Free specialty workspace
Statistics calculator
Use this statistics calculator to summarize data, fit linear regression models and calculate normal or binomial probabilities with visible assumptions.
Method version1.0.0 · Reviewed November1,2026 · Calculated on your device
Your calculation
Choose a method, then enter your figures. Examples are illustrative, not recommended settings or rates.
How this calculation works
Mean = Σx/n; sample variance = Σ(x−mean)²/(n−1); least-squares slope = Σ((x−meanX)(y−meanY))/Σ(x−meanX)².
What information do I need?
Only fields for your selected method are required. Use consistent units and scope. Changing an entry method or unit system does not convert the numbers already entered. Previously calculated results are cleared when inputs change.
Input definitions and where to find them
- What would you like to calculate?
- Choose the method matching your inputs.
- Observed values
- Up to1,000 observations. Commas separate values, not thousands.
- Paired observations — one x,y pair per line
- Use matching independent x and observed y values on each row. Every row receives equal weight.
- x at which to estimate y
- Choose a point within the observed x range when possible.
- Distribution mean
- Assumed population mean, in the same units as the bounds.
- Distribution standard deviation
- A positive assumed population standard deviation.
- Lower bound
- Lower endpoint in the same units as the mean.
- Upper bound
- Upper endpoint, at least as large as the lower endpoint.
- Independent trials
- Whole number1–1,000; identical probability on every trial.
- Success count of interest
- Whole number between zero and the number of trials.
- Probability per trial (decimal0–1)
- Use0.5 for50%; this field is not a percentage.
Worked example
This example uses the initial method shown in the selector. Load an example in the form to try it.
Example inputs
- What would you like to calculate?: describe
- Observed values: 2, 4, 4, 4, 5, 5, 7, 9
Example answer: Count = 8
Use the result well
Each observation has equal weight. Sample SD uses n−1; population SD uses n. Quartiles use linear interpolation at (n−1)p. The approximate mean interval requires independent representative observations and a suitable sampling distribution; it is not a prediction interval.
Review the limits shown with each result. Browser arithmetic has finite precision; small differences may reflect rounding. Print or download important results before closing the tab. This workspace does not store projects online or replace an approved examination calculator, engineering design or professional judgment.
Method and sources
References support the stated mathematical concepts; the interface and implementation are original. Publishers do not endorse this site.