# Statistics Calculators

Compute common p-values, sample-size estimates, and confidence intervals locally, with the inputs visible.

Statistics Calculators is a scratchpad for a result you need to check while writing. It keeps the assumptions and inputs beside the answer, and the calculation stays on your device.

## Available calculators

### p-value

Compute an upper-tail or two-tailed p-value from:

- a t statistic and degrees of freedom;
- a z statistic;
- a chi-square statistic and degrees of freedom.

The calculator starts from a test statistic. It does not choose a statistical test from raw data.

### Sample size

Estimate a sample size for a proportion from the expected proportion, margin of error, and confidence level. Add a finite population size when sampling without replacement from a known population.

### Confidence interval

Calculate a t interval for a mean from the mean, standard deviation, and sample size, or a Wilson score interval for a proportion from successes and sample size.

## Run a calculation

1. Open **Statistics Calculators**, or run `/stats`.
2. Choose **p-value**, **Sample size**, or **Interval**.
3. Enter the values and select the method or tail.
4. Choose **Compute**.
5. Copy the result with its inputs, then record the method in the manuscript.

For example, a proportion of 81 successes from 100 observations uses a Wilson interval rather than the fragile `p ± 1.96 × SE` shortcut.

## What the calculator does not know

It cannot assess independence, sampling bias, model fit, missing data, multiple comparisons, power for a complex design, or whether a one-tailed test was justified before seeing the result. It also does not replace analysis code that should be preserved for reproducibility.

:::caution[Use as a check, not an analysis pipeline]
Confirm the formula fits the study design and report enough information for someone else to reproduce the number. For consequential analyses, keep the data, code, software version, and assumptions with the project.
:::

For a full analysis workspace, see [Data analysis](/docs/workflows/data-analysis/).