Hello All,
Does anyone know a way to make combined data tables consisting of only expanded results? I implemented a parallelism test comparing different levels of a substance to a standard. JMP produces result tables of all possible comparisons and making a combined data table outputs all results of all comparisons without a flag to know which is which. I am only interested in the combined tables of only the the different levels against the standard. If I could save only the tabs opened into a combined data table without including the closed tabs, that would be amazing (See image below). I could save individual tables but I have done the analysis by several factors. Any help?
If you want only the report against the STD compound, then use the Make Into Data Table command instead of the one to combine reports.
If you want more than one report, then you might use the Rows > Data Filter on the current data table to select what you want or do not want. The rows can be deleted, excluded and hidden, or subset.
Thank you for your response @markbailey The Make Into Data Table command doesn't help here because like I said, I have several of that same tables (in the picture) so if anything, the combined data table is the closest; However, jmp make a combined table with all the other comparisons (versus L1, versus L2,......., plus versus the standard) for all the analyses similar to the picture i showed without a key to distinguish between which result comes from which "versus" tab. So my thought was, if I could make a combined datatable of only the similar results of what has been opened (leaving out the other comparisons collapsed), that would be the solution.
So here is the strange way I have been able to do what you want. The below script takes the display output from the Fit Y by X platform that was run, copies it to a journal, goes through all of the outline boxes and deletes the ones that are closed, and then runs the Make Combined Data Table and then finally, closes the journal. JMP is nice enough that when you do this, none of the journal creation ends up being displayed to the user, just the resulting new data table is displayed.
names default to here(1);
dt = open("$SAMPLE_DATA/semiconductor capability.jmp");
fg = Fit Group(
Oneway( Y( :NPN1 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
Oneway( Y( :PNP1 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
Oneway( Y( :PNP2 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
Oneway( Y( :NPN2 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
Oneway( Y( :PNP3 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
Oneway( Y( :IVP1 ), X( :SITE ), Means and Std Dev( 1 ), Mean Error Bars( 1 ), Std Dev Lines( 1 ) ),
<<{Arrange in Rows( 1 )}
);
fgr = fg << report;
nw = New Window( "Close Unwanted Tables",
Text Box( "Close the Outline Boxes of the Means and Std Deviation table" ),
Text Box( "you do not want to be in the Make Combined Data Table" ),
Text Box( "and then click OK" ),
Button Box( "OK",
nw << close window;
fgr << journal;
mjr = Current Journal();
test = (mjr << xpath( "//OutlineBox" ));
For( i = N Items( test ), i >= 1, i--,
If( test[i] << getOpen == 0,
test[i] << delete
)
);
mjr["Means and Std Deviations"][Table Box( 1 )] << Make Combined Data Table;
close all( journals, nosave );
)
);
The script from the Button Box does all of the work, so all you need to do is to take it an modify it to fit into your script.
My suggestion was more than just using the Make Combined Data Table command. The result includes the comparison you want and ones that you don't want. You could then use a Rows > Data Filter to select the rows you want for a Tables > Subset or the rows you don't want for Cols > Delete. Seems pretty simple to me.
