39476_en-US

liliconcom

box plot basics

Hello,

 

My data table has several columns of the same data type (but different test runs). The reason for not having the same data always in the same column is that I would like to be able to create new columns (e.g. "change from run1 to run2" = run2 - run1) and also plot those in graph builder. The number of columns of similar data type is dynamic (because the user can select how many runs to compare).

 

I would like to have the option to create a box plot based on the data coming from several columns. Is this possible? If so, how?

If I add another variable to the y-axis, an additional box lot is created and the data is not merged.

 

Any suggestion for workaround in my workflow is also highly appreciated.

 

Thank you!

 

Cheers,

Alex

Re: box plot basicsThank you taking the time to look into this.

Per label I have only a single data item per subject, you seem to have multiple (but I guess this is not a problem).

What could be a problem is the following:
My assumption is that the box plots in the upper part of your plot only consider data from a single label and a single subject.

I would like to have a single box plot for a label that merges the data of all subjects.Re: box plot basics

Is the image below close to what you are trying to achieve?

 

Re: box plot basics

I do not see a way to do what you want within one Graph Builder instance.  However, if I had the requirement to do what you want, JMP, using a script, could produce the 2 graphs you want, and then take the parts from the graphs and bring then together into a final display.

 

maybe another user has a better insight on how to use  a single Graph Builder to do what you want?

Re: box plot basicsThank you for the feedback.

I think I have done a poor job trying to explain what I would like to achieve.

I would like to build a graph that has the lines for the different columns and within the same graph builder instance another below that contains the box plot (derived out of the combined data of the columns).

With your approach (nice solution by the way, I was not familiar with the stack command), I can do either or but not both at the same time because either I have the data separated (-> line plots ok, no box plot) or combined (box plot ok, no line plots).

Any other idea?

Thank you for your help!Re: box plot basics

What you need to do is to simply create a new data table where the columns you want to combine are stacked into a new column.  To do this, use

     Tables==>Stack