Sampling and inference
Problem-solving and data analysis · 5 questions and templates for SAT prep, in order from warm-up to top rank. Check each answer, then read the full solution.
New to this topic? Read the lesson first: Samples, margin of error and what you can conclude.
Loading questions… (they need JavaScript; nothing you type is sent anywhere)
Written questions with solutions
The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.
-
MCQExam level
A researcher randomly chose 200 volunteers and randomly assigned half to drink green tea daily and half to drink water. After 8 weeks, the tea group had lower average blood pressure. Which conclusion is best supported?
- Drinking green tea daily likely causes lower blood pressure among people like these volunteers.
- Drinking green tea lowers blood pressure for everyone in the country.
- Green tea is related to blood pressure, but no cause can be suggested.
- Water raises blood pressure.
Answer and solution
Answer: A
Random assignment to groups lets you conclude cause and effect. But the people were volunteers, not a random sample of the country, so the result applies to people like them. Answer: the first option.
-
MCQExam level
To estimate how many hours residents of a city exercise each week, a researcher surveys 300 people leaving a gym. Why is the estimate likely to be unreliable?
- The sample is not representative of all residents, since people at a gym probably exercise more than average.
- The sample is too small to say anything.
- Surveys can never be used to estimate averages.
- The researcher should have surveyed 300 people at a library.
Answer and solution
Answer: A
The problem is who was asked, not how many. People leaving a gym exercise more than typical residents, so the estimate will be too high. A library sample would be biased too; a random sample of all residents is what is needed.
-
MCQHard
Two random samples were taken from the same population to estimate a mean. Sample A has 100 people and sample B has 900 people. Which statement is most likely true?
- Sample B's estimate has a smaller margin of error.
- Sample A's estimate has a smaller margin of error.
- Both have the same margin of error.
- Sample B's estimate must equal the true mean.
Answer and solution
Answer: A
Larger random samples give more precise estimates, so sample B has the smaller margin of error (about a third as large here, since the margin shrinks with the square root of the sample size). No sample is guaranteed to hit the true mean exactly.