Association Table visualization
The Association Table visualization let you associate metrics with metrics, dimensions, and dimension elements using Cramer’s V algorithm.
The Association Table compares values with Cramer’s V calculation rather than using Pearson’s correlation coefficient as used in the Correlation Matrix and Correlation Chord visualizations (these can only compare metrics, while the Association Table and Association Chord can compare metrics, dimensions, and elements).
Build an Association Table
The Association Table compares metrics over a countable or non-countable dimension. The table can be modified to highlight associations within the visualization through color picking or to render it as a text map, heat map, or both.
-
Open an Association Table.
Right-click Visualization > Predictive Analytics > Association Table.
-
Select an extended dimension—a Clickthrough, Hit, Product, Visit, or Visitor dimension. An Association Table will open with the extended dimension identified in the corner and its associated metric placed in both the row and column.
The Associations Table uses Cramer’s V as a symmetrical correlation, resulting in selected metrics, dimensions, and elements values reflected in both the columns and rows of an Association Table. For example, selecting the Product extended dimension uses the Visits metric as the associated metric in both the row and column of the table, resulting in a perfect yet useless comparison (1.00) because the compared values are identical.
-
Add more values to the Association Table.
Right-click in a column or row and select Add Metric or Add Dimension. You can also drag metrics and dimensions from a Finder panel. Dimension elements can also be dragged and dropped from an open table to the table visualization.
note note NOTE There is a limit of ten rows and columns allowed in the Association Table.