Average Calculator — Mean, Median, Mode, Weighted & Geometric Mean with Steps

What is an average calculator?

An average calculator reduces a list of numbers to a single value that stands for the whole list. The most common type of average is the arithmetic mean — add up every value and divide by how many there are. But in statistics, "average" is not one number; it's a family of measures, each answering a slightly different question about the data. Choosing the wrong one can give you a technically correct answer that still misleads.

This tool computes the full picture at once: the arithmetic mean, median, mode, geometric mean, harmonic mean, RMS, and weighted average, plus all the statistics that describe how spread out and how shaped the data is — variance, standard deviation, standard error, coefficient of variation, quartiles, IQR, skewness, kurtosis, and a confidence interval. An interactive histogram with sigma bands and a fitted normal curve shows the shape of the data at a glance, and a plain-English summary explains what the numbers mean.

Three input modes cover the common workflows. Simple list accepts any paste-in list of numbers and parses commas, spaces, tabs, and line breaks. Weighted assigns importance to each value — the standard way to compute a GPA, a weighted course grade, or a portfolio return. Running tally adds values one at a time, which is handy when you're counting things in real time.

How to use the average calculator

1. Pick a mode. Simple list for a paste-in dataset, Weighted if values have different importance, or Running tally to add numbers as you go.
2. Enter your numbers. Commas, spaces, tabs, and line breaks all work. Anything that isn't a number is ignored, so you can paste a column from Excel or Google Sheets directly.
3. Confirm the count. The pill above the textarea shows how many numbers were parsed, so you can verify nothing was missed.
4. Read the results. The mean appears at the top with sum, count, and a 95% confidence interval bar, followed by a mean-vs-median comparison, a plain-English interpretation, the interactive charts, and the essential statistics.
5. Explore. Drag the percentile slider to see the value at any point in the distribution. Hover over the histogram to see bin ranges. Expand the advanced section for skewness, kurtosis, confidence interval details, and the empirical rule.
6. Try what-if scenarios. See how the mean changes if you add or remove a value from the dataset.

💡 Tip: When the mean and median differ noticeably (visible in the comparison strip), the data is skewed. The interpretation card explains which one better represents a typical value.

The formulas behind every statistic

Arithmetic mean (the standard average)

Example. Test scores 78, 85, 92, 70, 88. The sum is 413, and the count is 5, so the mean is 82.6.

Median

Sort the values and take the middle. For an odd count, it's the middle value. For an even count, it's the average of the two middle values.

Example. Sorted 70, 78, 85, 88, 92 — the median is 85. For 70, 78, 85, 88, 92, 95, the two middle values are 85 and 88, so the median is (85 + 88) ÷ 2 = 86.5.

Mode

The value that appears most often. A dataset can have one mode, several modes, or no mode at all — if every value is unique, there is no mode.

Example. 70, 78, 85, 85, 92 → the mode is 85. 1, 1, 2, 2, 3 → the modes are 1 and 2 (bimodal). 1, 2, 3, 4, 5 → no mode.

Variance and standard deviation

Variance measures how far values spread from the mean. Population variance divides by N; sample variance divides by N−1 (Bessel's correction) and is the right choice when your data is a sample from a larger population.

The standard deviation is the square root of variance — s = √(s²). It has the same units as the data, which is why it's the more interpretable number. Standard error (SEM) divides by √n to express how precisely the mean estimates the true population mean.

Geometric mean

Requires all values to be positive. Use it for growth rates, investment returns, ratios, and anything that compounds.

Worked example. A portfolio returns +50% one year and −50% the next. Arithmetic mean = (+50 + −50) ÷ 2 = 0%, but the actual result is 1.5 × 0.5 = 0.75, a −25% loss. The geometric mean is √(1.5 × 0.5) − 1 = √0.75 − 1 = 0.866 − 1 = −13.4%, which correctly reflects what actually happened.

Harmonic mean

Requires all values to be non-zero. Use it for rates, speeds, and ratios where the denominator varies.

Worked example. You drive 60 mph one way and 40 mph back over the same distance. The harmonic mean is 2 ÷ (1/60 + 1/40) = 2 ÷ (0.0167 + 0.025) = 2 ÷ 0.0417 = 48 mph — the actual average speed for the round trip. The arithmetic mean of 50 mph would be wrong.

Weighted mean

Worked example — GPA. Course grades 92 (3 credits), 88 (2 credits), 78 (4 credits), 85 (3 credits). Weighted mean = (92×3 + 88×2 + 78×4 + 85×3) ÷ (3 + 2 + 4 + 3) = (276 + 176 + 312 + 255) ÷ 12 = 1019 ÷ 12 = 84.92. The arithmetic mean of 85.75 would overstate the result because the lowest grade carries the most credits.

