Quartiles split ordered data into four sections, each representing about 25% of the observations. Learn how Q1, Q2, and Q3 relate to percentiles and the median, and how these measures help describe a data set’s center, spread, and possible outliers.

Multiple Choice

What are quartiles used for?

Quartiles are a statistical tool used to divide a dataset into four equal parts, with each part representing 25% of the data. This means that when you arrange the data in ascending order, the first quartile (Q1) marks the 25th percentile, the second quartile (Q2) corresponds to the median or 50th percentile, and the third quartile (Q3) indicates the 75th percentile. By using quartiles, you can easily analyze the distribution of data, identify outliers, and gain insights into the spread and center of the data set. This method is beneficial when assessing the relative standing of a particular value within the context of the overall data distribution. The understanding of quartiles is essential in both descriptive statistics and exploratory data analysis.

Quartiles turn a long list of numbers into a simple picture of how the values are spread. They split ordered data into four groups, each representing roughly 25% of the observations. That makes them useful for describing a typical range, comparing groups, and spotting values that sit far from the rest. A few well-chosen quartile figures can reveal patterns that a single average might hide.

Four groups, three dividing lines

Imagine arranging a set of values from smallest to largest. Quartiles mark three points along that ordered list:

  • First quartile (Q1): the point below which about 25% of the data falls.

  • Second quartile (Q2): the median, with about half the data below it and half above.

  • Third quartile (Q3): the point below which about 75% of the data falls.

These points create four sections: the lowest quarter of values, the next quarter, the third quarter, and the highest quarter. The boundaries are Q1, Q2, and Q3.

The word “about” matters. If a dataset has a number of observations that doesn’t divide neatly into four, a quartile boundary may fall between two values. Different calculation methods can also produce slightly different results, especially for small datasets. The central idea stays the same: quartiles show where values sit in an ordered distribution.

A small example makes it clear

Consider these nine delivery times, listed in minutes:

12, 15, 18, 20, 22, 24, 27, 31, 45

The middle value is 22, so Q2—the median—is 22. The values below it form the lower half, and those above it form the upper half. Using a common method for this example, the middle of the lower half is 15, so Q1 is 15. The middle of the upper half is 31, so Q3 is 31.

The quartiles now tell a compact story. About a quarter of deliveries took 15 minutes or less; about half took 22 minutes or less; and about three-quarters took 31 minutes or less. The 45-minute delivery sits well above the upper quartile, which may be worth a closer look. Perhaps there was heavy traffic, a routing issue, or simply an unusual day.

Notice what the quartiles don’t tell us: they don’t explain why that delivery took longer. Statistics can point to a pattern, but context supplies the story.

Why a median can be more useful than a mean

Quartiles often appear alongside the median because both help describe data without being pulled around by unusually high or low values. The mean, or arithmetic average, can shift a lot when one observation is far from the rest.

Suppose several employees earn similar salaries, but one senior executive earns far more. The mean salary may rise, even though it doesn’t describe what most employees earn very well. The median gives the value at the center of the ordered list, while Q1 and Q3 show where the middle half of salaries lies.

That middle half is especially useful. It gives a sense of the typical range without letting a few extreme observations dominate the picture. Quartiles don’t replace the mean in every situation; they answer a different question. The mean uses every value in its calculation. Quartiles focus on position within the ordered data.

The interquartile range: a quick look at spread

Subtract Q1 from Q3 and you get the interquartile range, usually shortened to IQR:

IQR = Q3 − Q1

The IQR measures the spread of the middle 50% of the data. In the delivery example, Q3 is 31 and Q1 is 15, so the IQR is 16 minutes. That means the central half of the delivery times spans 16 minutes, from the first quartile to the third.

A small IQR suggests that the middle values are clustered fairly closely. A larger IQR suggests more spread among them. This measure is less sensitive to extreme values than the overall range, which runs from the minimum to the maximum. If one delivery takes an unusually long time, the range may balloon. The IQR is less affected because it focuses on the middle portion.

That distinction comes in handy when comparing data. A store might have a higher average wait time than another, but a smaller IQR. In plain terms, its wait times could be longer on average yet more consistent. One summary number rarely says everything.

Quartiles help flag unusual values

Quartiles also provide a common way to identify possible outliers. A value may be flagged if it falls below:

Q1 − 1.5 × IQR

or above:

Q3 + 1.5 × IQR

These cutoffs are often used when creating a box plot or exploring a dataset. A value outside them is a signal to investigate, not automatic proof of an error. It could reflect a data-entry mistake, a rare event, or a genuine feature of the group being measured.

For example, a very high household electricity bill might come from a typo. It might also belong to a large household during a heat wave. The calculation tells you where to look; it doesn’t tell you what you’ll find.

This is why “outlier” should be treated as a useful label, not a verdict. Removing an unusual value just because it looks different can erase important information. First check the source and consider what the observation means.

Reading quartiles in a box plot

A box plot puts quartiles into a visual format. The box stretches from Q1 to Q3, and a line inside it marks the median. The length of the box represents the IQR. Lines, often called whiskers, extend out from the box to show more of the data; depending on the plotting convention, they may reach the most distant values within the outlier cutoffs. Possible outliers are then shown as separate points.

At a glance, a box plot can show whether one group has a higher median than another, whether its values are more spread out, or whether its distribution seems uneven. A median line near one edge of the box, for instance, can hint that values are more spread out on one side of the center. Still, a plot is a summary, not a full account of the data. Pair it with sample size and context when those details matter.

A few cautions worth keeping in mind

Quartiles depend on the order of values, not on the gaps between them. A dataset with uneven jumps can still be split into four ranked sections. That’s useful, but it means quartiles don’t describe every detail of the distribution.

There’s also no single calculation convention used in every software package or textbook. Some methods include the median when dividing the data into lower and upper halves; others leave it out. With a large dataset, the differences are often minor. With a small one, Q1 and Q3 can vary enough to notice. When comparing results, check that the same method was used.

Finally, quartiles work best alongside other measures. The minimum and maximum show the full range. The mean summarizes the arithmetic center. The median marks the halfway point. Together, these figures offer a more rounded view than any one number can provide.

Quartiles are simple, but their value is real: they divide ordered data into four roughly equal parts and show where the center and middle spread lie. Used thoughtfully, they help turn a crowded table into a clearer picture—one that shows not just what’s typical, but how much the data varies around it.