SAT Reading Strategy

Scientific Method & Study Design on the SAT

How to Evaluate Research Like a Pro

Track the author’s logic, locate evidence quickly, and sharpen your reasoning.

9 Min Read

Reading Skill

Evidence-First

5 Practice Qs

Anchor every answer in the exact line that proves it. If you cannot point to the words, it is not the answer.

Read the question, then scan for the line that directly supports a choice.
Match wording, not vibe: synonyms are fine, new ideas are not.
If two answers feel close, eliminate the one with any extra claim.

When the SAT presents science and social science passages, it tests your ability to evaluate study design rather than your background knowledge of the subject. You are expected to think like a researcher: read a description of a study and judge whether the conclusions actually follow from the evidence. This is one of the most valuable skills the SAT measures, and it is also one of the most learnable.

If you've already worked through our Data Interpretation guide, you know how to read graphs and pull numbers from tables. This post goes one level deeper: instead of asking "What does the data show?" we're asking "What can we actually conclude from this study, and what can't we?" That distinction is where the SAT separates students who skim from students who score.

The good news? Study design questions follow predictable patterns. Once you learn to recognize the five key concepts and the four trap answer types the SAT relies on, these questions become some of the most straightforward on the test. Let's break it all down.

What Is This Skill?

Evaluating study design means understanding how a piece of research was conducted and what that method allows (or doesn't allow) the researchers to claim. On the SAT, this boils down to five concepts you need to recognize on sight.

1. Experimental vs. Observational Studies

This is the single most important distinction. In an experiment, the researchers actively change something, they assign participants to different groups, give one group a treatment and the other a placebo, or manipulate a variable on purpose. In an observational study, researchers simply watch and record what happens without intervening. They survey people, track outcomes over time, or compare groups that already exist.

Why does this matter? Because only experiments can establish cause and effect. If researchers just observed that people who eat breakfast tend to have lower rates of heart disease, they can't conclude that breakfast causes better heart health, maybe health-conscious people are simply more likely to eat breakfast in the first place.

2. Control Groups

A control group is the baseline, the group that doesn't receive the treatment or intervention. Without a control group, there's no way to know whether the observed result was caused by the treatment or by something else entirely (time passing, placebo effect, seasonal changes, you name it). When you read a study on the SAT, one of your first questions should be: "Was there a control group, and what did it look like?"

3. Sample vs. Population

The sample is the specific group of people (or animals, or plants) actually studied. The population is the larger group the researchers want to generalize to. A study of 200 college freshmen at one university tells you about those 200 students, extending those findings to "all young adults" or "all college students" is a leap that the data may not support. The SAT tests whether you can spot that leap.

4. Correlation vs. Causation

This is the concept most frequently exploited by SAT trap answers. Correlation means two things tend to occur together, when one goes up, the other goes up (or down). Causation means one thing directly produces the other. Ice cream sales and drowning rates are correlated (both rise in summer), but ice cream doesn't cause drowning. On the SAT, if a passage describes an observational study, any answer choice that uses causal language ("leads to," "results in," "causes") is almost certainly wrong.

"Strong pattern" does not mean "strong study design." A survey might find a dramatic correlation, say, students who sleep 8+ hours have GPAs a full point higher. That's a striking finding, but if the study is observational, no amount of statistical strength allows a causal conclusion.
"Did not prove" is not the same as "disproved." If a study fails to find a link between two variables, that doesn't mean the link doesn't exist, it means this particular study didn't detect it. The SAT will offer answer choices that confuse "no evidence found" with "evidence against," and that's a trap.

How the SAT Tests This Skill

Study design questions tend to arrive through a small number of predictable question stems. Recognizing these stems instantly tells you what the question is really asking:

"Which conclusion is best supported by the study described in the text?"
"Which finding, if true, would most weaken the researchers' conclusion?"
"Based on the study design, which of the following is a limitation of the researchers' findings?"
"The study's results most directly support which of the following claims?"

The Four Trap Answer Types

The SAT builds its wrong answers from the same playbook. Learn to name these traps and you'll reject them on instinct:

The Causation Jump: The passage describes an observational study (no manipulation, no random assignment), but the answer choice uses causal language. If the study only observed, the conclusion can only describe an association, not a cause.
The Overgeneralization: The study examined a specific group (say, 150 eighth-graders in rural Ohio), but the answer choice extends the findings to "all students" or "people in general." Watch for answer choices that drop the qualifiers.
The Reversed Direction: The passage says Variable A is associated with Variable B, but the answer choice flips it, claiming B causes A. On the SAT, directionality matters, if the study tracked sleep habits and then measured GPA, the answer shouldn't claim that GPA affects sleep (unless the passage explicitly says so).
The Absolute Language Trap: Words like always, never, proves, all, none, guarantees are red flags in answer choices. Scientific studies deal in probabilities and tendencies, not absolutes. The correct answer almost always uses hedged language: suggests, is associated with, tends to, may indicate.

Here's a five-step method you can apply to every study design question. It takes seconds once you've practiced it, and it keeps you from falling into the traps above.

