Sampling & Statistics

Margin of Error

Definition

The half-width of a confidence interval; how far the estimate may sit from the true value.

Why it matters

It is the precision you are buying with a given sample size.

In practice

The margin of error is the half-width of a confidence interval: how far the estimate may be from the true value at the stated confidence level.

It is quoted in percentage points of the whole sample, not relative to the estimate. A ±2.5-point margin is tight for a 50% estimate but very loose for a 3% one.

Margins shrink slowly. Halving the margin takes about four times the sample.

Formula

Margin of error = z × √(p × (1 − p) ÷ n)

Worked example

At 95% confidence and the most cautious assumption (p = 50%), about 385 documents give a ±5% margin, 1,537 give ±2.5%, and 2,401 give ±2%.

Illustrative numbers, not from a real matter.

Compare with

Work it out

← Back to glossary