Learn how to find the median by arranging values from smallest to largest and locating the center. See why an even-sized set uses the average of its two middle values, and how the median differs from the mean, mode, and range—useful context when numbers have an uneven spread.

Multiple Choice

What is the median in a data set?

Median is the middle value of a data set when the values are arranged from smallest to largest. To find it, order the data and pick the central value; if there is an even number of observations, take the average of the two middle values. This divides the data into two halves with equal numbers of observations. For contrast, the most frequently occurring value is the mode; the sum of all values divided by the number of scores is the mean; the range is the difference between the largest and smallest values. Example: 3, 1, 4, 2, 5 → ordered: 1, 2, 3, 4, 5 → median is 3. If the set were 1, 2, 3, 4 → middle values are 2 and 3, so the median is 2.5.

A median can tell a clearer story than an average when a few unusually high or low values pull a data set off balance. In environmental and public health work, that distinction matters. A handful of very high readings can raise the mean, even when most measurements sit much lower. The median offers another way to describe the center: it marks the point where half the observations fall below and half fall above.

The middle has a precise meaning

To find the median, first arrange the values from smallest to largest. Then locate the center of the ordered list.

Take the numbers 3, 1, 4, 2, and 5. In order, they become 1, 2, 3, 4, 5. The middle value is 3, so the median is 3.

The order matters. If the same values appear in a different sequence—say, 5, 1, 3, 4, 2—the middle position in that unsorted list does not tell you the median. Sorting puts the data into a useful shape, like lining up a group by height before identifying the person in the middle.

When the number of observations is odd, there’s one central value. When it’s even, there are two central values, and the median is their average. For 1, 2, 3, and 4, the central pair is 2 and 3. Their average is 2.5, so the median is 2.5. It doesn’t need to be one of the original observations. It’s the value halfway between the two middle observations.

Why the median can be more useful than the mean

The mean is the familiar arithmetic average: add all the values, then divide by how many values there are. It’s useful, but it can be sensitive to outliers—observations that sit far from the rest.

Imagine five household water samples with nitrate readings of 2, 3, 3, 4, and 18 milligrams per liter. The median is 3. The mean is 6. The high reading pulls the mean upward, so it no longer describes a typical sample particularly well. The median stays near the cluster where most of the readings fall.

That doesn’t make the median universally better. The mean uses every value’s size, while the median focuses on position. If unusually high readings are important to the public health picture, smoothing their influence away could hide useful information. A sound summary may report both the median and the mean, then explain why they differ.

This is especially helpful when data are skewed. A distribution is skewed when values stretch farther on one side than the other. For example, most homes might have low levels of a contaminant while a small number have much higher levels. The long tail of high readings can pull the mean upward; the median is less affected.

A quick tour of nearby terms

Several measures of a data set’s center or spread can sound similar, but each answers a different question.

  • Median: What is the middle value after the data are ordered?

  • Mean: What is the arithmetic average?

  • Mode: Which value occurs most often?

  • Range: What is the difference between the largest and smallest values?

Consider the set 1, 2, 2, 4, 9. The median is 2, the mode is 2, the mean is 3.6, and the range is 8. They describe different features, even though they all come from the same five values.

A data set can have more than one mode if several values tie for the highest frequency. It can also have no useful mode when every value appears only once. The median, by contrast, can be found whenever the observations can be put in order and there is at least one value.

What “half on each side” really means

It’s common to say the median divides a data set into two equal halves. More precisely, it divides the ordered observations into groups of equal size, or as close to equal as possible. With an odd number of observations, the central value itself sits between the two groups. With an even number, the two groups contain the same number of observations, and the median lies between the middle pair.

This is about the number of observations, not the total of their values. In the ordered set 1, 2, 3, 4, 100, the median is 3. Four of the five values are at or below 4, but the single value 100 is far larger than the rest. The median doesn’t mean the data are evenly spread around the center; it only identifies the center by rank.

That distinction is easy to miss. A median of 3 does not tell us whether the nearby values are 2 and 4 or 2 and 90. To understand the full picture, pair the median with a measure of spread, such as the range or interquartile range, and consider the distribution itself.

Where this shows up in environmental health

Environmental health data often contain uneven patterns. A monitoring program may record air pollutant levels across several locations, water quality readings from different wells, or the number of days needed to complete an inspection. Some locations or time periods may produce extreme values. The median can offer a steady summary of what’s typical across those observations.

For example, a small set of inspection times might be 20, 22, 24, 25, and 70 minutes. The median is 24 minutes, while the mean is 32.2. The 70-minute visit may have involved a complicated site, a long interview, or an unexpected issue. It’s still worth noting, but it may not represent the usual inspection time.

In a public report, a median can also be easier to explain: “Half of the recorded readings were below this value, and half were above it.” That’s a useful statement, provided the data have been collected and summarized carefully. A single median cannot reveal which neighborhoods were sampled, whether readings changed by season, or whether the measurement method was consistent. Numbers need context; otherwise, a tidy summary can give a misleading sense of certainty.

A few details worth checking

Before calculating a median, confirm that the values are comparable and that none have been entered incorrectly. A misplaced decimal point can shift the order dramatically. So can mixing units, such as recording some temperatures in Celsius and others in Fahrenheit.

Also check whether missing observations were handled transparently. If a data set includes only the available readings, its median describes those readings—not necessarily every location or household of interest. And if observations come from very different groups, one combined median may hide meaningful differences. Separate medians by region, season, or population group may reveal a more useful pattern.

For a large data set, software can sort the observations and calculate the median quickly. The underlying idea stays the same: arrange values by size, then identify the center. A spreadsheet can do the arithmetic, but it can’t decide whether the data belong together or whether the result answers the real question. That part calls for judgment.

A simple measure, not a complete story

The median is a sturdy, easy-to-understand measure of the center of ordered data. It is especially helpful when a few extreme values make the mean feel unrepresentative. But it doesn’t describe every feature of a distribution, and it shouldn’t be treated as a substitute for looking at the data.

Think of it as one marker on a map. It tells you where the midpoint lies, but not how steep the hills are or how far the landscape stretches. In environmental and public health work, that marker can be valuable—particularly when paired with clear methods, context, and a measure of spread.