Automatic add connecting letters in Graph BuilderWhat inspired this wish list request?
What is the improvement you would like to see?
An option included in graph builder, like in Summary Statistics for example, to have it generated automatically, in a single click, (taking into account all the By variables, data filters etc). Then add the letters on the graph, on top of the bars. See image attached for an example of what it may look like.
Why is this idea important?
I believe it'sa must have, since other softwares can do this as standard.
Understanding whether means are significantly different from each other to make decisions is a very basic thing to do, but currently it is unnecesarily cumbersome and time consuming. I believ this would benefit very many users.
Basic Data Analysis and ModelingData Exploration and VisualizationRe: Automatic add connecting letters in Graph Builder - Status changed to: AcknowledgedRe: Automatic add connecting letters in Graph Builder - Status changed to: AcknowledgedRe: Automatic add connecting letters in Graph Builderyes actually, looks like the same thing ^^
Re: Automatic add connecting letters in Graph BuilderThis is maybe somewhat related Display letters of significance in graph builder when using boxplot and in fit y by x when comparison is performed
Most likely what you want could also be scripted, but would be nice to have it directly in JMP
Re: Automatic add connecting letters in Graph BuilderWhat inspired this wish list request?
I often analyze data in the form of bar charts with error bars. It can be difficult to know which ones are different from each other, so at some point one needs to do a Tukey, Student, or other test to compare the means. Currently the process is :
Fit Y by X, click red triangle--> compare means--> Tukey HSD (or whichever you're using), then find the connecting letters table (this process has to be done for every analysis if you have "By" variables, which is annoying). Export your graph in Excel, add a text box with the letters for each bar. (I don't know of a simple, non-convoluted way to do this on JMP).
This can take quite a bit of time to do, especially if you have to do this on several graphs.
