Average Calculator: Mean, Median, Mode and More Explained

📖 Table of Contents

The Many Faces of "Average"
Arithmetic Mean
Median
Mode
Geometric Mean
Weighted Average
Moving Average
Which Average Should You Use?
Programming Implementation

The Many Faces of "Average"

When someone says "the average is…," they usually mean the arithmetic mean. But "average" is actually a broad term that encompasses several different measures of central tendency, each with distinct properties and use cases. Choosing the wrong type of average can lead to misleading conclusions — sometimes dramatically so.

Consider income data. In 2025, the United States had a median household income of roughly $80,000 but a mean (arithmetic average) household income of over $105,000. The mean is pulled upward by a relatively small number of very high earners. If a politician wanted to make incomes look high, they'd quote the mean; if they wanted to show what a "typical" household earns, they'd use the median. Neither is wrong — they answer different questions.

This guide covers every major type of average, explains when each is appropriate, and shows you how to calculate them all.

Arithmetic Mean

The arithmetic mean is what most people think of as "the average." It's calculated by summing all values and dividing by the count.

Mean = (x₁ + x₂ + ... + xₙ) / n = Σxᵢ / n

Example: The test scores 78, 85, 92, 88, 72 have a mean of:
(78 + 85 + 92 + 88 + 72) / 5 = 415 / 5 = 83

Properties of the Mean

Uses every data point: Every value contributes to the mean, making it informationally efficient.
Sensitive to outliers: A single extreme value can dramatically shift the mean. The dataset {1, 2, 3, 4, 100} has a mean of 22, which doesn't represent any of the "typical" values.
Unique algebraic property: The sum of deviations from the mean equals zero: Σ(xᵢ - x̄) = 0. This makes the mean the "center of gravity" of the data.
Minimizes squared errors: The mean is the value that minimizes the sum of squared deviations — this is why it's used in least squares regression.

When to Use the Mean

The mean is ideal for roughly symmetric distributions without extreme outliers. It's the standard measure in physics, engineering, and most scientific contexts. Use the mean when you need a mathematically efficient summary of your data and you're confident outliers aren't distorting it.

Median

The median is the middle value when data is sorted in order. If there's an odd number of values, it's the exact middle one. If there's an even number, it's the average of the two middle values.

Example (odd count): {12, 15, 18, 22, 28} → Median = 18 (3rd value out of 5)

Example (even count): {12, 15, 18, 22, 28, 31} → Median = (18 + 22) / 2 = 20

Properties of the Median

Resistant to outliers: The dataset {1, 2, 3, 4, 100} has a median of 3 — much more representative than the mean of 22.
Works with ordinal data: You can find the median of rankings (1st, 2nd, 3rd) even though you can't meaningfully calculate their mean.
Divides data in half: By definition, 50% of values fall at or below the median, and 50% fall at or above it.
Less algebraically convenient: The median doesn't have the nice mathematical properties of the mean, making it harder to work with in advanced statistical formulas.

When to Use the Median

The median is the preferred measure for skewed distributions and data with outliers. It's standard practice for income, housing prices, response times, and any data where a few extreme values could distort the mean. When the mean and median differ significantly, the data is skewed, and the median better represents the "typical" value.

Mode

The mode is the most frequently occurring value in a dataset. Unlike mean and median, a dataset can have no mode (all values occur equally), one mode (unimodal), two modes (bimodal), or more (multimodal).

Example: {2, 3, 3, 5, 7, 7, 7, 9} → Mode = 7 (appears 3 times)

Bimodal: {2, 2, 3, 5, 7, 7} → Modes = 2 and 7

When to Use the Mode

The mode is the only measure of central tendency that works with categorical (nominal) data. You can't calculate the "mean" or "median" eye color, but you can identify the mode (brown, for most populations). The mode is also useful for identifying the most common or popular item in a dataset — the most common shoe size sold, the most frequent customer complaint, or the peak of a distribution.

Geometric Mean

The geometric mean is calculated by multiplying all values together and taking the nth root. It's specifically designed for data that's multiplicative rather than additive.

Geometric Mean = (x₁ × x₂ × ... × xₙ)^(1/n) = (Πxᵢ)^(1/n)

In practice, it's easier to compute using logarithms:

Geometric Mean = exp[(ln(x₁) + ln(x₂) + ... + ln(xₙ)) / n]

Example: Investment returns of 5%, 10%, -3%, and 8% over four years.
Geometric Mean = [(1.05)(1.10)(0.97)(1.08)]^(1/4) = (1.205)^(0.25) = 1.0479
Average annual return ≈ 4.79%

Note: the arithmetic mean of these returns would be (5 + 10 - 3 + 8) / 4 = 5%, which overstates the actual compounded return.

When to Use the Geometric Mean

Investment returns: The geometric mean gives the true average compound growth rate over time.
Ratios and indices: When comparing things using ratios (price indices, population growth rates).
Log-normal distributions: Many natural phenomena (income, biological measurements) follow log-normal distributions where the geometric mean is the natural center.
Always positive data only: The geometric mean requires all values to be positive (or all negative).

Geometric vs. Arithmetic Mean

The geometric mean is always less than or equal to the arithmetic mean (AM-GM inequality). The gap between them increases with the variability of the data. When returns are volatile, the geometric mean can be significantly lower than the arithmetic mean — this difference represents the "volatility drag" that erodes compounded returns.

Weighted Average

A weighted average assigns different importance (weights) to different values. It's essential when some data points should count more than others.

Weighted Average = Σ(wᵢ × xᵢ) / Σwᵢ

Where wᵢ is the weight assigned to value xᵢ. The standard arithmetic mean is just a weighted average where all weights equal 1.

