Discussions
Solve problems, and share tips and tricks with other JMP users.
I need help with table scripting, I think...
Goal: run paired tests on 93 variables and save the combined data table of results to apply a multiple-test correction
What's working: the analysis
What I'm struggling with: the output of the combined data table. I'm looking for an output where, ideally, the two-tailed P value is included in the table with a descriptor column that relates back to the comparison. The combined data table looks like this:
I can't figure out a way to get a column that contains a title/name for that paired test for each 'block' of output. I can use table tools, etc. to get a list of the P values, but not in a way that contains a reference back to the paired test. Ideally, I would have something like: "Difference: Log10....." "Prob > |t|" then I can easily apply the test correction.
Hope I make sense to someone!
Thanks so much in advance!
Shiz
Yeah, but that results in no relationship between the P value and the test description (see below). The only notation of the variables in the merged table output (above) is in the first two rows in that block of data. I feel like I'm missing something silly and obvious. Thanks for the help!
I can get there by shifting/cut/pasting to something I can run with...essentially moving the P descriptor (col 3) and value (col 5) up two cells so it now has some indication of the variable used in the test (col 1). So a win for this, but would like to use this as a chance to work on the table scripting.
Ideally, need an additional column that blocks each test. For example, in my initial table...having a column with the first 8 rows indicating "Diff: Log[xxx] = Log[yyy]", and so forth. But there are other parts I don't understand about the table output. I was expecting 4 columns (based on the report) but the P values all shift a column over:
Thanks for your input!
