How exactly does the analysis for scatterplot stories work? This article will explain the different analytic types that power each sentence in your scatterplot narrative.
Scatterplot story types are used to better understand the relationship between two measures. This story type requires 2-3 measures. The narrative analysis adds the relationship (regression) between two measures, and if there are any groups (clusters) within the data.
Say we have a scatterplot that measures profit and sales across some dimension.
The first two bullets might read:
- As Profit increased, Sales increased based on the data provided. Specifically, when Profit increased by one, Sales increased $2.89. There may be other factors contributing to Sales, but there is evidence of a very strong relationship.
- Few entities deviated from this general relationship, indicating a good fit.
These bullets are powered by regression analytics. Regression shows how one measure affects another. Notice in the first bullet, the story has identified a relationship between profit and sales.
The next bullet could read:
- When organized into groups of similar Profit and Sales values, one distinct group stands out. There were 1,459 entities that had values of Profit between one and 1,996 and Sales between six and 4,902.
This bullet is derived from clustering. Clustering analytics tries to identify key groups or clusters across all the variables in the data.
Scatterplot stories will also call out any existing outliers, like such:
- Canon Copier, Fellowes PB500 with Manual Bind and Hewlett Packard LaserJet 3310 Copier, among others were exceptions with very high Profit and Sales values.
Remaining bullets for scatterplot stories use range & average analysis.
- The minimum value for Profit is one (Acco Expandable Hanging Binders, Avery Self-Adhesive Photo Pockets for Polaroid Photos and HON 5400 Series Task Chairs for Big and Tall, among others) and the maximum value is $25,200 (Canon imageCLASS 2200 Advanced Copier), a difference of $25,199. The average Profit per entity is $235.53 and the median is $61.
- The minimum value for Sales is six (Avery 5 and Xerox 20) and the maximum value is 61,600 (Canon imageCLASS 2200 Advanced Copier), a difference of 61,594. The average Sales per entity is 1,129 and the median is 269.