The Box-and-Whisker Plot, or Box Plot, is another effective visualization choice for illustrating distributions. along side histograms and stacked area charts, Box-and-Whisker plots are among my favorite chart types used for this purpose. They work particularly well once you want to match the distributions across two different dimension members side-by-side, where one set of dimension members structure the X-axis, and therefore the other dimension member is employed because the visualization’s level of detail.
As you’ll see, each set of circles corresponds to the dimension members on the X-axis for the Sub-Category dimension. the extent of detail, or most granular level of the analysis, is Month of Order Date. Since the extent of detail is month of order date, each Sub-Category column has 12 circles, one for every month of the year.

In short, this visualization is showing how the distribution of monthly sales vary between product sub-categories. While I can easily find several insights during this visualization and believe box-and-whisker plots to be among the foremost effective ways to speak distributions, I find them to be one among the foremost misunderstood chart types once I plan to share them with an external audience.

For this reason, this post shares not only the way to make Box-and-Whisker Plots in Tableau, but the way to read them.

How to Make a Box-and-Whisker Plot in Tableau
Box-and-Whisker Plot is one among the out-of-the-box Show Me options in Tableau, but they’re actually created with reference lines – which is what we’ll show here. to make a box-and-whisker plot, start by creating a bar graph with the dimension and measure of interest. within the example above, we are watching Sales by Sub-Category.
Next, add the distribution that you simply care close to the Detail Marks Card. during this case, we are watching how Sales are distributed by Sub-Category, by Month of Order Date. So during this example, Month of Order Date is added to the Detail Marks Card.
By increasing the extent of detail, a stacked bar graph is made , with each stack per sub-category representing one among the twelve months of the year. to line the inspiration of the box-and-whisker plot, convert this stacked bar graph to a dot plot by changing the mark type from Automatic (Bar), to Circle.

Lastly, to make a box-and-whisker plot, right-click on the Y-Axis, and choose “Add Reference Line”. When the add reference line panel appears, click on the selection for Box Plot. There are some formatting options available, but the default settings are usually best:
IQR stands for Interquartile Range, which are the info points between the primary and third quartile. therefore the default options are telling Tableau to form all of the info points on the box-and-whisker plot fit into 1.5 times the IQR; anything outside of that range is an outlier. This sounds confusing and is perhaps why I don’t see this chart type getting much traction, but this chart type provides tons of helpful context and is straightforward to read once someone explains it to you.
Each line on the box-and-whisker plot provides a bit of statistical context. the foremost important line is that the one right within the middle of every “box”, which represents median. With median displayed, you’ll quickly look out on the dimension members and compare medians, no matter how big or small the range of values is within each column.

That alone is extremely useful for an analysis, but the remainder of the lines even have a meaning. understanding from the median, subsequent set of lines is showing plus or minus one quartile from the median.

Lastly, the upper whisker is 50% above the IQR, or “middle fifty”, which are the info points within the primary and third quartile. The lower whisker is 50% less than the IQR. Any data points outside of the box-and-whisker are considered outliers.

Now not only are you able to make a box-and-whisker plot in Tableau, you recognize the way to use them to urge the foremost out of your analyses!