SAT OwlQuestion bank / Evaluating statistical claims: observational studies and experiments
A news website invited its readers to vote in an online poll about a proposed city tax. Of the readers who voted, opposed the tax. Which of the following is the best reason the poll results should not be used to estimate the percentage of all city residents who oppose the tax?
A sample of is too large to be accurate.
The respondents chose to participate, so they may not represent all city residents.
The poll did not use random assignment to treatment groups.
The percentage, , is too high to be believable.
Show answer and explanation
Answer: B
The respondents selected themselves, and only readers of the website could vote. People with strong opinions are more likely to respond to voluntary polls, so the sample is likely biased. A large sample does not fix a biased selection method.
Choice C names a requirement for experiments; this is a survey, where the issue is random selection, not random assignment.
Got it wrong? On SAT Owl, Owl looks at the answer you chose and explains the exact mistake, then gives you similar questions until it sticks. Get your predicted SAT score free.