Statistical Experiments
Statistics gathers data three main ways: surveys ask questions, observational studies watch without interfering, and experiments impose a treatment and compare results.
Only an experiment can show cause and effect, because it controls what happens. Classifying a study correctly tells you what conclusions are allowed.
The three study types
A survey collects opinions or facts by asking. An observational study records what already happens without changing it. An experiment assigns treatments — like a drug versus a placebo — and compares outcomes.
The dividing line is intervention: experiments act on the subjects, observational studies and surveys do not.
Why it matters
Because experiments control the treatment and use random assignment, they can establish cause and effect. Observational studies can only show association.
So the same data pattern supports a stronger claim from an experiment than from an observational study.
Worked examples
Example 1: classify the study
Researchers randomly assign patients to a new drug or a sugar pill, then compare recovery. Which type of study is this?
Answer: An experiment
Example 2: observational
A researcher records exercise habits without changing them. Which type is that?
Answer: An observational study
Example 3: survey
A pollster asks 500 voters which candidate they prefer. Which type is that?
Answer: A survey
Try one yourself
Common questions
What separates an experiment from an observational study?
An experiment assigns a treatment and controls conditions; an observational study only watches without intervening.
Which study type can show cause and effect?
An experiment, because random assignment of treatments rules out many other explanations. Observational studies show only association.
Where do surveys fit?
Surveys collect data by asking questions. They are observational — they measure without imposing a treatment.
Want the video version?
Allday Everyday Math has video lessons, practice, and an AI tutor for every topic, Pre-Algebra through Algebra 2.