How To Make A Box Plot | Visualizing Data’s Story

Understanding data is a core skill across many fields, and sometimes numbers alone don’t tell the full story. Box plots offer a wonderfully clear way to see the spread and central tendency of your data at a glance.

Think of it as a helpful snapshot. This guide will walk you through each step, making the process straightforward and accessible. You will gain a solid grasp of this valuable statistical tool.

Understanding the Core Concepts of a Box Plot

A box plot distills a dataset into five key numbers. These numbers provide a robust summary of the data’s distribution.

It shows you where the middle of your data lies and how spread out the rest of the values are. You can also spot unusual data points with ease.

The Five-Number Summary

The foundation of every box plot rests on these five specific values:

• Minimum Value: This is the smallest number in your entire dataset, excluding any identified outliers.
• First Quartile (Q1): This marks the 25th percentile of the data. It means 25% of your data falls below this value.
• Median (Q2): This is the middle value of your dataset, representing the 50th percentile. Half the data is above it, and half is below it.
• Third Quartile (Q3): This marks the 75th percentile. 75% of your data falls below this value.
• Maximum Value: This is the largest number in your entire dataset, again excluding any identified outliers.

These five points give us a complete picture of the data’s central location and variability.

Gathering Your Data and Ordering It

The first practical step is always to collect your raw data. Make sure all your observations are present and accurate.

Once you have your data, the next critical step is to arrange it in ascending order. This means going from the smallest value to the largest value.

Ordering your data makes all subsequent calculations much simpler and prevents errors. Let’s use an example dataset to illustrate the process:

Our example dataset: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20]

This dataset has 11 data points, already sorted for our convenience. We will use this set throughout our calculations.

Calculating the Five-Number Summary

With your data ordered, you can now pinpoint the five essential values. Each calculation builds upon the previous step.

1. Determine the Minimum and Maximum Values

These are the easiest to find from your ordered list.

• Minimum: The smallest number in the dataset. For our example [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20], the minimum is 1.
• Maximum: The largest number in the dataset. For our example, the maximum is 20.

These values establish the overall spread of your data.

2. Calculate the Median (Q2)

The median is the true center of your data. Its calculation depends on whether you have an odd or even number of data points.

1. Count your data points (n). Our example has n = 11.
2. If n is odd: The median is the value at the (n + 1) / 2 position.
   • For n = 11, the position is (11 + 1) / 2 = 6.
   • The 6th value in [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20] is 6. So, Q2 = 6.
3. If n is even: The median is the average of the two middle values. These are at positions n / 2 and (n / 2) + 1. You add them together and divide by two.

3. Calculate the First Quartile (Q1)

Q1 is the median of the lower half of your data. This means you find the middle of all values below the overall median (Q2).

1. Identify the lower half of the data. Do not include the median (Q2) if your total dataset (n) was odd. If n was even, the median is between two numbers, so both halves are distinct.
   • Our example data: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20]. The median (6) is a data point.
   • The lower half is [1, 2, 3, 4, 5].
2. Find the median of this lower half. This will be Q1.
   • The lower half has 5 data points. (5 + 1) / 2 = 3.
   • The 3rd value in [1, 2, 3, 4, 5] is 3. So, Q1 = 3.

4. Calculate the Third Quartile (Q3)

Q3 is the median of the upper half of your data. You find the middle of all values above the overall median (Q2).

1. Identify the upper half of the data. Again, do not include the median (Q2) if your total dataset (n) was odd.
   • Our example data: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20]. The median (6) is a data point.
   • The upper half is [7, 8, 9, 10, 20].
2. Find the median of this upper half. This will be Q3.
   • The upper half has 5 data points. (5 + 1) / 2 = 3.
   • The 3rd value in [7, 8, 9, 10, 20] is 9. So, Q3 = 9.

Interquartile Range (IQR)

The IQR is the range of the middle 50% of your data. It is a measure of statistical dispersion.

You calculate it by subtracting Q1 from Q3.

IQR = Q3 – Q1

For our example: IQR = 9 – 3 = 6.

The IQR is very helpful for identifying outliers, which we will discuss next.

Our Example’s Five-Number Summary

Summary Point	Value
Minimum	1
First Quartile (Q1)	3
Median (Q2)	6
Third Quartile (Q3)	9
Maximum	20
Interquartile Range (IQR)	6

Identifying Potential Outliers

Outliers are data points that fall significantly outside the general range of the rest of your data. They can skew your analysis if not handled thoughtfully.

Box plots use a specific rule to identify these points, based on the IQR.