Skewness and kurtosis

Skewness measures the asymmetry of the distribution; kurtosis measures how heavy the tails are. The sample-adjusted skewness used by this calculator is:

Positive skew (right tail longer) pulls the mean above the median — as with income data, where a few very high earners inflate the mean. Negative skew (left tail longer) pulls the mean below the median. Values near zero indicate a symmetric distribution. Kurtosis is interpreted as "excess kurtosis": 0 means normal tails, values above 0 mean heavier tails (more extreme outliers than a normal distribution), and values below 0 mean lighter tails.

95% confidence interval for the mean

The interval estimates the range in which the true population mean is likely to lie, with 95% confidence. Larger samples produce narrower intervals — more data means more certainty about the center.

When to use each type of average

Why the mean and median diverge — the income example

Five people earn $40K, $55K, $60K, $75K, and $300K. The mean income is $106K; the median is $60K. The mean is what each person would have if the total were split evenly. The median is what the person in the middle of the lineup actually makes. In any dataset with a long right tail — income, wealth, house prices, follower counts — the mean is pulled upward by a small number of very large values. That's why median household income, not mean income, is the statistic governments and news outlets report.

This calculator makes the divergence visible. When the mean and median differ by more than about 5% of the standard deviation, the data is skewed, and the mean-to-median comparison strip at the top of the results will show which direction. The histogram's sigma bands and normal curve overlay make the shape of the data obvious at a glance.

Real-world use cases

• Students and teachers. Calculate average test scores, GPA, or weighted course grades. The weighted mode handles course credits directly.
• Business analytics. Average sales per day, average customer order value, average employee productivity — with standard deviation and confidence interval to communicate variability and uncertainty.
• Finance. Average returns (use the geometric mean for compounding returns), average P/E ratios, average portfolio weights. The geometric mean is essential for measuring true investment performance.
• Science and research. Mean, standard deviation, standard error, and 95% confidence intervals for experimental results — the standard trio reported in almost every scientific paper.
• Sports analytics. Batting averages, points per game, average possession percentage — often alongside median values to spot skewed performances.
• Quality control. Six Sigma, control charts, and process monitoring all rely on the mean and standard deviation of measurement data.
• Everyday decisions. Average grocery spend, average commute time, average ratings from review data.

Understanding outliers and shape

An outlier is a value that sits unusually far from the rest of the data. The standard statistical test is the Tukey rule: any value more than 1.5 × IQR below Q1 or above Q3 is considered an outlier. The IQR (interquartile range) is the distance between the 25th and 75th percentiles — the middle 50% of the data.

Outliers aren't automatically wrong. A salary outlier in a company dataset might be a CEO whose pay is legitimately much higher than everyone else's. An outlier in a set of experimental measurements might indicate a genuine anomaly worth investigating. The important thing is to notice that the outlier exists, understand its effect on the mean, and decide whether to use a more robust statistic like the median.

Skewness and kurtosis round out the picture of the distribution's shape. Skewness tells you whether the tail is heavier on the right or left; kurtosis tells you whether the tails are heavier or lighter than a normal distribution. Together they help you decide which averages and which statistical tests are appropriate for your data.

Common mistakes

• Using the arithmetic mean for compounding data. Investment returns, growth rates, and ratios require the geometric mean. A 10% return followed by a 20% return averages to 14.89% geometrically, not 15% arithmetically.
• Using the arithmetic mean for speeds. If you drive at different speeds over equal distances, the harmonic mean is the correct average speed.
• Forgetting that mode can be missing. If every value in a dataset appears once, there is no mode — not "the value that appears once."
• Mixing population and sample standard deviation. Use population (divide by N) when you have the entire population, and sample (divide by N−1) when your data is a sample from a larger group.
• Reading the mean as typical. In a skewed dataset, the mean is not the typical value. Use the median.
• Ignoring sample size. A mean from 3 data points is far less reliable than a mean from 300. The standard error and confidence interval communicate this difference.

Limitations

• The calculator does not compute hypothesis tests, regression coefficients, or ANOVA. Those require a dedicated statistics package.
• Weighted mode requires the values and weights to have the same count. A red badge above the input area flags any mismatch.
• Geometric mean requires all values to be positive; harmonic mean requires all values to be non-zero. Both display "N/A" when those conditions aren't met.
• Outlier detection uses the standard 1.5 × IQR rule. Some fields (like finance) use 3 × IQR; the highlighted outliers are a starting point, not a verdict.
• The 95% confidence interval assumes a roughly normal sampling distribution. For very small samples (n < 30), a t-distribution-based interval would be more accurate.

Frequently asked questions

Related calculators on AIToolsPros