This week’s #WorkoutWednesday challenge came from @AnnUJackson . The goal was to analyse multivariate data using Parallel Coordinates.
Honestly, I did not build this kind of chart before, so I was so excited.
What is the Parallel Coordinates chart?
It was developed by Alfred Inselberg (his book is on my wish list: Parallel Coordinates: Visual Multidimensional Geometry and Its Applications ) in the 1970s as a way to visualize many variables together and seeing the relationship between them. This chart most common subjects of academic papers in visualization. This causes you don’t find it often in business and consumer data visualizations.
I love Stephen Few’s first thought about the chart (http://www.b-eye-network.com/view/3355)
„The first time that I saw a parallel coordinates visualization, I almost laughed out loud. My initial impression was “How absurd!” I couldn’t imagine how anyone could make sense of the dense clutter caused by hundreds of overlapping lines. This certainly isn’t a chart that you would present to the board of directors or place on your Web site for the general public. In fact, the strength of parallel coordinates isn’t in their ability to communicate some truth in the data to others, but rather in their ability to bring meaningful multivariate patterns and comparisons to light when used interactively for analysis.”
Let’s do it in Tableau:
Requirement: Create a parallel coordinates chart that shows Sales, Profit Ratio, and # Customers (CountD Customer Name) per Sub-Category.
The major challenge in building a parallel coordinates chart is getting the ranges for each variable into a common scale. First of all we have to normalize our variables (Sales, Profit Ratio and Countd Customer).
If you want to get a min-max scaling or min-max normalization just use this simple and general formula:
- For Sales
- For Profit Ratio
- For Countd Customer
These table functions normalize/map different value of the variable onto a scale of 0 to 1. The minimum value been mapped as 0 and the maximum been mapped as 1. Using this method we got the same scale for Sales, Profit Ratio and Countd Customer.
Tip: (You can use LOD expressions for rescaling the measures) for example for Sales:
Let’s visualize the chart
- Drag the Measure Values to the Rows , Measure Names to the Columns and filter for these 3 new Calculated fields.
- Add the Sub-Category to Detail Shelf. Compute the Table Calculations to this field.
- Select Line Marks Type.
Now we have the connected lines, but we don’t have the vertical line for each measures.
- Duplicate the Measure Values and create a dual axis.
- Use the Sub-Category field on the Line Shelf on the second Marks window.
The difficult part is now ready. YOU built your own Parallel Coordinates 🙂 Congrats!
If you want to get Ann’s final view, follow her instructions here : http://www.workout-wednesday.com/blog/ and take part in your first #WorkoutWednesday.