Blog
Statistics on data

Descriptive statistics calculator online

Seventeen measures computed on a real column of a real file, not on a dozen numbers typed by hand. Every formula used here is stated explicitly below, so the numbers match what your spreadsheet returns rather than quietly differing in the third decimal.

Your data never leaves this device

Your data

Paste a table straight from Excel or Google Sheets, or drop a CSV file. Nothing is uploaded anywhere — the whole calculation runs inside this browser tab.

CSV, TSV or plain text. Files saved in windows-1251 are detected and re-read automatically, so Cyrillic headers do not turn into garbage.

Parsing options

Columns

Result

The drawing library is downloaded only when you press the button, and only on this page.

This page has no server side at all. The table you paste is parsed by JavaScript inside your own browser, the chart is drawn on a canvas element on your machine, and the export file is assembled locally. Nothing is uploaded, stored or written to any log. You can disconnect from the network after the page has loaded and everything will still work — which is the simplest way to verify the claim yourself.

Facts and limits of this method

Measures returned 17: count, sum, mean, median, mode, min, max, range, Q1, Q3, IQR, variance, standard deviation, standard error, coefficient of variation, skewness, kurtosis
Quantile definition R type 7 — the same as Excel PERCENTILE.INC, Google Sheets and numpy default
Variance divisor n−1 (sample). The population value is available on the dedicated variance page.
Skewness formula G1, the bias-corrected estimator used by Excel SKEW
Kurtosis formula G2 excess kurtosis, used by Excel KURT — a normal distribution gives 0, not 3
Minimum rows 2 for most measures, 4 before kurtosis becomes defined
Missing values Skipped, not treated as zero. The count reflects valid numbers only.

When it misleads you

How it is calculated

The mean is the sum divided by the count. The median is the middle value of the sorted series, or the average of the two middle values when the count is even. The mode is the most frequent value; if every value occurs exactly once there is no mode, and the tool says so instead of inventing one.

Quartiles use the R type 7 definition: the position is (n−1)·p, and when it falls between two elements the result is interpolated linearly between them. This matters more than it sounds — the nine competing quantile definitions can differ by several percent on small samples, and a tool that does not tell you which one it uses cannot be reconciled with your spreadsheet.

The variance is computed with the n−1 divisor, which is the unbiased estimator for a sample drawn from a larger population. This is what Excel VAR.S returns and what almost every textbook means by "variance" unless it explicitly says "population". The standard deviation is its square root, and the standard error of the mean is that divided by the square root of the count.

Skewness is the G1 estimator and kurtosis is the G2 excess estimator, both bias-corrected for sample size, matching Excel SKEW and KURT. Excess kurtosis subtracts 3, so a perfectly normal distribution scores 0: positive values mean fatter tails than normal, negative values mean thinner ones.

The coefficient of variation is the standard deviation divided by the absolute value of the mean, expressed as a percentage. Because it has no units, it is the only measure here that lets you compare the spread of quantities measured in completely different scales.

Questions and answers

Why does my standard deviation differ from another calculator?

Almost always because of the divisor. This page uses n−1, the sample formula. Calculators aimed at school problems often use n, the population formula. Both numbers are on the dedicated standard deviation page so you can see the difference directly.

Why is my kurtosis near 0 and not near 3?

This is excess kurtosis, which subtracts 3 so that a normal distribution scores exactly 0. It is the convention used by Excel and by most statistical software. Add 3 if you need the raw fourth-moment version.

What happens to empty cells?

They are skipped entirely, and the reported count refers only to valid numbers. They are never silently converted to zero, which would drag the mean towards it and quietly corrupt every other measure.

How many rows can it handle?

Hundreds of thousands. Everything is a single pass or a sort, and it all happens in memory in your browser — the practical limit is the size of the file your browser is willing to read, not the calculation.

Is any of this sent to a server?

No. There is no server. Open the page, disconnect from the network, paste your data — it still works. That is the whole point of running the statistics locally: payroll, medical and financial columns never have to leave the machine they are on.

The opposite tool

Need it the other way round? Mean, median and mode Three centres of the same data, and why they disagree

Related tools

All tools