The objective of data visualization is to present data so that it's easier for people to consume, spot trends, and understand the story of the data. There is a debate over the use of three dimensional in charts, and people looking at Tableau may wonder how they can make visualizations that are three dimensional. Three dimensional is not available in Tableau. Three dimensional requires the viewer to spend more time trying to understand the data being presented to them than necessary.
People often consider three dimensional as a way to bring more information into the view. We don't need to use three dimensional to make visualizations beautiful and informative. Beauty can come in the form of simplicity and understanding the data story as easily as possible. Flashy isn't always better, particularly if it misleads the viewer.
If you can't use three dimensional but still want to show different metrics on the visualization, then what are your options? If you want to display a number of variables, then we can use a scatterplot matrix, also known as a splom. This is a grid of scatter charts. What is a scatter chart? It is also known as a scatter graph, scatter chart, or simply a scattergram, and it is simply a dataset plotted as points on a graph. The x axis represents one variable, and the y axis represents another. We can arrange them in a grid so that the viewer can easily compare along the variable, and also vertically between the graphs. We can also enhance the scattergram by adding in trend lines and using color to convey a story.
In this recipe, we will look at creating a scattergram matrix using Tableau and adding in a few reference lines. This is a good option when you are asked to create something that shows the relationships between variables. In this recipe, we will create a small dashboard segment that looks at whether the sales differ on a quarterly basis.
- Create a new worksheet by clicking on File and then New Worksheet.
- Drag Year(FullYearAlternateDate) onto the Columns shelf.
- On the pill Year(FullYearAlternateDate), click on the small plus sign on the left-hand side so that you now see the Quarter(FullYearAlternateDate) pill.
- On the Measures pane at the left-hand side, rename Number of Records to
Number of Sales Transactions. - Drag Number of Sales Transactions onto the Rows shelf.
- Change the measure calculation to Count by right-clicking on Number of Sales Transactions, navigating to Measure (SUM), and selecting the Count option.
- Let's create a new calculation called Diff from All Year's Average. Firstly, let's right-click on the Measures part of the Data sidebar and select the option Create Calculated Field.
- In the textbox, type the following formula and then click on OK to return to the main Tableau workbook canvas:
ZN(SUM([SalesAmount]))-Window_AVG(SUM([SalesAmount]))
- Let's change the visualization to a scattergram by going to the Marks shelf and selecting Shape from the drop-down list. The chart will now change to a scattergram. Here is an example in the following screenshot:
- Take the calculation Diff from All Year's Average and drag it onto the Color button. The color scheme will change to red and green.
- Click on the Color button and select Edit Color….
- Change the red color to royal blue and click on OK.
- Change the green color to gray and click on OK.
- Next, let's add in a reference line to show the trend lines. To do this, go to the Analysis menu item and choose the option Show Trend Lines.
- The trend lines will appear for each year. We will swap it around to show a quarter by dragging Quarter(FullDateAlternateYear) onto the Columns shelf to the left-hand side of Year(FullDateAlternateYear).
- Click on one of the reference lines and select the Format option. The Data pane will now change to the Format Lines option pane.
- For the Trend Lines option, click on the drop-down list and select the thinnest line. You can see an example in the following screenshot:
- Let's soften the color of the trend line by selecting the light purple color at the bottom, which is located on the bottom row at the far-right side. You will see the changes made on the trend line.
- Next, let's remove the headers in order to make the visualization as compact as possible. Remove the header FullDateAlternateKey by right-clicking on it and selecting Hide Field Names for Columns.
- On the y axis, right-click and deselect the option Show Header.
- On the x axis, right-click and deselect the option Show Header.
- The completed visualization now looks like the following screenshot:
