I have a data table that has a datetime column modelled as numeric-continuous. I have various Serial Numbers ("SN", Categorical-Nominal) in another column, along with another Categorical-Nominal column, we can call "Y". I want to see when certain "Y" events occur, so I put the Categorical Nominal Variable on the Y-axis and datetime on the X-axis. For the purposes of this analysis, I choose to split each SN by Page. When I do this, all Pages show all levels of "Y" on the Y-axis. I only want to show the levels applicable to each SN on each Page. Is there a way to exclude levels on specific Pages that don't exist for that group?
If you run Graph Builder and use the Dialog to enter the specifications for the graph
and then use the By selection instead of the Page
It will treat each level of the By grouping as an independent graph. In my example, I deleted all rows for 12 year old females from the data table
A similar issue which I face very often:
too many labels on a x axis when using X group - or too many labels on a y axis when using Y group.
@shampton82 pointed out that "page" could be THE solution - with it's ability to have non-linked axes settings: direct Link
Unfortunately, the setting is ignored for categorical values : (
related entry in the wish list: