Statistics Calculator
Drag a live curve for probabilities, or paste your data for t-tests and more.
Type a mean and standard deviation and drag the cut-offs to see any probability as a shaded area under the normal curve, or pick a calculator: z-scores, p-values and sample size, or tests on your own data such as t-tests, chi-square and regression.
Need the mean and SD of your data first? The standard deviation calculator works them out from a list.
ExampleProbability
68.27%
P(85 < X < 115) = 0.6827.
Type your own numbers, or drag the cut-off.
Drag the cut-offs along the axis; the shaded area is the probability.
- z-scores
- −1 and 1
- As a decimal
- 0.6827
- The rest
- 31.73%
How this works: 7 sources,
How this works
Method
Probabilities come from the normal, t, chi-square, F and binomial distributions (series and continued fractions for the incomplete gamma and beta functions), agreeing with SciPy to about 11 significant figures. Intervals and sample sizes use the textbook formulas. The tests on pasted data follow SciPy’s and statsmodels’ conventions and were checked against them on about 1,300 values. Combinations are exact whole numbers.
Sources
- NIST/SEMATECH e-Handbook of Statistical Methods
- Abramowitz and Stegun, Handbook of Mathematical Functions, chapters 6, 7 and 26 (1964)
- SciPy 1.18 scipy.stats (reference values for the tests)
- statsmodels 0.15 (reference values for the A/B test and regression intervals)
- Welch, The generalization of Student’s problem when several different population variances are involved (Biometrika) (1947)
- Fleiss, Levin and Paik, Statistical Methods for Rates and Proportions, 3rd edition (sample size for two proportions) (2003)
- Hyndman and Fan, Sample quantiles in statistical packages (The American Statistician) (1996)
How it’s tested
57 worked examples for this page are checked by automated tests before every release: given the inputs, the tool must show the expected answer.
Privacy
On your device. Nothing you enter or open leaves this browser. About privacy labels
Changes
- First published: live curves with draggable cut-offs, and tests on pasted data (t-tests, ANOVA, chi-square, Fisher’s exact, correlation, regression, A/B), checked against SciPy.
Key facts
- About 68.27% of a normal distribution lies within one standard deviation of the mean, 95.45% within two and 99.73% within three.
- A two-tailed test at the 5% level rejects when |z| exceeds 1.96, the value that leaves 2.5% in each tail.
- A poll of about 1,000 people has a margin of error of about ±3 percentage points at 95% confidence.
All 25 statistics calculators
The normal curve · Tests · Probability and counting · Estimates and surveys · Compare groups (your data) · Counts and A/B tests · Correlation and your data
The normal curve
Probability and counting
Estimates and surveys
Compare groups (your data)
Counts and A/B tests
Correlation and your data
Common questions
What does the shaded area mean?
Under a normal curve the total area is 1, so the shaded part is the probability of a value landing there. With a mean of 100 and an SD of 15, the area between 85 and 115 is about 0.6827: roughly two in three values fall within one standard deviation of the mean.
Where do I work out the mean and standard deviation of my data?
In the Math Calculator: paste a list of numbers into its mean, median and mode page or its standard deviation page. Then bring the mean and SD here to read probabilities and percentiles off the curve.
Can I test my own data here?
Yes. Paste columns from a spreadsheet into the t-test, ANOVA, Mann-Whitney, chi-square, correlation, regression, A/B test or descriptive statistics pages. Headers, blank cells, thousands separators and decimal commas are handled, and the page says what it left out. Small data sets travel in the link; large ones never leave your device.
How accurate are the probabilities?
They are computed from series and continued fractions in double precision, and checked against SciPy, the standard scientific library, to about 11 significant figures, including tail probabilities as small as 10⁻³⁰⁰.
Does anything I type leave my browser?
No. Every calculation runs on your device; nothing is uploaded. The numbers you type are kept only in the page’s address, so a link reproduces the same curve.