Give a chestnut! Tableau Tips (84): Use the box and whisker chart to see your data distribution

About the box must

If you want to show the distribution of a set of data: for example: understand the data at a glance, see how the data skews to a certain segment, and look at the outliers in the data. A box-and-whisker plot is recommended, which is an important way to show the distribution of data.

If you want to show the distribution of a set of data: for example: understand the data at a glance, see how the data skews to a certain segment, and look at the outliers in the data. A box-and-whisker plot is recommended, which is an important way to show the distribution of data.

The box and whisker diagram, as can be seen from the name, this chart contains two parts:

- Box, containing the median of the data, and the 1st and 3rd quartiles (25% larger and smaller than the median);

- Required, generally representing data within 1.5 times the interquartile range (difference between the 1st and 3rd quartiles). “Should” can also be used to display the highest and lowest points in the data.

For example, the following figure: Compare housing sales prices. In a certain period of time, the median price of homes for sale was the highest in San Francisco, but the distribution in Los Angeles was even wider. In fact, the most expensive home sales price in Los Angeles is several times the median.

In many analysis tools, the process of making the box and whisker is very cumbersome, and often makes the data powder prohibitive. However, in Tableau's smart display, you can directly use the box and whisker map to make data presentation a breeze!

Here, let's share how to use the box and whisker diagram to see your data distribution in Tableau.

In this issue of "Climbing a Chestnut", the Tableau technique that Ada wants to share with you is: Use the box to map to see your data distribution.

For the convenience of learning, we use the supermarket data that comes with Tableau.

Specific steps are as follows:

Step1: Create a chart

Double-click on "Sales" and "Product Type" to generate a column chart by default.

Step2: Select box and whisker

Click on the top right corner of the mouse, or use the shortcut Ctrl+1 (Mac User: Command-1) to open "Smart Display" and select "Box and Stake".

Step3: Double click on "Order Date"

Next, double-click on "Order Date" to present the order data using the box and whisker chart.

Finally, we will help you better understand the box and whisker by interpreting this chart.

Through the above box and whisker diagram, the following data results can be roughly obtained:

It is not difficult to find that the highest sales in 2017 is the highest, and the lowest in 2016 is the lowest; the upper hub in 2017 is close to the maximum in 2014, but the median in 2017 is generally low. In 2016, the lower hub was the lowest and the median was the lowest.

More suggestions：

- Hide the points inside the box. This can help the viewer to focus on the outliers;

- Compare box plots between categories. To quickly compare the distribution between data sets, the box plot is ideal.

Today's Tableau trick, are you getting it?Open your Tableau and try it out!