I'm a newbie to JMP but have used SAS for years, with much emphasis on data preparation (e.g., PROC FORMAT). What's the best way to take a data table in JMP and group values for analysis? (e.g., age ranging from 0 to 99, want grouped into 1-5, 6-10, etc.)... or is the assumption that your data is already "prepped" once it hits JMP? I've looked at value labels and row states and nothing's jumping out at me.
I'm sure you could use a formula for a column that takes the actual age and puts it into a "bin" 1-5, 6-10, etc. It would be tedious, but I am not aware of any non-tedious way to do this (even in SAS) if you really want the bins to be formatted as "1-5", "6-10", etc. If you just want bins without clear labels, things get much easier, create a formula that divides the age by 5 and then takes the floor of that division.
Of course, part of me is screaming ... don't turn continuous variables into bins, that's rarely a good thing to do for statistical analyses.
If you want to bin a continuous variable there are at least a couple of ways to do this.
One is to use the Distribution platform on the continuous column and choose Save Level Numbers or Save Level Midpoints.
These options will use the histogram bins to create new columns in your data table.
If you need adjust the histogram bins you can:
a) use the hand tool (press H on your keyboard) and drag in the histogram
b) use the Histogram OptionsSet Bin Width
If you want to specify your own cut points to bin your column you can use this add-in.
I hope that helps.
Thank you! What are the steps to deploy the app? I attempted to use it in JMP 8.0 and was unsuccessful. We are soon to upgrade our SAS release (currently we're 9.2) and get JMP 10.0 installed, so that may have a bearing on the deployment steps. Also my machine will be upgraded from Windows XP to Windows 7 if that makes a difference. My use of JMP at this time is entirely experiemental in preparation for learning JMP 10 once all these upgrades have taken place.