Margin of error
First taught in High school statistics
Definition
The margin of error tells how far a result from a sample might be from the true value for the whole population.
Advertisement
It is written with (plus or minus).
Example: A poll finds that 52% of voters support a plan, with a margin of error of . The true support is probably between and . This range is a confidence interval.
Because 49% is below 50%, the poll cannot be sure that a majority supports the plan.
What affects the margin of error:
- Sample size: a larger random sample gives a smaller margin of error. For a proportion, a rough estimate at 95% confidence is , so 1,000 people give about .
- Confidence level: 99% confidence needs a larger margin than 95%.
The margin of error only covers random sampling error. It does not fix a biased sample.