Evaluating Statistical Claims

Unit 3 · Problem-Solving & Data Analysis

What the SAT asks here

These questions test two principles: random selection (sampling) enables generalization from sample to population, and random assignment (in an experiment) enables cause-and-effect conclusions. Combining both is the only way to claim 'X causes Y in the broader population.' About one problem per test.

Decision rule

Random selection + random assignment → causal AND generalizable

Random selection only → association, generalizable

Random assignment only → causal, NOT generalizable beyond study group

Neither → association only, limited scope

SAT Tip: Read the study description for two flags: 'randomly selected from the population' and 'randomly assigned to groups'. Each flag unlocks one stronger conclusion.

Caution: Observational studies (no random assignment) can never justify causation, even with random selection. Confounding variables remain.

Type 1

Identify whether causation is justified

Causation requires random ASSIGNMENT. Without it, only association can be claimed.

Example 1

A study finds that students who eat breakfast score higher on tests. The students chose for themselves whether to eat breakfast. Can we conclude breakfast CAUSES higher scores?

Practice more of this type— AI-generated · always-new problems

Generate Problems →

Type 2

Identify whether generalization is justified

Generalization to a broader population requires random SELECTION from that population.

Example 2

Researchers randomly assigned 200 volunteer patients at one hospital to either a new drug or a placebo. The drug group recovered faster. Does this generalize to all patients with the condition?

Practice more of this type— AI-generated · always-new problems

Generate Problems →

Type 3

Identify the stronger conclusion

Apply both rules at once. Random selection unlocks generalization; random assignment unlocks causation. Both together give the strongest claim.

Example 3

A nationwide study randomly selects 5,000 adults and randomly assigns each to one of two diet plans. The plan-A group loses more weight on average. What kind of conclusion is justified?

Practice more of this type— AI-generated · always-new problems

Generate Problems →

← Inference from Sample Statistics & Margin of Error Area & Volume →