Information is uneven because attention is costly. People often learn deeply about matters tied to their work, identity, or community and use party labels, endorsements, and trusted groups as shortcuts elsewhere. A cue can be sensible without guaranteeing a correct conclusion.

Good measures separate factual accuracy from education, interest, confidence, and ideological agreement. They sample more than one kind of fact, allow uncertainty, and ask whether the information tested is relevant to the decision being studied.

A school nurse may know exactly how a state vaccination rule is funded and administered but be unable to name a distant committee chair. A general quiz could miss that expertise. A confident respondent might name many national figures yet misunderstand the eligibility rule being debated. The first person has useful domain knowledge; the second has name recognition without accurate policy understanding.

Video lesson: Public Opinion: Crash Course Government and Politics #33