Subscribe Bookmark
audrey_shull

Staff

Joined:

Jun 23, 2011

JMP add-in summarizes sample data (or your data)

Sometimes when you hear about a new feature in JMP, you want to try it out. You may not have the right kind of data on hand, or you may just want to see a canned example. The JMP sample data is a rich resource full of these good examples. Some examples are showcased in the documentation, but many are not. Over the summer, one of our summer interns created a JMP application that makes searching for these demonstrative examples much easier. Evan Zayas has already returned to his senior year at MIT where he is majoring in physics, so I have the privilege of showcasing this gem.

The Data Table Summary Application has been designed with JMP Sample Data in mind, but the add-in is also customizable if you have your own directory full of JMP tables that you’d like to search. Just Browse to your own folder and press the button to Generate Tables & Reports. The JMP Sample Data directory is analyzed on my computer in about 30 seconds. It works by scanning the tables in the specified directory to find table scripts that use each platform.

Once the application analyzes all the tables, you are presented with a Data Table Summary dialog. The left tab for Histograms will show a summary of basic summary statistics about the data, such as number of rows and columns. The most versatile functionality, however, is found in the Platforms Used tab, shown below. There are two ways to browse through the JMP Platforms. On the left side, you’ll find an alphabetical drop-down list of all the JMP platforms. If you know which platform you’re looking for information about, that is the quickest way to find the relevant tables. A more exploratory approach can be taken on the right side by browsing a collection of outline nodes. The nodes are laid out identically to the JMP menu items. The tables revealed for each platform will be identical regardless of the path you took to get there, and can be clicked to open the table immediately for exploration.

A neat feature for this add-in is its customizability – you can add your own JSL keywords to the tool to correspond to your own needs and interests. An example is shown below, where I want to quickly find example tables using the Bootstrap feature. The documentation is an excellent place to go for a detailed example, but in this case, I am looking for a canned example. Choose the Edit button to get to the Edit Keywords dialog shown below. Enter your item of interest and press Add. When you press OK, the directory of tables will be scanned again for your JSL keyword(s) and a new category will be added to the results.

In the image below, you can see that the Bootstrap keyword was added both to the left and right lists. There is one canned example of Bootstrap in the sample data, in the US Store.jmp table. Another use might be getting a list of all tables in your own folders of stored tables that use certain custom functions.

The Data Table Summary add-in can be found on the JMP File Exchange (download requires a free SAS profile), along with several other new add-ins created by our summer interns, such as functionality for Gantt Charts, Combining Map Shapes, Statistical Dot Plots, Geocoding and a KML Polygon Importer for maps. It was a productive and exciting summer for new JMP innovations in the JMP Development group. I wish the best of luck to Evan and all the other interns who worked with JMP this summer.

1 Comment
Community Member

Lee Creighton wrote:

I think these add-ins are one of the most important additions to JMP since version 4 added scripting. I've got lots of repetitive tasks that I need to do that really aren't up to script level, but by making them a menu item or a pushbutton, I get the increased functionality with jumps great user interface.

For example, I commonly have to clean data. Sometimes the data is in all caps, sometimes all lowercase sometimes mixed case and what I want is camel case. I've written An add-on that lets me highlight a block of data, and thence that's a fine how I wanted to appear. So handy.