Math calculators

Margin Of Error Calculator

Updated Jul 2, 2026 By Jehan Wadia

Survey Inputs

How certain you want to be that your result captures the true population value. 95% is the research standard — it means if you repeated the survey 100 times, 95 of those results would include the true answer.
The number of people or items you actually surveyed. Larger samples produce smaller, more precise margins of error.
Your best estimate of the percentage of the population that holds the characteristic you're measuring. If unknown, leave at 50% — this produces the largest (most conservative) margin of error.
The total number of individuals in the group you're studying. Leave blank if the population is very large or unknown. Entering a value here applies a correction that reduces the MOE when your sample is a meaningful fraction of the population.
Leave blank for an infinite / unknown population.

Your Margin of Error

Margin of Error The margin of error is a ± range around your result. For example, if 58% of respondents prefer Option A with a ±3% MOE, the true population preference is likely between 55% and 61% at your chosen confidence level.
±3.10%
(± 0.0310)
Z-Score Applied 1.9600
Confidence Interval
At 95% confidence, the true population proportion lies between 46.90% and 53.10%.

Margin of Error Quality Scale

Good Your margin of error of 3.10% falls in the Good range.
0%2%5%7%10%+
Excellent (< 2%)
Good (2–5%)
Acceptable (5–7%)
Poor (> 7%)
Aim for a margin of error between 2% and 7%. The lower the value, the more reliable and precise your results.
Step-by-Step Solution

How Sample Size Affects Your Margin of Error

Z-Score Reference Table

Confidence Level Z-Score

Introduction

The Margin of Error Calculator tells you how precise your survey results are. When you survey a group of people, you can't ask everyone in the whole population. Instead, you ask a smaller group — called a sample. The margin of error shows how close your sample's answers are likely to be to the true answer for the entire population.

For example, if a poll says 60% of people like a product with a margin of error of ±3%, the real number is most likely between 57% and 63%. A smaller margin of error means your results are more reliable. A larger one means there is more uncertainty.

To use this calculator, enter your confidence level, sample size, and population proportion. If you know the total population size, you can add that too for a more exact result. The tool will instantly calculate your margin of error, show you the confidence interval, give you a step-by-step solution, and display a chart so you can see how changing your sample size affects precision.

How to Use Our Margin of Error Calculator

Enter your survey details below to find out how precise your results are. The calculator takes your inputs and gives you a margin of error, a confidence interval, and a step-by-step breakdown of the math.

Confidence Level: Pick how sure you want to be that your results are correct. Most surveys use 95%. A higher number means more certainty, but it also makes the margin of error bigger.

Sample Size (n): Type in the number of people you surveyed. A bigger sample gives you a smaller margin of error, which means more precise results. If you need to figure out how many people to survey before running your study, try our Sample Size Calculator.

Population Proportion (p%): Enter your best guess for the percentage of people who hold the trait you are measuring. If you are not sure, leave it at 50%. This gives you the largest and safest margin of error.

Population Size (N): This field is optional. Enter the total number of people in the full group you are studying. If the group is very large or you do not know the number, leave this blank. When your sample is a large part of the population, filling this in will lower your margin of error.

What Is Margin of Error?

Margin of error tells you how close your survey results are to what the whole group really thinks. When you ask a small number of people a question instead of asking everyone, your answer won't be perfect. The margin of error is a range — shown as a plus-or-minus number — that tells you how far off your results might be from the true answer.

For example, if a poll says 60% of people like chocolate ice cream with a margin of error of ±3%, the real number is most likely between 57% and 63%.

How Is Margin of Error Calculated?

The margin of error formula is:

MOE = z × √(p(1 − p) / n)

Here is what each part means:

  • z (Z-score): A number tied to your confidence level that comes from the normal distribution. A 95% confidence level uses a Z-score of 1.96. A higher confidence level means a larger Z-score and a wider margin of error. You can also think of this as a critical value for your chosen confidence level.
  • p (Population proportion): Your best guess of the percentage of people who hold the trait you are measuring. If you don't know, use 50% — this gives you the largest and safest margin of error.
  • n (Sample size): The number of people you surveyed. A bigger sample gives you a smaller margin of error.

What Is the Finite Population Correction?

When your sample is a large chunk of the total population, the basic formula overstates the error. The finite population correction (FPC) fixes this. It multiplies the margin of error by:

√((N − n) / (N − 1))

Here, N is the total population size and n is your sample size. This factor shrinks the margin of error because you have already surveyed a big part of the group. If your population is very large or unknown, you can skip this step.

What Is a Confidence Level?

The confidence level is how sure you want to be that your range captures the true answer. A 95% confidence level means that if you ran the same survey 100 times, about 95 of those results would contain the true value. Higher confidence levels give you more certainty but also produce a wider margin of error. To build a full interval around your survey result, use our Confidence Interval Calculator.

