Margin of Error Calculator
The margin of error tells you how much the results of a survey might differ from the true opinion of the whole population you are studying. When a poll reports that 50% of respondents support a policy with a margin of error of plus or minus 4.90 percentage points, it means the true figure in the population is likely somewhere between 45.10% and 54.90%. Every survey that uses a random sample has a margin of error, because you are only measuring a portion of the population, and that sample may not perfectly represent the whole. The margin shrinks as you survey more people and widens if you use a higher confidence level or if the proportion is close to 50%, which is the point of maximum variability. The confidence level is the probability that the true value falls within the stated interval. At 95%, the standard choice, you accept a 5% chance that the true population value lies outside your interval. At 99% you are more certain but the interval is wider. This calculator uses the standard formula: MOE equals the z-score for your chosen confidence level multiplied by the square root of p times (1 minus p) divided by n, where p is the proportion and n is the sample size. For 95% confidence the z-score is 1.96, for 90% it is 1.645, and for 99% it is 2.576. Enter your sample size, sample proportion, and confidence level and you get the margin of error, the lower and upper bounds of the confidence interval, and the standard error. This tool suits students, researchers, journalists, and anyone interpreting survey or polling results. The formula assumes simple random sampling from a large population and does not apply a finite population correction.
How it works
The calculator reads your sample size n, sample proportion p (converted from a percentage to a decimal), and the z-score matching your confidence level: 1.645 for 90%, 1.96 for 95%, and 2.576 for 99%. Standard error is the square root of p(1-p)/n. Margin of error is z times the standard error, then multiplied by 100 to express the result in percentage points. The confidence interval lower bound is p minus MOE and the upper bound is p plus MOE, both reported as percentages. The formula assumes simple random sampling from a large population.
Worked example
You survey 400 people and 50% say they support a proposed local change. Using a 95% confidence level, z = 1.96. Standard error is the square root of (0.5 x 0.5 / 400) = square root of 0.000625 = 0.025, or 2.50%. Margin of error is 1.96 x 2.50% = 4.90%. The confidence interval runs from 45.10% to 54.90%. These match the default values pre-filled above.
Related calculators
- Standard Error Calculator: compute the standard error of the mean or a proportion from your data.
- P-value Calculator: find the probability of a result at least as extreme as your observed statistic.
- Correlation Coefficient Calculator: measure the strength and direction of a linear relationship.
- Percent Error Calculator: calculate how far a measured value deviates from a known accepted value.