The 1.5 IQR Rule

We define fences, or boundaries, beyond which data points are considered outliers.

• Lower Fence: Any value below Q1 - (1.5 IQR) is a potential outlier.
• Upper Fence: Any value above Q3 + (1.5 IQR) is a potential outlier.

Let’s apply this to our example dataset:

• Q1 = 3, Q3 = 9, IQR = 6.
• Lower Fence: 3 - (1.5 6) = 3 - 9 = -6.
• Upper Fence: 9 + (1.5 6) = 9 + 9 = 18.

Now, we check our data points against these fences:

Our dataset: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20]

• No data points are less than -6.
• The value 20 is greater than 18. Therefore, 20 is an outlier.

When drawing the box plot, outliers are typically marked individually. The “whiskers” extend only to the highest and lowest data points that are not outliers.

For our example, the new “maximum” for the whisker will be 10 (the largest non-outlier). The minimum remains 1.

How To Make A Box Plot: Drawing the Visual

Now that you have all the necessary numbers, it’s time to bring your box plot to life visually. This is where the story of your data truly unfolds.

1. Draw a Number Line: Start by drawing a horizontal or vertical number line. This line should cover the entire range of your data, from your minimum to your maximum (including potential outliers). Make sure it has appropriate, evenly spaced tick marks.
2. Mark the Five-Number Summary:
   • Place a small vertical line or dot at your Q1, Median (Q2), and Q3 values directly above your number line.
   • For our example, mark 3 (Q1), 6 (Q2), and 9 (Q3).
3. Draw the Box:
   • Connect the Q1 mark to the Q3 mark with horizontal lines to form a box. This box represents the middle 50% of your data (the IQR).
   • Draw a line inside the box at the Median (Q2) mark. This line shows the central tendency.
4. Draw the Whiskers:
   • From the edge of the box at Q1, draw a line (a “whisker”) down to the lowest data point that is not an outlier.
   • From the edge of the box at Q3, draw a line (another “whisker”) up to the highest data point that is not* an outlier.
   • For our example, the lower whisker extends from 3 down to 1. The upper whisker extends from 9 up to 10 (since 20 is an outlier).
5. Plot Outliers:
   • If you identified any outliers, mark them individually with a distinct symbol (like a star, circle, or ‘x’) beyond the whiskers.
   • For our example, place a distinct mark at 20 on your number line.

Box Plot Components and Their Meaning

Component	Represents
The Box	The middle 50% of the data (IQR)
Line inside Box	The Median (Q2)
Whiskers	Range of non-outlier data
Individual Marks	Outliers

Your completed box plot provides a clear visual summary. You can quickly see the spread, the center, and any unusual values in your dataset. This visual approach helps you understand your data’s characteristics quickly.

Interpreting Your Box Plot

Once your box plot is drawn, it offers immediate insights into your data’s distribution. The length of the box shows the data’s spread around the median.

A longer box means more variability in the middle 50% of your data. A shorter box indicates data points are more clustered.

The position of the median line within the box indicates the skewness of your data. If the median is closer to Q1, the upper 50% of the data might be more spread out.

Conversely, if the median is closer to Q3, the lower 50% of the data might show greater dispersion. The whiskers also offer clues about the overall range and any extreme values.

How To Make A Box Plot — FAQs

What is the primary purpose of a box plot?

The primary purpose of a box plot is to visually display the distribution of a dataset. It summarizes the data’s spread, central tendency, and potential outliers using the five-number summary. This visual representation allows for quick comparisons between different datasets or groups.

Can a box plot tell me if my data is symmetrical?

Yes, a box plot can offer clues about data symmetry. If the median line is roughly in the center of the box and the whiskers are of similar length, the data tends to be more symmetrical. A median shifted to one side or uneven whisker lengths suggest skewness in the data distribution.

What do the “whiskers” on a box plot represent?

The whiskers on a box plot extend from the edges of the box to the lowest and highest data points that are not considered outliers. They illustrate the range of the main body of the data. Any points beyond these whiskers are individually marked as outliers.

Why is identifying outliers important in a box plot?

Identifying outliers is important because these extreme values can significantly influence statistical calculations, such as the mean. A box plot clearly flags these points, prompting you to investigate their cause. They might be data entry errors or genuine, but unusual, observations.

Can I compare multiple datasets using box plots?

Yes, box plots are excellent for comparing distributions across multiple datasets. You can place several box plots side-by-side on the same scale. This arrangement makes it easy to compare their medians, interquartile ranges, and the presence of outliers at a glance.