constant? Look for patterns.
3
Compare groups. How does the experimental group differ from the control at each data point?
4
Watch for outliers. Does one data point break the pattern? The ACT may ask about it.

Interpolation vs Extrapolation

Time (days) Growth (cm) MEASURED DATA Interpolate Extrapolate Day 0 Day 28
Interpolation
Estimating a value within the range of measured data. More reliable.
Extrapolation
Predicting a value beyond measured data. Less reliable: the trend may change.

Modeling with Equations

Sometimes the ACT presents a mathematical model that describes experimental data. You may be asked to evaluate whether the model fits, predict a value, or identify which model best represents the data.

Linear
\( y = mx + b \): Constant rate of change. Straight-line graph.
Exponential
\( y = a \cdot b^x \): Rapid growth or decay. Curved graph.
Inverse
\( y = \frac{k}{x} \): As one variable increases, the other decreases.

Common Mistakes to Avoid

✕ Confusing Correlation with Causation
Just because two variables change together does not mean one causes the other. Look for a controlled experiment.
✕ Ignoring Controlled Variables
If multiple variables changed between groups, you cannot isolate which one caused the effect.
✕ Over-Extrapolating
Extending a trend far beyond the data range is unreliable. The ACT often tests whether a conclusion is supported by actual data.

Summary of Key Concepts

Concept What to Look For ACT Question Type
Independent Variable What was deliberately changed "Which variable was manipulated?"
Dependent Variable What was measured as outcome "What was the measured result?"
Control Group Group with no treatment "What served as the control?"
Design Flaw Missing control, confounding variables "Which weakens the conclusion?"
Model Fit Does the equation match the data trend? "Which model best represents...?"

Question 1 of 5

A scientist tested whether temperature affects enzyme reaction rate by running reactions at \(20^\circ\text{C}\), \(30^\circ\text{C}\), \(40^\circ\text{C}\), and \(50^\circ\text{C}\) while keeping pH and substrate concentration constant. What is the independent variable?

Question 2 of 5

A study found that students who eat breakfast score higher on tests. The researchers concluded that eating breakfast causes better test scores. What is the flaw in this conclusion?

Question 3 of 5

Which best describes the problem with the student's claim?

Question 4 of 5

In an experiment testing a new fertilizer, Group A (50 plants) received the fertilizer and Group B (50 plants) received only water. After 6 weeks, Group A grew an average of 12 cm taller. What would most strengthen the conclusion that the fertilizer caused the growth difference?

Question 5 of 5

A population of bacteria doubles every hour. At \( t = 0 \), there are 100 bacteria. Which model best represents the bacteria count \( N \) at time \( t \) hours?