Statistics Calculator

What This Is For

You have a list of numbers. You need to know the center, the spread, and whether anything smells off. Copy a response-time log from your monitoring dashboard, a week of error rates, or experiment results -- paste, and the tool returns every descriptive statistic in under a second.

What It Computes

Central tendency: mean, median (50th percentile), mode, geometric mean. Dispersion: range, variance, standard deviation, interquartile range, mean absolute deviation. Shape: skewness (>0 = right tail), kurtosis (>3 = heavy tails, lots of outliers). Percentiles: Q1, Q2, Q3, P95, P99, P99.9. Correlation: paste two equal-length columns for Pearson's r. Visualization: histogram and box plot rendered in SVG.

Sample vs Population: n-1 vs n

The tool defaults to sample standard deviation (n-1 denominator, Bessel's correction). If your data is a subset of a larger population, use sample (n-1). If you literally have every observation, use population (n).

Outliers and the Mean-Median Gap

A mean of 64ms with a p99 of 200ms doesn't mean "the average is 64." It means "the average is 64 but something is dragging the tail up." The large gap between mean and median flags skewness. Compare them to decide whether trimming is appropriate.

Reporting Context

Means without context are misleading. Always report n alongside your summary measures. A mean of 50ms over 15 observations has different weight than over 15,000.

FAQ

Q: Median vs mean? A: Median whenever the distribution is skewed or contains outliers. Response times, incomes, file sizes are classic right-skewed measures. Q: Does the tool do hypothesis testing? A: No. It provides the descriptive summaries that you feed into a separate t-test or ANOVA calculator.