Margin of error
A margin of error is the plus-or-minus range around an estimate at a stated confidence level.
What margin of error is
A margin of error is a plus-or-minus amount attached to an estimate. If a survey reports 52 percent with a margin of error of 3 percentage points, the corresponding interval is often read as 49 percent to 55 percent at the stated confidence level.
Connection to confidence intervals
The margin of error is half the width of a confidence interval when the interval is symmetric around the estimate. It does not stand alone; it should be read with the confidence level, such as 90, 95, or 99 percent, and with the method used to calculate the interval.
Standard error
A margin of error is usually built from a standard error and a critical value. The Office for National Statistics gives the common 95 percent form as margin of error equals 1.96 times the standard error. Other confidence levels use different multipliers.
Survey estimates
In survey reporting, the margin of error summarizes uncertainty caused by using a sample instead of measuring the full population. The U.S. Census Bureau publishes margins of error for many American Community Survey estimates so readers can judge how precise the reported values are.
What affects it
Margin of error is affected by sample size, sampling design, response patterns, weighting, variability in the measured quantity, and the chosen confidence level. Larger samples usually reduce sampling uncertainty, but design effects, nonresponse, and weighting can make the real calculation more complex than a simple poll formula.
What it does not cover
A margin of error usually describes sampling uncertainty, not every possible source of error. It may not include biased questions, bad measurement, nonresponse bias, coverage gaps, data processing mistakes, or model assumptions. A small margin of error can still accompany a biased estimate.
Comparing estimates
Two estimates with overlapping margins of error are not automatically the same, and two non-overlapping intervals are not the only way to show a difference. Formal comparison often requires a standard error for the difference, especially when estimates come from related samples or complex surveys.
Why it matters
Margins of error help readers avoid treating sample estimates as exact facts. They make uncertainty visible, which is especially important in polling, public statistics, health surveys, market research, and any decision that depends on whether an observed difference is large enough to trust.