Scatter plot displays lots of data points at once yet still enable you to quickly identify clusters, correlations, trends, outliers.
You can use scatter plot to identify and explore relationships between 2 measures.
Firstly, we add 2 measures on opposite axis (row and column)
  • Tableau automatically sum up the 2 measures for the entire data set, resulting in 1 mark

  • To add more marks into the view, add dimensions to level of detail

  • To reveal the pattern in data better, we could also add dimension to color.

  • Things we can see from scatter plot chart:
    • Clusters: well-define groups
    • Correlation:
      • Positive: both measures show similar trends in high and low values
      • Negative: measures show opposite trend to each other, one measure shows high value while other measure shows low value and vice versa.
      • No relation: values show no relationship
    • Trend
    • Outlier: values that appear from outside of the trend

From the scatter plot above, we can see a positive correlation between average marketing cost and average sale. The higher the marketing cost, the higher the sale.

We could also spot out 2 clusters groups that appear from outside of the trend:

  • Coffee: low marketing cost and high sale
  • Espresso: high marketing cost but low sale

This could create points for investigation and lead to insights and stories from the data set.

In summary, scatter plot is useful for exploring the data and spotting trends and groups, to see if the two interested measures are related to one another.

To help with deciding if scatter plot is good for use, you could consider the following:
  • Do you want to see if a change in one measure would increase or decrease the other measure?
  • Do you want to see if there is a relationship between the two measures and which effect does it have?

It has brought it to the end of this blog, I hope you find it helpful.

Thanks for reading!


The Data School
Author: The Data School