Mean vs. Median vs. Mode

Executive Summary

Mean, median, and mode are the three measures of central tendency — each answers the question "what is a typical value in this dataset?" in a different way. The mean (arithmetic average) is the most commonly used but is sensitive to extreme values. The median (middle value) is resistant to outliers. The mode (most frequent value) works for both numerical and categorical data. Choosing the right measure for a given situation is one of the most important decisions in descriptive statistics.

Key Takeaways
  • Three Different 'Average': Mean, median, and mode each define 'typical' differently and give different answers for the same dataset.
  • Use Median When Data Is Skewed: Outliers pull the mean away from the centre. The median is immune to outliers and better represents a 'typical' value in skewed distributions.
  • Mode Is for Categories: Mode is the only measure of central tendency that works for non-numerical (categorical) data — like the most common blood type or most-purchased product.
Visual Explanation

How Skew Separates Mean from Median

In a perfectly symmetric dataset (like heights in a large population), mean, median, and mode are all equal — they all point to the same central value. But in a skewed distribution (like household incomes), these three values diverge sharply. In right-skewed income data, a small number of very high earners pull the mean significantly to the right, while the median stays near the true centre. This is why "average income" overstates what most households actually earn.

Mean vs. Median Divergence in Skewed Data (Annual Incomes, 000s $)

Individual Income
Individual Income
$0$150$300$450$600Person 1Person 5
Formula
Mean = Σxᵢ / n
Median = Middle value of sorted set

Descriptive Statistics Formulas

Variable Glossary
SymbolMeaning & Description
MeanArithmetic Average (Sum of values divided by count)
MedianMiddle value of the sorted dataset
ModeValue(s) that occur with the highest frequency
Step-by-Step Worked Example
1. Mapped Variables
Data Set Input
3, 7, 5, 13, 7, 2, 9, 7, 5
2. Equation Substitution
Equation with standard inputs
Mean = 58 / 9
3. Calculation Steps
Step 1: Calculate Mean

Mean = Sum / Count = 58 / 9 = 6.4444

Step 2: Find Median (Middle value)

Sorted dataset: [2, 3, 5, 5, 7, 7, 7, 9, 13]. Median = 7

Final Resolved Arithmetic Mean6.4444

The Three Formulas

Mean (Arithmetic Average):

xˉ=i=1nxin=x1+x2++xnn\bar{x} = \frac{\sum_{i=1}^{n} x_i}{n} = \frac{x_1 + x_2 + \cdots + x_n}{n}

Add all values, divide by the count. Sensitive to outliers.

Median (Middle Value):

  • Sort the dataset in ascending order.
  • If n is odd: median = the middle value.
  • If n is even: median = average of the two middle values.

Resistant to outliers.

Mode (Most Frequent Value):

The value that appears most often in the dataset. A dataset can have:

  • No mode (all values appear once)
  • One mode (unimodal)
  • Two modes (bimodal)
  • Multiple modes (multimodal)

Worked Comparison

Dataset: {3, 7, 5, 13, 7, 2, 9, 7, 5}

Mean: (3+7+5+13+7+2+9+7+5) ÷ 9 = 58 ÷ 9 = 6.44

Median: Sort → {2, 3, 5, 5, 7, 7, 7, 9, 13}. Middle value = 7

Mode: 7 appears 3 times (most frequent) = 7

Note that in this dataset, median and mode agree (both = 7) but the mean (6.44) is lower due to the lower values dragging the average down.

Now with one outlier — add 1,000 to the dataset:

{2, 3, 5, 5, 7, 7, 7, 9, 13, 1000}

  • New Mean: 1058 ÷ 10 = 105.8 (wildly distorted)
  • New Median: (7 + 7) ÷ 2 = 7 (unchanged)
  • New Mode: 7 (unchanged)

The outlier (1,000) completely destroys the mean but has zero effect on the median and mode. This is why median is preferred for skewed data.

Mental Model & Analogy

The Salary at the Party Analogy

Imagine 9 people at a party with salaries of $40k, $45k, $50k, $50k, $55k, $55k, $60k, $65k, and $70k. The mean salary is $54,444. Now a billionaire walks in earning $1,000,000,000. The mean salary in the room instantly jumps to over $100 million — making it sound like everyone in the room is rich. The median (the 5th highest person in a now 10-person group) barely changes. The mode is still the salary that most people earn. The mean is the measure that was hijacked by the outlier.

Real-World Applications
  • Economic Reporting: The U.S. Census Bureau reports both mean and median household income. Median is the primary figure because the distribution of income is heavily right-skewed. Mean income is always higher than median income in unequal economies.
  • Real Estate: House price data uses median, not mean, as the standard measure. A single ultra-luxury listing can pull the mean up dramatically, while the median accurately reflects what most buyers pay.
  • Retail Analytics: Mode is used for inventory management — identifying which product sizes, colours, or SKUs sell most frequently to optimise stock levels.
  • Medical Research: Survival times in clinical trials (how long patients live after treatment) are reported as median survival time, not mean, because a small number of very long survivors would distort the mean upward.
Common Mistakes to Avoid
  • Using Mean for Skewed Data: Reporting mean income, mean house prices, or mean hospital wait times is almost always misleading because these distributions are heavily skewed. Median is the correct choice.
  • Assuming "Average" Means Mean: In everyday language, "average" usually means arithmetic mean. But statistically, any of the three measures of central tendency can legitimately be called an average.
  • Ignoring Bimodal Distributions: When a dataset has two peaks (e.g., bimodal), summarising it with a single mean or median hides the true structure. Always visualise data before summarising it.
  • Forgetting That Mode Can Be Multiple Values: Datasets can be bimodal or multimodal. Reporting only one mode when two exist omits important distributional information.
Concept Comparison

AMean vs. Median

Mean is the arithmetic centre of gravity — every value contributes equally, including outliers. Mean is ideal for symmetric, normally distributed data where extreme values don't exist or are rare. Mean supports more powerful statistical analyses (variance, standard deviation, confidence intervals) than median. Choose mean for scientific measurements, exam averages, and symmetric data.

BMedian vs. Mode

Median vs. Mode: Median is the physical middle value — 50% of data is above and below it. It is ideal for ordinal data and skewed distributions. Mode is the most common value — it is the only measure that works for categorical (non-numerical) data, like the most common blood type (O+) or most popular product colour. Use median for skewed quantitative data; use mode for categorical frequency analysis.

Decision Framework

Which Measure to Use

SituationBest MeasureWhy
Symmetric numerical data (heights, test scores)MeanMost representative; mathematically richest
Skewed numerical data (incomes, house prices)MedianOutlier-resistant; more representative of "typical"
Categorical / nominal data (colours, blood types)ModeOnly measure that works for non-numerical categories
Reporting to the publicMedianHarder to manipulate; more intuitive
Input into further statistical calculationsMeanVariance, SD, confidence intervals all built on mean
Distribution with two peaksBoth modesReport the bimodal structure; single summary is misleading
Frequently Asked Questions
Live Simulation

Mean, Median & Mode Calculator Sandbox

Tweak variables below to see the formula calculate instantly.

Calculating...
Interactive Tool

Mean, Median & Mode Calculator

Calculate the mean (average), median (middle value), mode (most frequent value), and range of any data set.

Launch Interactive Calculator