Mean Calculator
Compute the arithmetic mean (average) of any comma-separated list of numbers — the most common measure of central tendency in statistics. The result is the sum divided by the count, equivalent to balancing a see-saw of the data points.
Last updated: May 2026
Compare with similar
About this calculator
The arithmetic mean is defined as the sum of all values divided by the count: x̄ = Σxᵢ / n, where xᵢ are individual observations and n is the sample size. The calculator parses your comma-separated input, ignores non-numeric tokens, and computes the mean only over the valid numbers. Statistical properties: the mean is the value that minimises the sum of squared deviations from the data (the ‘least squares’ estimator of central tendency), and for a normally distributed population it is also the maximum-likelihood estimator. It is the building block of variance, standard deviation, t-tests, regression, and most parametric statistics. Variables: x̄ is the sample mean, xᵢ is the i-th observation, n is the sample size. The result has the same units as the input data. Edge cases: a single value returns that value as both ‘the mean’ and the data itself. An empty input or all-NaN input returns 0 by convention. The mean is highly sensitive to outliers — a single extreme value can shift it dramatically while the median barely moves; this is the main reason to choose median over mean for skewed distributions like incomes, house prices, or response times. For weighted averages where some observations should count more than others, use a weighted-mean calculator instead. For very large datasets (thousands of points), enter via a spreadsheet rather than typing into this field.
How to use
Example 1 — Test scores. A student scored 78, 85, 92, 88, and 79 on five exams. Enter values = 78,85,92,88,79. Sum = 78+85+92+88+79 = 422. n = 5. Mean = 422/5 = 84.4. ✓ Verify: the mean lies between the lowest (78) and highest (92) values, as it always must. Example 2 — Skewed data showing why the mean misleads. Five annual salaries at a small company: 35000, 38000, 42000, 45000, 320000 (a CEO). Enter values = 35000,38000,42000,45000,320000. Sum = 480,000. n = 5. Mean = 480,000/5 = $96,000. ✓ The mean of $96,000 is higher than four of the five salaries — none of these workers makes anywhere near it. The median (middle value when sorted = $42,000) is a far more honest ‘typical’ summary. This is exactly the failure mode that makes mean-based reporting misleading for income, home prices, response times, and other right-skewed data.
Frequently asked questions
When should I use the mean versus the median?
Use the mean when your data is numeric, roughly symmetric, free of extreme outliers, and you plan to do further calculations like variance, standard deviation, regression, or hypothesis tests — all of which build on the mean. Use the median when data is skewed or contains outliers: incomes, house prices, response times, file sizes, and recovery times all show heavy right tails where one or two extreme values pull the mean far above the typical observation. A classic illustration: ‘average’ US household income is meaningfully higher than the median because high earners pull the mean upward, while the median better describes what a typical household experiences. Reporting both together gives the fullest picture; if mean and median agree closely the distribution is roughly symmetric and any works. If they diverge, the divergence itself signals skewness or outliers worth investigating.
Why is the mean so sensitive to outliers?
The mean uses every value equally weighted in the sum, so adding one extreme observation changes the total by exactly that value while the divisor (n) increases by only 1 — net effect can be enormous on the average. Concrete example: data {1, 2, 3, 4, 5} has mean 3 and median 3. Add one outlier of 100: mean becomes (1+2+3+4+5+100)/6 ≈ 19.2, while the median moves only from 3 to 3.5. The mean tripled while the median barely shifted. This breakability is exactly why the median is preferred for ‘typical value’ reporting whenever the data has heavy tails or contamination by outliers, and why news stories that quote ‘average’ income reliably overstate what most people actually earn. For statistical robustness, the trimmed mean (drop the top and bottom k% before averaging) trades some efficiency for outlier resistance.
How is a weighted mean different from the arithmetic mean?
The arithmetic mean treats every observation equally; the weighted mean assigns importance (‘weights’) to each value: x̄_w = Σ(wᵢ × xᵢ) / Σwᵢ. The arithmetic mean is the special case where all weights equal 1. For a GPA, grades are weighted by credit hours: a 4-credit A counts more than a 1-credit A. For a portfolio return, individual asset returns are weighted by dollar allocation. To compute by hand: multiply each value by its weight, sum the products, then divide by the sum of all weights. This calculator computes only the unweighted (equal-weight) arithmetic mean. For grades use a GPA calculator, for portfolios a weighted-return calculator, and for survey data with different sample sizes per stratum use a stratified-mean tool that accounts for the sampling design.
What are the most common mistakes people make computing means?
The single most common mistake is averaging averages without re-weighting — if Class A averaged 85 across 30 students and Class B averaged 75 across 10 students, the combined mean is not 80 but (85×30 + 75×10) / 40 = 82.5. The second is using the mean on percentages, ratios, or growth rates where the geometric mean is mathematically correct (e.g., averaging annual returns of +10%, −5%, +15% requires geometric, not arithmetic, mean). The third is including outliers that should have been flagged as data errors rather than averaged in. The fourth is forgetting that the mean has the same units as the data — averaging 70 °F with 80 °C without converting produces nonsense. The fifth is presenting the mean of a small sample as if it were the population mean; with only 5 data points, uncertainty around that mean is large, and a single additional value can shift it noticeably. Always pair the mean with a sample size and a measure of spread (standard deviation or interquartile range).
When should I not use this calculator?
Skip it for datasets longer than a few dozen values that would be tedious to type — use a spreadsheet’s AVERAGE function or a statistics package. Do not use it for categorical or ordinal data where the mean has no meaning (favourite colour, customer satisfaction labels); for those, use the mode or a frequency table. It is the wrong tool for percentages and growth-rate averages, which require the geometric mean. Skip it for heavily skewed distributions where the mean misrepresents typical experience — report the median (or both) instead. Do not use it for weighted-average problems (GPA, portfolios, stratified samples) — use a weighted-mean calculator. Finally, never report only a mean without a measure of spread (SD, IQR, or range) and a sample size; the mean alone hides everything about the shape and uncertainty of the data.