cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar
0 Kudos

Gardner-Altman and Cullen-Frey plots

Dear JMP,

 

being able to have these graphs available would be very useful for statistical inference purposes and would give more visibility than R which instead has dedicated packages to make them.
I would add box plots and a half violin plot on the dot plot to the Gardener-Altman plot (as shown in the photo), in this way the information that would be obtained would be optimal.

 

Thank you

 

Giuseppe

Box Blot and Half-Violin plotBox Blot and Half-Violin plot

Gardner-Altman PlotGardner-Altman Plot

Cullen-Frey PlotCullen-Frey Plot

3 Comments
SamGardner
Staff
Status changed to: Needs Info

Can you provide more informatin on the use case and why you want to have these types of plots available?  From my quick searching on these plots, it appears:

  • Cullen-Frey plots (also known as Pearson Plot) is just a way to plot the estimated skewness and kurtosis of a data set on a plot that shows the theoretical values (as curves on the plot) for the skewness and kurtosis for known distributions (normally exponential family distributions)
  • Gardner-Altman plots are a simple way to plot the effect size when comparing two groups.  You get similar information in the Fit Y by X (Oneway) analysis:
    SamGardner_1-1653998010707.png

     

  • In JMP, you can already create a combined dot plot, box plot, and violin plot

SamGardner_0-1653997856257.png

 

GiuseppeC
Level II

Dear Sam,

 

I know these graphics proposed by JMP, but they are not as informative as the ones you can get with the R that I use very little.
Observing the distribution of the points is important to understand if there is the presence of multi-modality and in the Gardner-Altman plot observing the mean or the median with the relative confidence intervals (as shown in the paper by the two authors) would be very informative.
Cullen-Frey plot (or Pearson plot), the authors in their book write that they are not the inventors of this graph, it would not hurt to be able to implement it in JMP instead of having to do it with R.
In general, making these graphics available, in my opinion, would not be a bad idea.

 

Thank you

 

Giuseppe

SamGardner
Staff
Status changed to: New