All tools

Margin of error calculator

You already have responses and want to know how much you can trust the percentages. Enter how many responses you collected and, if you know it, the population. It tells you the margin of error.

Leave empty if it is very large or unknown

Margin of error

± 5.7 %

Your percentages can be off by ± 5.7 points

The calculation runs in your browser. We don’t store what you type.

How it’s calculated

Margin of error = Z × √(p × (1 − p) / n), with Z from the confidence level, n the number of responses and p = 0.5 (the most demanding case).

If you know the population (N), it is multiplied by √((N − n) / (N − 1)). When a large share of the population answered, the margin drops a lot.

How to read the result

  • A ±4% margin means a 52% in your survey could really be anywhere between 48% and 56%.
  • If two options are closer than the margin, you can’t say which one wins. To compare two groups, use the compare two percentages tool.
  • The margin only covers random error. It doesn’t cover badly worded questions or a sample that doesn’t represent the population.

Frequently asked questions

What is a good margin of error?
Opinion polls usually accept ±3% to ±5%. For internal decisions or exploratory studies, ±7% or ±10% can be enough.
Why doesn’t my margin drop much when I add responses?
Because it shrinks with the square root of n: to halve it you need four times as many responses.
Does it work for subgroups?
Yes, but use that subgroup’s number of responses. If you only analyze 80 women out of 300 responses, the margin is the one for 80.
Is it for your thesis?See the step-by-step guide to running your thesis survey: sample, questionnaire, consent, pilot test and analysis.Thesis surveys

Know what you want to ask?

Build the survey in SurvIQ, share it by link, WhatsApp or QR code, and watch results come in live. The Free plan is free forever.

Create a free survey

More tools