In assessing whether a visual or statistical result is of interest, it would be useful to have a way to generate metadata column(s) where we should not expect to see a pattern. This action will allow a user to create random metadata groupings, either based on an existing column in their metadata or completely at random. That metadata column can then be provided to statistical or visual group analysis methods, and compared by the user to results from actual metadata columns.
In assessing whether a visual or statistical result is of interest, it would be useful to have a way to generate metadata column(s) where we should not expect to see a pattern. This action will allow a user to create random metadata groupings, either based on an existing column in their metadata or completely at random. That metadata column can then be provided to statistical or visual group analysis methods, and compared by the user to results from actual metadata columns.