How To Make A Modified Box Plot Fences And Outliers?

A modified box plot is a tool for spotting unusual data points. It uses a special rule to find the “fences” that separate normal values from outliers. To make one, you calculate the quartiles, find the interquartile range (IQR), multiply it by 1.5, and add or subtract that from the quartiles to set the fences. Any data point falling outside these fences is marked as an outlier. This method is standard in statistics and gives you a clear visual of where your data clusters and which points stand out.

What Are Fences in a Box Plot?

Fences are the boundaries that determine which data points count as outliers. In a standard box plot, the “whiskers” extend to the most extreme data points that are still inside the fences. Points beyond the fences are plotted individually as dots or asterisks.

The lower fence is calculated as the first quartile (Q1) minus 1.5 times the interquartile range. The upper fence is the third quartile (Q3) plus 1.5 times the interquartile range. The interquartile range is simply Q3 minus Q1, which represents the middle 50 percent of your data.

This 1.5 multiplier is a convention, not a law of nature. It was popularized by statistician John Tukey in the 1970s. It works well for most datasets because it catches points that are genuinely far from the bulk of the data while not flagging too many normal values.

How To Calculate the Fences Step by Step

Start by arranging your data from smallest to largest. Find the median, which is the middle value. Then find Q1, the median of the lower half of the data, and Q3, the median of the upper half.

Subtract Q1 from Q3 to get the IQR. Multiply the IQR by 1.5. Subtract that result from Q1 to get the lower fence. Add it to Q3 to get the upper fence.

Here is a simple example. Suppose your data is 2, 4, 6, 8, 10, 12, 14. The median is 8. Q1 is 4. Q3 is 12. The IQR is 8. Multiply 8 by 1.5 to get 12. The lower fence is 4 minus 12, which is negative 8. The upper fence is 12 plus 12, which is 24. None of your values fall outside these fences, so there are no outliers.

Now change one value. If the data is 2, 4, 6, 8, 10, 12, 40, the median is still 8. Q1 is still 4. Q3 is now 12. The fences are the same. The value 40 is above the upper fence of 24, so it is an outlier.

What Makes a Box Plot “Modified”?

A standard box plot draws whiskers to the minimum and maximum values. A modified box plot draws whiskers only to the most extreme values that are still inside the fences. Outliers are shown separately as individual points.

This distinction matters because the standard version can hide outliers. If your data has an extreme value, the whisker stretches to include it, and the box plot looks like the data is more spread out than it really is. The modified version separates the outliers visually, so you can see both the central clustering and the unusual points clearly.

Most statistical software packages now produce modified box plots by default. When you see a box plot with dots beyond the whiskers, you are looking at a modified version.

How To Identify Outliers Using the Fences

Once you have your fences, identifying outliers is straightforward. Any data point below the lower fence or above the upper fence is an outlier. You do not need to make any judgment calls. The rule is mechanical.

Some datasets have extreme outliers that even the 1.5 IQR rule does not catch well. In those cases, some analysts use a 3 IQR multiplier to identify “far out” points. These are sometimes called extreme outliers. The 3 IQR rule is more conservative and flags only the most extreme values.

Outliers are not automatically errors. They can be genuine measurements that reflect real variation. Before removing an outlier from your analysis, check whether it is a data entry mistake, a measurement error, or a legitimate value that just happens to be unusual.

How To Draw a Modified Box Plot by Hand

Draw a number line that covers your entire data range. Mark Q1, the median, and Q3. Draw a box from Q1 to Q3 with a line at the median. This box represents the middle 50 percent of your data.

Determine the smallest data point that is still above the lower fence. Draw a whisker from the box down to this point. Do the same for the upper side. Draw a whisker from the box up to the largest data point that is still below the upper fence.

Plot every outlier as a small dot or asterisk beyond the whiskers. Each outlier gets its own point. This gives you a clear visual of both the central distribution and the unusual values.

Software like R, Python, Excel, and SPSS can all create modified box plots automatically. In most cases, you do not need to draw them by hand. But understanding the calculation helps you interpret what the software shows you.

Common Mistakes in Calculating Fences

The most common error is using the wrong quartiles. Different software packages calculate quartiles slightly differently. Some use the median of the lower half including the median itself. Others exclude it. For most datasets, the difference is small, but it can change whether a borderline point is flagged as an outlier.

Another mistake is confusing the whiskers with the fences. The whiskers extend to the most extreme data points within the fences. They are not drawn at the fence values themselves unless a data point happens to land exactly there.

A third issue is applying the 1.5 IQR rule to data that is not roughly symmetric. For heavily skewed data, the fences can be misleading. Some statisticians recommend using transformations or alternative methods for skewed distributions.

When the 1.5 IQR Rule Falls Short

The 1.5 IQR rule works well for roughly symmetric distributions. It can be too aggressive for small samples, flagging normal values as outliers. It can also miss true outliers in highly skewed data.

For example, income data is heavily right-skewed. The upper fence may be so high that only extreme values get flagged, while the lower fence may fall below zero, which is meaningless for income data. In these cases, you may want to use a different outlier detection method, such as the median absolute deviation or a statistical test tailored to your data distribution.

The rule also assumes your data has no clusters. If your data has two distinct groups, the box plot will treat the gap between them as a wide spread, and points in one group may appear as outliers relative to the other group. This is not a flaw in the method. It is a signal that your data may contain subgroups worth investigating separately.

Why Outlier Detection Matters

Outliers can distort your summary statistics. A single extreme value can pull the mean away from the center of your data. It can inflate your standard deviation and weaken statistical tests. Identifying outliers helps you decide whether to investigate, transform, or exclude them.

In quality control, outliers can signal defective products. In medical research, they can indicate measurement errors or genuinely unusual patient responses. In finance, they can reveal fraud or extreme market events. The modified box plot gives you a quick, standardized way to spot these cases.

Remember that identifying an outlier is not the same as explaining it. The box plot tells you a point is unusual. It does not tell you why. That requires domain knowledge and further investigation.