Calculate the confidence interval for a sample mean at a chosen confidence level (90%, 95%, or 99%).
📖 Standard Confidence Interval Formula (Z-based)Fill in the fields on the left with your information, then press Calculate to see your result instantly. No sign-up required, and no data is sent anywhere — everything is calculated right in your browser.
A confidence interval is a range of values expected to contain the true mean of the entire population, at a given confidence level. Instead of relying on a single sample's mean as a certain, exact value, a confidence interval gives a realistic range that reflects the natural uncertainty in any statistical sample. It's calculated by multiplying the "Z-value" associated with the desired confidence level (1.645 for 90%, 1.96 for 95%, 2.576 for 99%) by the standard error; this result is called the "margin of error," which is added to and subtracted from the sample mean to get the interval's upper and lower bounds. Confidence intervals are widely used in opinion polls and scientific and medical research to show how accurate results estimated from a limited sample are. A wider confidence interval is not necessarily a weaker result — it often simply reflects a smaller sample size or higher variability in the underlying data, both of which are honestly disclosed rather than hidden by reporting a single misleadingly precise number.
News reports on opinion polls almost always cite a 'margin of error', which is the practical output of a confidence interval calculation, yet the underlying concept is frequently misunderstood even by people who read these figures regularly.
A 95% confidence interval does not mean there is a 95% probability that the true population value falls within the reported range for this specific sample. The more precise, technically correct interpretation is that if the same sampling process were repeated many times, approximately 95% of the confidence intervals calculated from those samples would contain the true population value. This is a subtle but genuinely important distinction that statistics courses emphasize precisely because the more intuitive-sounding interpretation is technically incorrect.
The width of a confidence interval is driven by three factors, each with a common-sense reason behind it. A higher confidence level (99% instead of 90%) produces a wider interval, because being more certain of capturing the true value requires casting a wider net. A larger sample size produces a narrower interval, because more data reduces the uncertainty inherent in estimating from a subset rather than the whole population. Higher variability in the underlying data produces a wider interval, because more scattered data makes the sample mean itself a less reliable estimate of the true population mean.
The Z-values used in the formula (1.645, 1.96, and 2.576 for the 90%, 95%, and 99% confidence levels respectively) come directly from the standard normal distribution, representing how many standard errors need to be spanned on each side of the sample mean to capture the desired percentage of possible sample outcomes under the assumption that the sampling distribution is approximately normal — a reasonable assumption for most sample sizes encountered in practice, by the Central Limit Theorem.
Confidence intervals appear constantly in published research, political polling, and clinical trials specifically because reporting a single number (like '52% support this policy') without any indication of its uncertainty would be misleading about how precisely that number is actually known. A poll based on 200 respondents and a poll based on 20,000 respondents might report the identical central estimate, but the confidence interval reveals that the larger poll's estimate is considerably more reliable — information a bare percentage alone cannot convey.
It means that if the same sampling process were repeated many times, about 95% of the resulting intervals would contain the true population mean.
Higher confidence requires a larger margin of error to be more certain the true value falls within the range.
Yes, larger samples reduce the standard error, which narrows the interval for the same confidence level.