📊 Statistics Calculators

Standard deviation, z-scores, confidence intervals, p-values: the stats toolkit for students, researchers, and the statistically curious.

Core Statistics 7

Probability & Odds 10

More tools 5

Which statistics calculator do you need?

Statistics splits into two jobs: describing the data you have, and deciding what it implies about data you do not have.

Sample or population: the mistake that changes your answer

Every spread calculation asks whether your numbers are the whole population or a sample drawn from it. The distinction is not pedantry. A sample divides by n minus 1 instead of n, which makes the result slightly larger, because a sample almost always understates the true spread of the thing it came from. On small data sets the gap is big enough to matter.

The rule of thumb: if you measured every member of the group you are describing, it is a population. If you measured some in order to say something about all, it is a sample. When you are unsure, use the sample formula, since it is the more cautious of the two.

What a p-value does not tell you

A p-value is the probability of seeing data at least this extreme if nothing were going on. It is not the probability that your hypothesis is true, and it is not a measure of how large or important an effect is. A tiny p-value on a trivial difference is common in large samples, and a large p-value does not prove there is no effect. This is why we show confidence intervals alongside: a range communicates both the direction and the precision of a finding, and it is much harder to over-read.

Frequently asked questions

Should I use sample or population standard deviation?

If your numbers are the complete group you are describing, use population. If they are a subset measured in order to say something about a larger group, use sample, which divides by n minus 1 and returns a slightly larger figure. When in doubt use sample: it is the more conservative choice and it is what most statistical software defaults to.

What does a p-value of 0.05 actually mean?

It means that if there were genuinely no effect, you would see data at least this extreme about 5 percent of the time. It does not mean there is a 95 percent chance your hypothesis is correct, and it says nothing about whether the effect is big enough to care about. The 0.05 threshold is a convention, not a law of nature.

Why is my confidence interval so wide?

Usually sample size. Precision improves with the square root of n, so quadrupling your sample halves the width of the interval. High variability in the data widens it too. A wide interval is not a failure, it is an honest statement that your data is consistent with a broad range of possibilities.

Do these calculators show the working?

Yes. The statistics and math tools render the steps using the numbers you entered, generated by the same code that produced the answer, so the explanation and the result can never disagree. That is a deliberate rule on this site rather than a feature of a few pages.