A margin of error describes how much a result may vary because a sample was drawn instead of asking every person in the population. It is useful, but it covers only one part of a survey’s uncertainty. A precise-looking range cannot repair a weak design.
Question wording can push people toward different answers. The order of questions can also change what respondents have in mind. These effects are not normally included in the familiar plus-or-minus figure printed beside a poll result.
Coverage matters too. A survey may miss people who lack internet access, speak another language or are difficult to reach. Weighting can reduce known imbalances, but it relies on assumptions and cannot recreate information that was never collected.
Nonresponse creates another risk. People who answer may differ from people who ignore the survey. A large sample does not automatically solve that problem; it can produce a narrow margin of error around a biased estimate.
Read the methodology alongside the headline number. Check who was eligible, how they were contacted, when the fieldwork happened and how the question was phrased. The margin of error is one measurement, not a certificate of total accuracy.
Original educational commentary by Breaking Click. This explainer is designed to help readers evaluate information more carefully.



