What this calculator does
A confidence interval turns a sample estimate into a range. Instead of reporting a single average, it reports a band that is likely to contain the true population value, together with the confidence level that describes how often the method works. The width depends on how much the data vary, how large the sample is, and how confident you want to be. This calculator builds the interval for a mean using the normal approximation, the version most people meet first.
How to use it
- Enter the mean of your sample.
- Enter the sample standard deviation.
- Enter the number of observations.
- Choose 90%, 95%, or 99% confidence.
- Read the standard error, margin of error, and both bounds.
How it works
Example calculations
Example 1: mean 100, SD 15, n 30, 95%
Standard error = 15 / sqrt(30) = 2.74. Margin of error = 1.96 x 2.74 = 5.37. The interval runs from 94.63 to 105.37.
Example 2: mean 50, SD 10, n 100, 99%
Standard error = 10 / 10 = 1.00. Margin of error = 2.576 x 1.00 = 2.58. The interval runs from 47.42 to 52.58; the larger sample keeps the band narrow even at 99%.
Example 3: mean 200, SD 40, n 25, 90%
Standard error = 40 / 5 = 8.00. Margin of error = 1.645 x 8.00 = 13.16. The interval runs from 186.84 to 213.16, the widest of the three.
What the result means
A 95% confidence interval does not mean there is a 95% chance the true mean lies inside this range. The true mean is fixed; the randomness is in the sampling. The 95% figure describes the method: if you repeated the study many times and built an interval each time, about 95% of those intervals would contain the true value. A wide interval means the estimate is imprecise; a narrow one means the data pin it down more tightly. The interval is symmetric around the sample mean because it comes from a normal approximation, and it can extend past values that are impossible in principle.
Common mistakes
- Reading 95% as the probability that the true mean falls in this one interval.
- Using the standard deviation where the standard error belongs.
- Forgetting to divide by the square root of the sample size.
- Using z values with a very small sample instead of a t distribution.
Assumptions and limits
- The sample is random and the observations are independent.
- The sample mean is approximately normally distributed.
- The sample standard deviation estimates the population value well.
- Only sampling error is covered; bias and measurement error are not.
Frequently asked questions
Why are the z values 1.645, 1.96, and 2.576?
They cut off the middle 90%, 95%, and 99% of a standard normal distribution, leaving 5%, 2.5%, and 0.5% in each tail. These are the standard multipliers for those levels.
What does 95% confidence actually mean?
If the same sampling procedure were repeated many times, about 95% of the intervals produced would contain the true mean. It describes the method, not one interval.
Should I use z or t?
Use z when the population standard deviation is known or the sample is large. Use t for small samples with a sample standard deviation, because t intervals are wider.
What happens if I increase the sample size?
The standard error falls with the square root of n, so quadrupling the sample halves the margin of error. Cutting the margin in half again takes about sixteen times the original data.
Can the interval go below zero?
Yes, especially with small samples or variables that cannot be negative. That warns that the normal approximation may not suit the data.
Is a 99% interval always better than a 95% one?
It is wider, so it is more likely to contain the true value but less informative about where that value sits. The right level depends on the cost of a wrong conclusion.
Sources
2026 update
How to use the professional output
This page is not just a formula box. The result is designed to be checked, compared, and taken away.
- Professional worksheet. The result includes the inputs used, the method and scope, and a one-variable sensitivity check for finance calculators.
- A/B scenarios. Save two input sets in the browser and compare the primary, secondary, and breakdown figures side by side.
- Print / PDF and CSV. Print a branded worksheet, or export the year-by-year table when the calculator produces one.
The formulas and assumptions above still govern the number. The worksheet makes the estimate auditable; it does not make it a lender quote, tax return, or personalized advice.
Go deeper
