Evaluating statistical claims: Observational studies and experiments
Use random assignment to judge causation and random sampling to judge generalization.
核心知识
Whether a study can support a causal claim or only show an association depends on two design choices: how participants were assigned to groups (random assignment vs. self-selection) and how they were selected from the population (random sampling vs. convenience).
深入理解
The SAT tests your ability to read a study description and decide what kind of conclusion is valid. Two questions drive every judgment:
Was there random assignment to treatment groups? If yes, the study is an experiment and can support cause-and-effect claims. If no, it is an observational study and can only show an association.
Was the sample randomly selected from a larger population? If yes, results can be generalized to that population. If no, results apply only to the people actually studied.
These two features are independent. A study can have one, both, or neither. Most SAT questions on this topic give you a short paragraph describing a study, then ask what conclusion is justified. The wrong answers typically overstate what the design allows — claiming causation from an observational study, or generalizing from a convenience sample.
Your job is to match the study design to the strongest claim it can support, nothing more.
知识点教程
5Distinguish observational studies from experiments.
Experiments assign treatments; observational studies just record what already happened.
Understand why random assignment supports causal claims.
Random assignment makes cause-and-effect claims possible.
Understand how random sampling affects generalizing results to a population.
Random sampling lets you generalize findings to the population sampled.
Identify the population to which results can be extended (SAT).
The population is the exact group the sample was drawn from.
Evaluate whether a study design supports a stated statistical claim.
A valid conclusion must stay within what the study design allows.