Mean, Median, and Mode: Three Ways to Find the “Typical” Value

Not a member? Read it here.

When we look at data, one of the first questions we often ask is:

What is typical?

At first, this sounds like a simple question. But in statistics, “typical” can mean different things depending on what we are trying to understand.

Imagine we ask five people a simple question:

How many children do you have?

Their answers are:

0, 1, 2, 2, 5

Now we want to describe the center of this small dataset. In other words, we want to summarize these five values using one number.

But which number should we choose?

There are three common ways to describe the center of a distribution:

mode, median, and mean.

Each one answers a slightly different question.

1. Mode: What Value Appears Most Often?

The mode is the most common value in a dataset.

In our example:

0, 1, 2, 2, 5

The number 2 appears twice. Every other value appears only once.

So the mode is:

2

The mode answers the question:

What value happens most often?

This can be useful when we care about frequency. For example, if a shop wants to know the most commonly sold shoe size, the mode is useful. If one shoe size appears more often than all others, that is probably the size the shop should keep in stock.

But the mode is not always very helpful.

For continuous data, like income measured down to the dollar or even the cent, exact repeated values may not happen very often. Two people might both earn around $50,000, but one person earns $50,321.17 and another earns $50,318.92. Technically, those are different values.

So for many statistical problems, especially with continuous data, the mode is less commonly used than the median or the mean.

2. Median: What Is the Middle Value?

The median is the middle value when the data is ordered.

Our dataset is already ordered:

0, 1, 2, 2, 5

There are five values. The middle value is the third one:

2

So the median is:

2

The median answers the question:

What value has about half the data below it and half the data above it?

This is why the median is often used when talking about income, house prices, or anything where extreme values can distort the picture.

For example, when someone says “median income,” they mean that roughly half of people earn less than that amount and roughly half earn more.

The median gives us a sense of the “middle person.”

What Happens With an Even Number of Values?

The median is simple when we have an odd number of values. But what if we have an even number?

Suppose we add one more person who has zero children:

0, 0, 1, 2, 2, 5

Now there are six values.

There is no single middle value. The two middle values are:

1 and 2

So we take their average:

(1 + 2) / 2 = 1.5

The median is:

1.5

This may feel strange because nobody in the dataset has exactly 1.5 children. But that is fine. The median is not always one of the actual observed values. It is a summary of where the center lies.

3. Mean: What If Everything Were Shared Equally?

The mean is what most people usually call the “average.”

To calculate the mean, we add all the values together and divide by the number of values.

For our original dataset:

0, 1, 2, 2, 5

We add them:

0 + 1 + 2 + 2 + 5 = 10

There are five people, so we divide by five:

10 / 5 = 2

So the mean is:

2

The mean answers a different question from the median.

It asks:

What would each person have if the total were divided equally?

In this example, there are 10 children across 5 people. If we evenly distributed the children across everyone, each person would have 2 children.

Of course, we do not literally redistribute children. But this interpretation helps us understand what the mean is doing mathematically.

The mean spreads the total equally across all observations.

Why Outliers Matter

The mean is powerful, but it has one major weakness:

It is sensitive to outliers.

An outlier is an unusually large or unusually small value.

In our dataset:

0, 1, 2, 2, 5

The value 5 is higher than the others. It pulls the mean upward.

Now imagine something extreme:

0, 1, 2, 2, 5,000,000

The median would still stay close to the middle of the dataset. But the mean would become enormous, because the total sum is now dominated by one extreme value.

This is exactly why mean income can sometimes seem misleading.

If a few people earn extremely high salaries, they pull the mean upward. The average income may look high, even though most people earn much less.

That is why median income is often more useful when we want to describe the typical person.

The median is resistant to outliers.

The mean is not.

Mean vs Median: Which One Should You Use?

It depends on the question you want to answer.

If you ask:

“What does the middle person look like?”

Use the median.

If you ask:

“What would everyone get if the total were shared equally?”

Use the mean.

If you ask:

“What value occurs most often?”

Use the mode.

These are not just different formulas. They describe different ideas of “typical.”

A Simple Example With Income

Imagine five people have these yearly incomes:

$30,000, $35,000, $40,000, $45,000, $1,000,000

The median income is:

$40,000

That tells us the middle person earns $40,000.

But the mean is:

($30,000 + $35,000 + $40,000 + $45,000 + $1,000,000) / 5
= $230,000

The mean income is $230,000.

But does that describe a typical person in this group?

Not really.

Four out of five people earn far less than $230,000. The mean is being pulled upward by the one person who earns $1,000,000.

This is why the median can be more useful when data is skewed or contains extreme values.

The Big Idea

When we describe a distribution, we often want to describe two things:

1. Its shape
2. Its center

The shape tells us whether the data is symmetric, skewed, unimodal, or bimodal.

The center tells us where the data tends to be located.

Mean, median, and mode are three ways to describe that center.

But they are not interchangeable.

The mode tells us the most common value.

The median tells us the middle value.

The mean tells us the equal-share value.

In our small example:

0, 1, 2, 2, 5

All three measures point to the same number:

2

But that will not always happen.

In real datasets, especially datasets with outliers, the mean and median can tell very different stories.

And that is the important lesson:

Statistics is not just about calculating numbers. It is about understanding what question each number is answering.