SAT Math: Explain when an experiment (vs. observation) supports causation
14+ practice questions in Praczo
The concept, explained
- 1
An experiment supports a causal conclusion only when: (1) subjects are randomly assigned to treatment and control, and (2) the treatment is controlled by the researcher.
- 2
Observational data, even with strong correlation, cannot support a causal claim because confounding variables may explain the association.
- 3
Well-designed experiments can support the conclusion "the treatment caused the effect" for the population the participants were drawn from.
- 4
Random assignment alone supports causation in the sample. Random selection is additionally needed to generalize to a broader population.
- ✗ Extending a causal claim beyond the population of the experiment. The conclusion applies only to people similar to the participants.
- ✗ Accepting causation from observational data because the effect is large or statistically significant.
SAT-style practice
A team randomly assigned 500 participants, drawn from the patients of a single clinic, to receive either a new drug or a placebo. After 6 months, the drug group showed lower blood pressure. Which conclusion is best supported?
Ready to master this concept?
Praczo tracks your mastery on all 183 SAT concepts — not just broad topics. One sample question is a start; drilling to mastery is how scores move.
3-day free trial — no credit card required