AP Statistics: Correlation vs Causation in Study Design Questions

Last updated: 7 October 2026

AP Statistics: Correlation vs Causation in Study Design Questions

Two rules cover this part of AP Statistics: you can conclude cause and effect only from a randomized experiment, and you can generalise to a population only when the sample was randomly selected from it. Without one of these kinds of randomness, the conclusion has to be narrower.

Observational study vs experiment

Observational studyExperiment
What the researcher doesObserves and records; no treatment imposedImposes treatments on experimental units
Cause and effect?No, confounding is possibleYes, with random assignment
Typical wording“There is an association between…”“Giving… causes a change in…”

What your conclusion can claim

Random assignmentNo random assignment
Random samplingCause and effect, generalised to the populationAssociation only, generalised to the population
No random samplingCause and effect for the units studiedAssociation only, for the units studied

Example

Illustrative example. A survey finds that students who sleep more have higher grades. This is an observational study, so you can only state a positive association, not that more sleep causes better grades; a confounding variable such as workload or after-school activities may affect both.

Checklist before writing a conclusion

  • Is it an observational study or an experiment?
  • Was there random assignment?
  • Was there random sampling from a population?
  • Is the conclusion written in the context of the question?

Want a study plan that fits your child?

Tell us the curriculum, year group, exam date (if any) and goal. Our academic team will suggest a starting plan, 1-on-1 at our Bangkok centre or online.

Learn 1-on-1: 1-on-1 AP tutoring

Frequently Asked Questions

Does a very high r mean causation?

No. Correlation measures the strength and direction of a linear relationship, not what causes what.

What is a confounding variable?

A variable related to both the explanatory and the response variable, so you cannot tell whether the observed effect comes from the variable being studied.

Should I mention randomness in my answer?

Yes. State which kind of randomness the study used, because it justifies the scope of your conclusion.

Max Academia offers 1-on-1 AP tutoring by subject. See also AP Calculus FRQ justification.

Share the Post: