Discovery Summit Early User Edition: Solving data puzzles and increasing speed

[#1]

Discovery Summit Early User Edition has come to a close, but that doesn’t mean you can’t revisit the materials from the conference in the Community! On day one, Becca Feick, a data analyst at Cook Biotech, delivered a plenary talk about solving even your most complex data puzzles in JMP with basic methods and strategies. Day two featured JMP academic ambassador Ross Metusalem, who highlighted how new users can boost their speed in JMP with shortcuts and best practices for efficiency. Read on for Becca and Ross’ answers to questions they didn't have time to respond to during the Q&A.

Additionally, we welcomed paper presenters from JMP, Adsurgo, NXP Semiconductor, and GE Aerospace, their presentations spanning a variety of topics. You can also access this content on demand.  

Solving Data Puzzles with JMP
Becca Feick
Data Analyst, Cook Biotech
On demand

Does your script handle the e-mailing [of results] as well?

My script does not directly handle e-mailing the results. I need to join two PDFs before emailing the results and automated e-mail wasn’t something in scope when the script was created. JSL does have the Mail function, but I’ve never personally used it while scripting.

How did you make the 'control chart data entry' form?
The user interface is a window with several nested display boxes in it. I used combinations of multiple types of display boxes: H List Box, V List Box, Spacer Box, Text Box, Check Box, Number Edit Box, Line Up Box, and Button Box. It may not be the most elegant solution, but it gets the job done. In some of my other scripts, I’ve also incorporated Border Box, Text Edit Box, Combo Box, Column List Box, and Panel Box into user interface creation.

Speed Squared with JMP
Ross Metusalem
Academic Ambassador, JMP
On demand

Is there a way to handle the different files with different number of columns?
With File > Open Multiple (File > Import Multiple Files on Windows), raw data files with different numbers of columns will be imported as separate JMP data tables. In the presentation, we used the Stack Similar Files option to open all Excel files as a single JMP data table. JMP looked for identical column structures across the individual files and automatically concatenated all files with identical structures, which in our example was all of them. Now, imagine instead that you have a group of data files that collectively exhibit two distinct column structures; Stack Similar Files would yield two JMP data tables, one for each column structure. 
Note that you can always choose One Table for Each File to force JMP to import each raw data file as a separate JMP data table, even when identical column structures are present. Regardless, if you end up with multiple JMP tables that need to be combined in some way, you can use the tools under the Tables menu to concatenate or join them. See the JMP Help on multiple file import and data table reshaping for more information.

Is there a way to run multiple saved scripts with one click?
Here’s a quick way to run multiple data table scripts: use left click on the script names to highlight all that you’d like to run and then right click and select Run Script to run all highlighted scripts. However, if you routinely run the same group of table scripts, you may find it more efficient to create a workflow with Workflow Builder or to create a single table script that produces multiple outputs. If you’re interested in the latter, first create a new table script and then paste multiple graphing or analysis scripts into that one table script (and be sure to separate each with a semicolon). Save Script > To Clipboard would be useful to get each JMP-generated script onto your clipboard.

Best invited paper
Parallel Plots for Visual Configuration Management of Model Applicability
Lee Becker, GE Aerospace

Best contributed paper
Using JMP to Win Your Fantasy Football League
Nick Holmes, JMP

More paper presentations
Advanced Decisions Using Predictive Modeling in Semiconductor Industry
Su-Heng Lin, NXP Semiconductor Inc.

Pictures from the Gallery 9: Innovative Graph Builder Views
Scott Wise, JMP

Take Heart: Establishing Causality Without DOE
James Wisnowski, Adsurgo

More Unsupervised Engineering: Hacking the Short Run Control Chart
Mike Anderson, JMP