Example — GPA Calculation:
Course A: 3.7 × 4 credits = 14.8
Course B: 4.0 × 3 credits = 12.0
Course C: 3.3 × 3 credits = 9.9
Course D: 3.5 × 2 credits = 7.0

GPA = (14.8 + 12.0 + 9.9 + 7.0) / (4 + 3 + 3 + 2) = 43.7 / 12 = 3.64

Common Applications of Weighted Averages

Academic GPAs: Credits act as weights for course grades.
Portfolio returns: Each asset's return is weighted by its proportion of the portfolio.
Price indices: The Consumer Price Index weights goods by their share of consumer spending.
Survey aggregation: Weighting responses to match population demographics.
Machine learning: Weighted averages are fundamental to neural networks, ensemble methods, and attention mechanisms.

Moving Average

A moving average (or rolling average) calculates the average of a fixed number of consecutive data points, "moving" the window forward one observation at a time. It's primarily used for time series data to smooth out short-term fluctuations and reveal trends.

Types of Moving Averages

Simple Moving Average (SMA): Equal weight to all observations in the window.

SMA(t) = (xₜ + xₜ₋₁ + ... + xₜ₋ₙ₊₁) / n

Weighted Moving Average (WMA): More recent observations receive higher weights.

Exponential Moving Average (EMA): Applies exponentially decreasing weights to older observations, responding more quickly to recent changes.

EMA(t) = α × xₜ + (1 - α) × EMA(t-1)

Where α = 2/(n+1) is the smoothing factor.

Applications

Stock market analysis: The 50-day and 200-day SMAs are widely watched technical indicators. A "golden cross" (50-day crossing above 200-day) is considered bullish.
Weather forecasting: Smoothing daily temperature data to see seasonal trends.
Business metrics: Tracking 7-day or 30-day rolling averages of sales, active users, or error rates to filter out daily noise.
Signal processing: Moving averages are the simplest form of low-pass filters, removing high-frequency noise from signals.

Which Average Should You Use?

Scenario	Best Average	Why
Symmetric data, no outliers	Mean	Most information-efficient
Skewed data (income, prices)	Median	Resistant to outliers
Categorical data (eye color)	Mode	Only measure that works
Growth rates, returns	Geometric Mean	Correct for compounding
Items with different importance	Weighted Average	Accounts for varying weights
Time series trends	Moving Average	Smooths noise
Log-normal distributions	Geometric Mean	Matches the distribution's center
Grades with different credits	Weighted Average	Fair credit weighting

The best practice is often to report multiple measures. Reporting both mean and median immediately tells your audience whether the data is skewed. Including the mode reveals whether there's a dominant value that the other measures might obscure. Together, these three measures — the "big three" of central tendency — give a much more complete picture than any single average alone.

Programming Implementation

Here's how to calculate each type of average in popular programming languages:

Python

import statistics from functools import reduce import math data = [78, 85, 92, 88, 72] # Arithmetic Mean mean = statistics.mean(data) # 83.0 # Median median = statistics.median(data) # 85 # Mode mode = statistics.mode(data) # Note: raises error if no unique mode # Geometric Mean (Python 3.8+) geo_mean = statistics.geometric_mean(data) # 82.96 # Weighted Average values = [3.7, 4.0, 3.3, 3.5] weights = [4, 3, 3, 2] weighted_avg = sum(v * w for v, w in zip(values, weights)) / sum(weights) # 3.64 # Moving Average (3-period) def moving_average(data, window): return [sum(data[i:i+window]) / window for i in range(len(data) - window + 1)]

JavaScript

const data = [78, 85, 92, 88, 72]; // Arithmetic Mean const mean = data.reduce((a, b) => a + b) / data.length; // 83 // Median const sorted = [...data].sort((a, b) => a - b); const mid = Math.floor(sorted.length / 2); const median = sorted.length % 2 ? sorted[mid] : (sorted[mid - 1] + sorted[mid]) / 2; // 85 // Mode const freq = {}; data.forEach(v => freq[v] = (freq[v] || 0) + 1); const maxFreq = Math.max(...Object.values(freq)); const mode = +Object.keys(freq).find(k => freq[k] === maxFreq); // Geometric Mean const geoMean = Math.pow( data.reduce((a, b) => a * b), 1 / data.length ); // Weighted Average const values = [3.7, 4.0, 3.3, 3.5]; const weights = [4, 3, 3, 2]; const wAvg = values.reduce((sum, v, i) => sum + v * weights[i], 0) / weights.reduce((a, b) => a + b);

SQL

-- Arithmetic Mean SELECT AVG(score) AS mean_score FROM students; -- Median (PostgreSQL) SELECT percentile_cont(0.5) WITHIN GROUP (ORDER BY score) FROM students; -- Mode (PostgreSQL) SELECT mode() WITHIN GROUP (ORDER BY score) FROM students; -- Weighted Average SELECT SUM(grade * credits) / SUM(credits) AS gpa FROM courses;

Conclusion

"Average" is not a single concept — it's a family of measures, each optimized for different situations. The arithmetic mean is the workhorse for symmetric data, the median is your shield against outliers, the mode handles categorical data, the geometric mean captures multiplicative processes, and weighted averages account for varying importance. Understanding which average to use — and why — is fundamental to data literacy. Report multiple measures when possible, and always consider the shape of your distribution before choosing a single summary statistic.

🔢 Calculate Any Type of Average Instantly

Our free Average Calculator computes mean, median, mode, geometric mean, weighted average, and more — with step-by-step solutions.

Try the Calculator →