Measuring public opinion
Polls are everywhere, but only well-designed polls tell us what the public really thinks.
Sampling
A good poll uses a random sample, in which every member of the population has an equal chance of being chosen. A sample of about 1,000 people can represent a whole nation with a margin of error of roughly plus or minus 3 percentage points. Pollsters weight results so the sample matches the population's characteristics.
Question design and bias
The wording and order of questions can change results. Leading questions push respondents toward an answer. Push polls are not real polls but campaign tactics that spread negative information disguised as questions. Self-selected online polls, where anyone can vote, are unreliable because respondents are not representative.
Worked example: reading a poll
A poll shows Candidate A at 48% and Candidate B at 45%, with a margin of error of plus or minus 3 points. A's true support could be 45% to 51%, and B's 42% to 48%. Because the ranges overlap, the race is statistically too close to call, even though A appears to lead.
Flashcard
Each member of the population has an equal chance of selection
Check your understanding
A poll shows 52% to 48% with a margin of error of plus or minus 4 points. What is the best conclusion?