Box plot

A box plot shows a group of numbers.

It uses a box and lines called whiskers. The box has a line in the middle. This line shows the middle number.

The whiskers show how the numbers spread out. Sometimes, a number is very far away. We call these outliers. They look like dots on the plot.

This tool helps us see many groups at once. It is a quick way to study data.

A box plot is a way to show numbers.

It helps us see how data spreads out. It also shows if data is bunched up on one side. This is called skewness.

Most box plots use five main numbers. First is the minimum, the lowest value. Next is the first quartile. This marks the bottom quarter of the data. Then comes the median. This is the middle number. The third quartile marks the top quarter. Finally, there is the maximum, the highest value.

A box plot has a box and whiskers. The box goes from the first quartile to the third. A line inside the box shows the median. Lines called whiskers reach out from the box. They show the range of most numbers. If a number is very far away, it is an outlier.

Outliers look like dots or stars.

People use different kinds of box plots. Some have notches. These help us compare medians. Others change width to show group size. John Tukey helped make these popular in 1970.

A box plot is a clever way to see a group of numbers at a glance.

Instead of looking at a long list of values, you can see how they spread out. It shows where most numbers live and if they are bunched up on one side. This bunching is called skewness.

Box plots are very helpful because they do not take sides. They do not assume your data follows a specific pattern. This makes them a great tool for a quick look at many different groups.

To build a box plot, you first find five special numbers. These are called a five-number summary. First, you find the minimum, which is the lowest value. Then you find the maximum, or the highest value. Next, you find the median, which is the exact middle number.

You also need the first quartile and the third quartile. The first quartile marks the 25th percentile, or the bottom quarter. The third quartile marks the 75th percentile, or the top quarter. The distance between these two quartiles is called the interquartile range, or IQR.

Most box plots have two main parts: a box and whiskers.

The box is drawn from the first quartile to the third quartile. A line is drawn inside the box to show the median. Whiskers are lines that reach out from the box. They show how far the rest of the data goes. Sometimes, a whisker ends at the minimum or maximum value. Other times, whiskers only go out a certain distance based on the IQR. If a number is way too far away, it is called an outlier.

These outliers are often drawn as tiny dots or stars.

People have been working on these charts for a long time. In 1952, Mary Eleanor Spear introduced a similar method called the range-bar method. She wrote about it in her book "Charting Statistics." Later, in 1970, John Tukey introduced the box-and-whisker plot. He helped make this style of drawing very popular. He even wrote a book called "Exploratory Data Analysis" in 1977. Since then, many new versions have been created to show even more detail.

There are many ways to change a box plot to show new things. Some plots have notches around the median line. These notches help you see if the medians of two groups are truly different.

Other plots change their width to show how big a group is. A wider box might mean there are more people in that group. You can also find plots that show more complex shapes, like violin plots. These variations allow scientists to compare many sets of data all at once. It is a very efficient way to organize a lot of information.

A box plot is a powerful tool used in descriptive statistics. It provides a visual way to show the locality, spread, and skewness of numerical data.

By using quartiles, these diagrams summarize how data points are distributed. This method allows researchers to see where most values cluster and where they spread out. Because box plots are non-parametric, they do not assume a specific underlying statistical distribution. This means they can describe a sample without making guesses about the population's shape.

To create a box plot, you must first calculate a five-number summary. This summary consists of the minimum, the first quartile, the median, the third quartile, and the maximum.

The minimum is the lowest value in the set, excluding any outliers. The maximum is the highest value, also excluding outliers. The median, or second quartile, is the middle value of the ordered data. The first quartile (Q1) is the median of the lower half of the data. The third quartile (Q3) is the median of the upper half. The distance between Q1 and Q3 is called the interquartile range, or IQR.

The visual structure of the plot typically includes a central box and extending whiskers.

The box is drawn from the first quartile to the third quartile. A horizontal line is placed inside the box to mark the median. Whiskers are lines that extend from the box to show variability. In one common method, whiskers extend to the minimum and maximum values. However, many statisticians use a different rule based on the 1.5 IQR value. In this method, the upper whisker reaches the largest data point within 1.5 times the IQR above Q3. The lower whisker reaches the smallest data point within 1.5 times the IQR below Q1.

Data points that fall outside these whisker boundaries are called outliers.

These are values that differ significantly from the rest of the dataset. Outliers are often represented as individual dots, small circles, or stars on the plot. Using the 1.5 IQR rule helps identify these unusual points clearly. For example, if a temperature reading is much higher than the rest, it may appear as a separate point. This distinction helps scientists identify errors or rare, extreme events in their observations.

The history of these diagrams involves several important contributors. Mary Eleanor Spear first introduced the range-bar method in 1952. She detailed this in her book, "Charting Statistics," and later in "Practical Charting Techniques" in 1969. In 1970, John Tukey introduced the box-and-whisker plot.

Tukey later expanded on these ideas in his 1977 book, "Exploratory Data Analysis." His work helped make these plots a standard tool for exploring data patterns.

Since Tukey's time, several useful variations of the box plot have emerged. Variable-width box plots change the width of the box to show the size of a group. Often, the width is proportional to the square root of the group size. Notched box plots include a narrowing around the median line.

These notches serve as a guide for comparing medians between two groups. If the notches do not overlap, it suggests a statistically significant difference between the medians. Other versions, like adjusted box plots, use the medcouple statistic to better describe skewed distributions.

Box plots offer several advantages over other methods like histograms. They take up much less space on a page. This makes them excellent for comparing many different groups or datasets in parallel. They also allow for a quick graphical examination of a dataset's spread. While histograms can be influenced by the width of the bins used, box plots provide a standardized summary. This efficiency makes them a staple in modern statistical analysis and data science.