Identify the study type. As you read, ask yourself: Did the researchers do something to the participants (experiment), or did they observe and record (observational)? This single determination controls which conclusions are valid.
Checkpoint: Can I label this study in one word, "experiment" or "observation"? If not, reread the passage.
Find the variables. What was measured? What was compared? In an experiment, identify the variable being manipulated (independent) and the outcome being measured (dependent). In an observational study, identify what was tracked and what was correlated.
Check for a control group. Was there a group that didn't receive the treatment? If not, the study's ability to draw conclusions is significantly limited. If the passage mentions a control group, note what it looked like, was it well-matched to the treatment group?
Checkpoint: If I can't find a control group, can the study prove anything was caused by the treatment?
Match the conclusion to the design. This is where most students go wrong. The correct answer will never overclaim. If the study is observational, the conclusion will mention association, not causation. If the sample is narrow, the conclusion will be narrow too. Eliminate any answer that goes beyond what the study design permits.
Watch the language. Scan each answer choice for absolute or causal words. If you spot proves, causes, always, all, guarantees, treat that choice with extreme skepticism. The SAT rewards careful, qualified language.
Checkpoint: Does my chosen answer use softer language than the alternatives? If so, that's usually a good sign.

Time management tip: Don't reread the entire passage for study design questions. Instead, scan for the paragraph that describes how the study was conducted, it usually appears in the first or second paragraph. That's where you'll find everything you need: the study type, the sample, the variables, and the presence (or absence) of a control group.

Note: The passages below are original, SAT-style constructions for practice; any names or details are fictionalized.

Now put the strategy to work. Each question below presents a short research scenario in the style you'll encounter on test day. Before selecting an answer, identify the study type, find the variables, and check for a control group. Then match your conclusion to the design.

Passage

A team of researchers at a large Midwestern university surveyed 340 first-year students about their average nightly sleep duration and then compared those responses to the students’ end-of-semester GPAs. Students who reported sleeping seven or more hours per night had an average GPA of 3.4, while those who reported sleeping fewer than seven hours had an average GPA of 2.8.

easy

Which conclusion is best supported by the study described in the text?

Passage

A pharmaceutical company conducted a trial in which 80 participants who experience chronic migraines were given a new medication daily for twelve weeks. At the end of the trial period, 65% of the participants reported a reduction in migraine frequency. The researchers concluded that the medication was effective at reducing migraines.

medium

Which of the following is the most significant limitation of the study described in the text?

Passage

Researchers at a suburban elementary school randomly assigned 60 fourth-graders to one of two groups. The first group listened to classical music for ten minutes before completing a reading comprehension test; the second group sat in silence for the same period before taking the same test. Students in the music group scored an average of 12% higher than those in the silence group, and the difference was statistically significant.

medium

Which claim is best supported by the results of this study?

Passage

Agricultural researchers divided a test field into 40 identical plots and randomly assigned each plot to receive either an organic fertilizer or a standard chemical fertilizer. After one growing season, they measured the yield of tomatoes per plot. Plots treated with organic fertilizer produced an average yield of 22 kg, while plots treated with chemical fertilizer produced an average yield of 19 kg. The researchers noted that soil composition, sunlight exposure, and irrigation were consistent across all plots.

medium

Based on the study described, which of the following conclusions is most justified?

Passage

A ten-year longitudinal study tracked the dietary habits and cardiovascular health of 12,000 adults across six U.S. cities. Researchers found that participants who reported eating breakfast daily had a 27% lower rate of heart disease compared to those who reported skipping breakfast most days. The study controlled for age and sex but did not account for differences in overall diet quality, exercise habits, or socioeconomic status among participants.

easy

Which finding, if true, would most weaken the conclusion that eating breakfast reduces heart disease risk?

Identify the study type first, everything flows from that. If it's observational, no conclusion can claim causation. If it's experimental, check for a control group before accepting causal claims. Ask yourself: Did the researchers change something, or just watch?
The correct answer never overclaims. It matches the scope of the study, same population, same variables, same hedged language. When in doubt, choose the answer that says less, not more. Ask yourself: Does this answer stay within the boundaries of what was actually studied?
Learn to name the traps. When you can label a wrong answer as a "causation jump" or an "overgeneralization," you stop falling for it. Recognition is elimination. Ask yourself: Can I name the specific trap this wrong answer is using?
Absolute language is almost always wrong. Words like proves, always, all, never, guarantees don't belong in scientific conclusions. Prefer answers that use suggests, is associated with, may indicate, tends to. Ask yourself: Would a careful researcher actually say this?

Study design questions on the SAT aren't testing your scientific knowledge, they're testing your ability to think critically about how evidence was gathered and what it can reasonably support. That's a skill that extends far beyond test day. Every time you read a news article claiming a new study "proves" something, you'll know to ask: Was it an experiment or an observation? Was there a control group? Does the conclusion match the design?

Master the five-step strategy, learn to spot the four trap types, and these questions transform from intimidating science passages into some of the most predictable points on the SAT.

Remember: A study can only conclude what its design allows. Match the conclusion to the method, watch the language, and let the overclaiming answers eliminate themselves.

About the Instructor

Thomas Meli

Founder, Deep Test Prep

"Viewing the SAT as a transformative rite of passage, my approach builds self-awareness, sharpens critical thinking, and fosters a love of learning that supports students far beyond test day."

With over 20 years of teaching and 500+ hours of training in compassionate communication, Thomas uses scientifically proven methods to help students target and achieve their SAT goals.