User Community
cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
  • Don’t miss your chance to experience Discovery Summit Europe at our best available rate! Early bird registration through 31 Oct.
  • Image to Table JMP extension imports single or batch images as data tables. Useful to extract curves from pictures or convert image data to curve data to use in analyses like Functional Data Explorer. Download and install the extension.

Discussions

Solve problems, and share tips and tricks with other JMP users.
Choose Language Hide Translation Bar
Ted
Ted
Level IV

X-axis format

Jun 28, 2019 12:53 PM (7755 views)

How to divide the X-axis into equal parts (SEE picture)?
P.S. X-axis is not interactive in the "Y by X platform"

Preview file
16 KB
0 Kudos
1 ACCEPTED SOLUTION

Accepted Solutions
jules
Community Manager jules
Community Manager Community Manager
Solution

Re: X-axis format

Jun 28, 2019 01:37 PM (7744 views)  |  Posted in reply to message from Ted 06-28-2019

Hi @Ted,

Mosaic plots (which is the default visual you will get in Fit Y by X when you use a categorical X and categorical Y) do not divide the x-axis evenly unless there is an equal number of observations in each of the categories represented by the x-axis. You can see the documentation for mosaic plots here. There isn't a way to disable this scaling in Fit Y by X because it's a feature of the plot itself. And, it's a good thing! Because of that scaling, you can see the marginal distribution of your X-variable, which is quite useful to see in addition to the conditional distribution of Y for each X category. 

 

If you absolutely need a plot that sizes the X-axis uniformly irrespective of the distribution of X, it won't be a mosaic plot. But, you can make such a plot in Graph > Graph Builder, by placing your Y variable in the Y Role, and your X variable not in the X-Role, but in the Group X role at the top. Then, toggle the mosaic plot at the top to make a mosaic-like plot.

 

gb.gif

 

 

I hope this helps,

@jules 

View solution in original post

3 REPLIES 3
Mark_Bailey
Staff Mark_Bailey
Staff

Re: X-axis format

Created: Jun 28, 2019 01:37 PM  |  Last Modified: Jun 28, 2019 10:40 AM (7745 views)  |  Posted in reply to message from Ted 06-28-2019

Sorry - I was mistaken that you could turn off the proportional spacing in the Contingency platform. (I never wanted to turn it off.) I must be thinking of Oneway, another platform under Fit Y by X.

0 Kudos
jules
Community Manager jules
Community Manager Community Manager
Solution

Re: X-axis format

Jun 28, 2019 01:37 PM (7744 views)  |  Posted in reply to message from Ted 06-28-2019

Hi @Ted,

Mosaic plots (which is the default visual you will get in Fit Y by X when you use a categorical X and categorical Y) do not divide the x-axis evenly unless there is an equal number of observations in each of the categories represented by the x-axis. You can see the documentation for mosaic plots here. There isn't a way to disable this scaling in Fit Y by X because it's a feature of the plot itself. And, it's a good thing! Because of that scaling, you can see the marginal distribution of your X-variable, which is quite useful to see in addition to the conditional distribution of Y for each X category. 

 

If you absolutely need a plot that sizes the X-axis uniformly irrespective of the distribution of X, it won't be a mosaic plot. But, you can make such a plot in Graph > Graph Builder, by placing your Y variable in the Y Role, and your X variable not in the X-Role, but in the Group X role at the top. Then, toggle the mosaic plot at the top to make a mosaic-like plot.

 

gb.gif

 

 

I hope this helps,

@jules 

Evan_Morris
Evan_Morris
Level IV

Re: X-axis format

May 19, 2022 05:10 AM (4109 views)  |  Posted in reply to message from jules 06-28-2019

Not sure if this is still an issue in the newer versions of JMP, but the problem I have is the following:

 

For the X by X contingency I can't adjust the y-axis which looks lousy for low frequency issues, so I go to the graph builder to use the mosaic.  But there the X-axis can't be adjusted.  Doing the Group X binning works ok but then I lose the built in chi-sq (I can solve it in x by x and then bring it back over but that's a little clunky).  I don't really care if I do it graph builder or x by x but I need it somewhere where I can get a clean, easy to read graph to go with my chi-sq.  

 

Maybe just make the y-axis scalable in the normal contingency plot?

0 Kudos

Recommended Articles