cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

JMP Wish List

We want to hear your ideas for improving JMP. Share them here.
Choose Language Hide Translation Bar
0 Kudos

Effect Summary Report PValue Formatting

What inspired this wish list request?

In the Model script, the Effect Tests Report has a helpful feature where the significance column (Prob > F) conditionally formats the values in eye-catching colors to make it easier to glean the information quickly, especially for a large number of effects. For example, values <0.05 are colored orange, and I assume that values <0.01 are colored red (correct me if I'm wrong for the latter threshold, though). However, the Effect Summary Report does not have this formatting for its PValue/FDR PValue column.

 

What is the improvement you would like to see?

It would be nice if the Effect Summary Report also conditionally formatted the PValue/FDR PValue columns just like the Effect Tests Report, especially since they are pretty much reporting the same PValues.

 

Why is this idea important?

Sometimes, when I am removing insignificant effects from the Effect Summary Report, I accidentally remove some that do have significance. I tend to make this mistake when I have a long list of effects with some bordering on significance, as after a while the numbers all start to blur together. That's why I appreciate the conditional formatting for the Effect Tests Report, where the orange and red coloring are much more readable and eye-catching. I feel that having this same formatting for the Effect Summary Report would make it a lot easier to identify insignificant effects quickly and avoid removing borderline significant ones by mistake. I recognize this may be a minor thing, but I think it would be a very helpful ease-of-use feature for those of us who have to regularly trim a lot of models for things like process characterization.

 

1 Comment
SarahGilyard
Staff
Status changed to: Not Planned For Now

Hello JMP-Rope. Thank you for posting. Part of the reasoning behind not formatting the p-values in the Effect Summary report is related to the nature of building Fit Least Squares models. It was intended not to adhere too strictly to the specific thresholds of .05 and .01 when doing model building by adding/removing effects in the Effect Summary dialogue. There are more considerations to adding/removing effects than strict adherence to the .05/.01 p-value paradigm. 

You can still apply color formatting to the p-value column in the Effect Summary if you would like. Right click on the p-value column and select Format Column. This option will allow you to format the p-value column. You can also make this a preset so that it can apply without setting it each time you use the platform. Please reach out with any further questions. Thank you!