Find the arithmetic mean, sum, count, minimum and maximum of a list of numbers. Enter positive or negative values, decimals or scientific notation.
Formula and sources updated · Data sources · 1View algorithm, constants and default inputs (JSON)
Find the arithmetic mean by adding every value and dividing by the number of values. Enter numbers with decimal points separated by commas or spaces, or use semicolons between values with decimal commas. Do not use thousands separators. An extreme value can pull the mean away from the centre of the data; consider the median when that matters. Results use limited precision and are displayed to four decimal places.
This page focuses on the arithmetic mean and also reports count, sum and extremes. If your task needs a geometric or harmonic mean, use the mean-type selector on the other tool and check the allowed values for that definition. These means are not interchangeable.
Recomputed from this page's implementation at build time · checked 2026-10-22
The mean is not really a question about arithmetic — it is a question about how many values the line contains. This page fixes the parsing rules, then shows the same data summarised two ways and the inputs where a separator quietly changes the answer.
| three values | plus 100 | plus 1,000 | plus 10,000 | |
|---|---|---|---|---|
| Mean | 2.00 | 26.50 | 251.50 | 2501.50 |
Every extra value carries a quarter of the total weight no matter how small the other three are, so one entry can move the mean further than the other three combined. The mean is a ratio, not a summary.
None of the five values equals the mean. A mean is a ratio, so it usually lands between two of your numbers rather than on one of them — that is arithmetic, not a rounding fault.
Here the middle value is a number you typed, while the mean was not. Two summaries of identical data disagree about what is typical, and both are defensible.
Most of the 8 entries sit at 4 or below while the largest is 100. The mean leans toward that single value, which is exactly why the next card deletes it and re-runs.
Removing one entry costs most of the mean. Compare this card with the previous one: when a data set has a long tail, a mean reported alone is closer to a description of its largest value than of its middle.
The semicolon marks commas as decimal points, so this line holds 3 values rather than six small integers. Same characters, different parse, different answer.
Read as 2 values, not as one thousand and a half. There is no thousands separator here at all: a comma splits values unless a semicolon appears. Strip formatting before pasting.
Exponents are normal numbers to this parser. A tool that accepted only digits would reject the line even though nothing in it is ambiguous.
average
Nothing but separators
An empty list has no mean, and the page says so instead of printing zero. Zero is a legal answer for other inputs, so using it here would make the two cases indistinguishable.
inputError
A word among the numbers
The whole line is rejected. Skipping the word would shrink the count and move the mean, changing the answer without any visible sign that it happened.
unavailable
Two values that overflow the sum
Each value is representable; their sum is not. Reporting infinity or falling back to the largest value would both be invented answers, so the calculation stops.
calculator.netchecked 2026-10-22
Why they built it that wayMean, median, mode and standard deviation live on four pages that share one form. That is a search-structure decision more than a mathematical one: each page owns a different query, so none of them has to explain how the other three differ.
The extra fields are the diagnostic. When the count disagrees with what you thought you entered, that number is the only visible evidence that a separator was read differently — a page showing just the mean would hide it.
Enter 2, 7, 19, 24, 25: the mean reads 15.4 and the count reads 5. Then replace the line with 1,000.5 and watch the count drop to 2 — the comma split one number into two.
Re-summed every list by hand and compared the compensated total against plain summation · impl d4b25859
Arithmetic mean is sum(values) / count(values). Compensated summation reduces cancellation error but retains floating-point limits. Commas or whitespace separate decimal-point values; semicolons select decimal-comma parsing. Every token must be finite. Empty text is preserved and rejected.
Not arbitrary-precision arithmetic. Displayed decimals are rounded. Empty lists, invalid tokens and non-finite running or corrected sums produce errors; values are not silently discarded.
These calculators give general mathematical results. They suit study, planning and estimation — verify anything that feeds a graded submission, a published figure or a financial commitment.
This record describes the implementation used by this page. It is not independent validation. External references, when available, support only the stated formula or definition; estimates and custom rules retain their own assumptions.
x̄ = Σ xᵢ ÷ n
Each value has equal weight. Separate values with commas or spaces; use semicolons with decimal commas, without thousands separators. Floating-point calculations have limited precision; displayed decimals are rounded. Invalid entries or a non-finite total produce an error.
Use semicolons between values with decimal commas: 1,5; 2,5; 3,5 gives an average of 2.5. With decimal points, use commas or spaces: 1.5, 2.5, 3.5. Do not use thousands separators: enter one thousand as 1000, not 1,000, which would be read as two values.
Because it does not have to be. The mean of 1, 2 and 4 is 2.33, a number nobody in the group has. It is a summary, not an observation, and that distinction matters when the mean is then used as if it were typical. For counts of people or items, always round deliberately and say which way you rounded.
Add the numbers and divide by their count: 10, 20, 30 gives 60 ÷ 3 = 20. Separate entries with commas or spaces. For decimal commas, use semicolons between entries, such as 1,5; 2,5. This tool calculates the arithmetic mean.