Average Calculator

The default list, worked through

The box starts with 10, 20, 25, 25, 30. The Sum is 110 and the Count is 5, so the Mean is 22. Sorted, the middle value is the third one, so the Median is 25. The value 25 appears twice and nothing else repeats, so the Mode is 25. Min is 10, Max is 30 and the Range is their difference, 20.

With an even count the median is the mean of the two middle values: for 1, 2, 3, 4 it is 2.5. When several values tie for the highest frequency, every one of them is listed, so 3, 7, 7, 9, 9 reports a mode of 7, 9. If every value is unique the Mode row says None.

How the text is turned into numbers

The parser scans the text for anything that looks like a number: an optional minus sign, digits, an optional decimal point and an optional exponent such as e3. Everything else is ignored, which is convenient for pasting messy text with currency symbols or labels mixed in.

Commas follow one rule. When the list also contains spaces or line breaks, a comma that sits between a digit and exactly three more digits is treated as a thousands separator, so 1,000 2,000 is read as 1000 and 2000 with a mean of 1,500. When commas are the only separator in the text, every comma splits: 100,200,300 is three values, and a lone 1,000 with no spaces around it becomes 1 and 0. A comma after a decimal fraction, as in 0.5,100 7, always splits.

How the results are formatted

• Values are trimmed to 12 significant digits and then displayed with at most three decimal places, so the sum of 0.1 and 0.2 is shown as 0.3 rather than the 0.30000000000000004 that floating-point addition returns, and a mean of 10 ÷ 3 appears as 3.333.
• Thousands separators are added using your browser's locale, so a sum of 1234.5 appears as 1,234.5 in English locales.
• Count is always a whole number. Everything else can be decimal, including the median of an even-length list.
• Negative numbers are handled throughout; a list of -5 and 10 has a mean of 2.5 and a range of 15.

Statistics it does not produce

The tool stops at the eight descriptive figures above. There is no standard deviation, variance, percentile, weighted mean or geometric mean. A weighted average can be imitated by repeating each value as many times as its weight: a score of 90 with weight 2 and 70 with weight 1 is entered as 90 90 70 and gives 83.333.

The mean can be dragged a long way by a single extreme value, while the median barely moves; comparing the two rows is the quickest check for outliers in a list of test scores, response times or prices.

Typical lists people paste

• Exam or assignment marks to find a class average and the middle score.
• Monthly bills or sales figures to see the typical month and the spread between best and worst.
• Sensor or lab readings where the mode shows the most frequently recorded value.
• Sports statistics, such as points per game, where Min, Max and Range summarise consistency.