What Is a Good Margin of Error?

In most surveys and polls, a margin of error between 2% and 5% is considered good. Below 2% is excellent but usually needs a very large sample. Between 5% and 7% is still acceptable for many purposes. Anything above 7% means your results may not be reliable enough to draw strong conclusions. You can explore related measures of accuracy with our Percent Error Calculator or review spread in your data using the Standard Deviation Calculator.

How to Reduce Margin of Error

There are three main ways to make your margin of error smaller:

  1. Increase your sample size. This is the most common fix. More responses lead to more precise results. Use our Sample Size Calculator to find the exact number of responses you need for your target margin of error.
  2. Lower your confidence level. Going from 99% to 95% shrinks the margin of error, but you give up some certainty.
  3. Use a proportion closer to 0% or 100%. The margin of error is largest at 50% and gets smaller as the proportion moves toward either extreme. You usually can't control this, but it helps explain why some surveys have tighter results than others.

Formulas used

Margin of Error (standard formula)
MOE = z \sqrt{\dfrac{\hat{p}(1-\hat{p})}{n}}
Margin of Error with Finite Population Correction
MOE = z \sqrt{\dfrac{\hat{p}(1-\hat{p})}{n}} \times \sqrt{\dfrac{N - n}{N - 1}}
Finite Population Correction Factor
FPC = \sqrt{\dfrac{N - n}{N - 1}}
Confidence Interval
CI = \hat{p} \pm MOE

Frequently asked questions

What does the ± symbol mean in margin of error?

The ± symbol means plus or minus. It tells you the range above and below your survey result. If your result is 50% with a margin of error of ±3%, the true answer likely falls between 47% and 53%.

Why should I leave the population proportion at 50%?

A proportion of 50% gives you the largest possible margin of error. This is the safest choice when you do not know the real proportion. It means your margin of error will not be too small by accident. If you already have data or a good estimate, you can enter that number instead.

Do I need to enter a population size?

No. Population size is optional. If your population is very large or you do not know the exact number, leave the field blank. The calculator will treat it as infinite. You only need to enter it when your sample is a big fraction of the total population, such as surveying 500 people out of 2,000.

What is a Z-score and where does it come from?

A Z-score is a number from the standard normal distribution that matches your chosen confidence level. For example, a 95% confidence level uses a Z-score of 1.96. The calculator picks the correct Z-score for you automatically. You can also see all the values in the Z-Score Reference Table at the bottom of the page.

Why does a bigger sample size give a smaller margin of error?

When you ask more people, your results get closer to the true answer for the whole population. In the formula, the sample size n is in the bottom of a fraction inside a square root. As n gets larger, that fraction gets smaller, which shrinks the margin of error.

What confidence level should I use?

Most surveys use 95%. It is the standard in research and gives a good balance between certainty and precision. Use 99% if you need extra certainty, like in medical studies. Use 90% if you are doing a quick informal poll and want a tighter range.

Can I use this calculator for small sample sizes?

Yes, but keep in mind that very small samples produce large margins of error. The formula works for any sample size of 1 or more. However, if your sample is very small (under 30), the results may be less reliable because the normal distribution assumption becomes weaker.

What is the difference between margin of error and confidence interval?

The margin of error is just the ± number. The confidence interval is the full range you get when you add and subtract the margin of error from your survey result. For example, if your result is 60% and the margin of error is ±4%, then the confidence interval is 56% to 64%.

When does the Finite Population Correction get applied?

The calculator applies the Finite Population Correction (FPC) only when you enter a population size that is larger than your sample size. It shrinks the margin of error because you have already surveyed a meaningful portion of the total group. If you leave the population size blank, no correction is applied.

Does this calculator work for non-survey data?

This calculator is designed for proportion-based surveys — questions where the answer is a yes/no or a percentage. It uses the formula for proportions. If you are working with means (averages) instead of proportions, you would need a different formula that uses standard deviation instead of p(1−p).

Why does my margin of error change when I change the proportion?

The formula uses p × (1 − p), which is largest when p is 50% and gets smaller as p moves toward 0% or 100%. So a proportion of 50% always gives the biggest margin of error, and extreme proportions like 10% or 90% give smaller ones.

What does the quality scale on the results page mean?

The quality scale rates your margin of error from Excellent to Poor. Below 2% is excellent, 2% to 5% is good, 5% to 7% is acceptable, and above 7% is poor. It helps you quickly see if your survey results are precise enough to trust.

Can margin of error be zero?

In practice, no. The only way to get a margin of error of exactly zero is to survey every single person in the population. As long as you are sampling a portion of the group, there will always be some margin of error.

How does the chart on the results page help me?

The chart shows how the margin of error changes as the sample size increases. It lets you see the point of diminishing returns — where adding more respondents stops making a big difference. Your current sample size is marked on the chart so you can compare it to other options.