Sampling is the process of choosing which people a poll will question. In a random sample, every member of the population has an equal chance of being picked, which lets pollsters calculate a margin of error and generalize the results to the whole population.
Why it matters
How a sample is drawn decides whether the poll reflects everyone or only a slice of the public. Smaller samples, such as one age group within a poll, have larger margins of error, so their results are less precise.
How the exam tests it
You may compare an opt-in web survey, a quota sample of mall shoppers and a random telephone sample, and pick the random one. In tables, add and subtract the margin of error from each candidate's number. On the exam, if the two ranges overlap, treat the lead as not clearly beyond sampling error.
Don’t confuse it with
Students think quota sampling is as good as random sampling because it matches census groups, but in a quota sample the interviewers choose who to ask, so not everyone has an equal chance of selection.