Average & Median

Paste a column of numbers — labelled, comma-separated or one per line — and read fourteen summary statistics back in a single table.

Runs on your deviceCalculatorsWorkingNo account needed

What this tool does

Average & Median reads every number out of whatever you paste and returns fourteen figures in one table: count, sum, mean, median, mode, minimum, maximum, range, variance, standard deviation, the lower and upper quartiles, the interquartile range and the geometric mean. That table is the entire output — Copy output hands back the same fourteen rows as two-column CSV, ready to paste into a spreadsheet. The values themselves are not echoed back in sorted order, so if you want to eyeball the outlier that pulled the mean away from the median you have the minimum, maximum and quartile rows to reason from rather than the list.

The input is free text rather than a strict list. Values may be separated by spaces, commas, semicolons, pipes or line breaks, so a column copied straight out of a spreadsheet works as it stands. A line beginning with # or // is treated as a comment and skipped, and a label ending in a colon is stripped off the front of a line, so "Week 1: 12" contributes 12 and not 1 as well. A leading currency symbol and a trailing unit are trimmed from a token, and a trailing per-cent sign is dropped rather than converted, so 15% counts as 15. Anything still not numeric after that trimming is discarded in silence — the table carries no warning and no tally of what was thrown out, which makes the Count row the one place a swallowed token shows itself.

One parsing rule catches people out, and it is worth knowing before you trust the answer: the comma separates values here, it is not a thousands mark. Paste 1,234 and the tool reads two numbers, 1 and 234, which quietly changes the count and every statistic derived from it. The same applies to a decimal comma, where 3,5 becomes 3 and 5. Strip the grouping commas first — Find and Replace will do it in one pass — or export the column without them, and check the Count row against the number of values you meant to paste.

Two settings change the numbers rather than their presentation. Standard deviation and variance use n − 1 by default, treating your figures as a sample of something larger; switch to Population (n) when the list is the entire set, and the population figure comes out the smaller of the two — unless every value is identical, in which case both are zero. Quartiles are the medians of the lower and upper halves, with the middle value excluded when the count is odd, so they will not always agree with Excel's QUARTILE or R's default — a difference of convention rather than of correctness. Decimal places affects display only. Sample variance and sample standard deviation need at least two values, so a lone number shows n/a in exactly those two rows; every other row still fills in, geometric mean included, since a single positive value is its own geometric mean. That row turns to n/a only when a zero or a negative appears anywhere in the list.

Every figure here is produced by JavaScript inside the page you are reading. Nothing is uploaded, no server sees the list, and the values live only in this tab until you close or reload it.

How to use it

  1. Paste the numbers. Spaces, commas, semicolons, pipes and new lines all work as separators. Load sample fills the box with a labelled six-week series if you want to see the expected shape first.
  2. Choose sample or population. Sample (n − 1) is the default and suits figures that stand in for a larger group. Population (n) is for a complete set — every employee, every unit shipped, every exam sat.
  3. Set the decimal places. Anywhere from 0 to 12, four by default. This rounds what is shown in the table; the arithmetic underneath is unaffected by it.
  4. Read the table. Fourteen rows, with the sample or population basis named in the variance and standard deviation labels so the working is unambiguous. Copy output puts those rows on the clipboard as CSV. Check Count first — it is your only confirmation that the tool read the same quantity of numbers you thought you pasted.

Processing & privacy

Runs on your device. Your file is processed directly in this browser.

Common use cases

Summarising a spreadsheet column

Copy the column, paste it, and get the spread as well as the average without writing STDEV.S and QUARTILE into four spare cells.

Checking whether the mean is representative

A mean sitting well above the median says a long tail is pulling it. Both figures are in the same table, next to the minimum and maximum responsible.

Reading benchmark timings or test scores

Paste the run times from a log and read the median and interquartile range, which survive one freakishly slow run better than the mean does.

Working through a statistics exercise

The sample and population options give the two answers a textbook question distinguishes between, with the basis printed in the row label so the working is unambiguous.

Questions

Are my numbers uploaded?
No. Average & Median runs in this browser tab, so the figures you paste are not transmitted or stored, and the tool still works with the network switched off.
Why is the count higher than the number of values I pasted?
Almost always thousands separators. A comma splits values here, so 1,234 is read as 1 and 234. Remove the grouping commas before pasting, then check the Count row against the number of values you expected.
Why is the count lower than I expected?
A token that is not numeric once the leading symbol and trailing unit have been trimmed is dropped, and the table says nothing about it. Ranges like 10-20, footnote markers and stray words all disappear this way, so a Count that falls short means something in the paste was not read as a number.
Should I pick sample or population standard deviation?
Sample (n − 1) when the numbers represent a larger group you did not measure in full, which covers most collected data. Population (n) when the list is everything there is. The population figure is the smaller of the two whenever the values differ at all.
Why does it say there is no mode?
A mode is only reported when some value occurs more than once, so a list of distinct values has none. Where several values tie for most frequent, all of them are listed together with the number of times they appear.
Do the quartiles match Excel?
Not always. Q1 and Q3 here are the medians of the lower and upper halves, with the middle value left out when the count is odd. Excel's QUARTILE and R's default interpolate instead, so the two can differ on short lists. Neither is wrong; they are different conventions.
What happens to percentages and currency symbols?
A leading symbol and a trailing unit are trimmed off the token, and a trailing per-cent sign is dropped without converting the value — 15% is counted as 15, not 0.15. Divide by a hundred yourself if you need the proportion.
Can I weight the values or give each a frequency?
No. Every number counts once, so a frequency table has to be expanded into repeated values before pasting. There is no grouping by category either — the whole input is treated as one set.

Related tools