nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

onway analysis with t-test and pooled-t for muliple columns

steo45

InputTable = Open( "Path\A+B.jmp" );

New Window( "MyWindow", tt = V List Box() );

im using jmp 15 and im trying to get oneway analysis for multiple columns about 340 columns and each of the analysis must have t-test pooled-t test , means

so far i have been able to generate the one way graph for each with the means, but when i add t test i get a error

For( i = 3, i <= N Col( InputTable ), i++, 
// <Do Stuff>;
    tt << append(
        V List Box(
            Text Box( "Chart " || Char( i - 1 ) ), 
            InputTable << oneway( Y( Column( i ) ), X( Column( 2 ), t Test( 1 ) ) )
        )
    )
);

Subset of A+B.jmp

Accepted answers

txnelson

Why are you creating a script? You can easily do this interactively by either changing the preferences for the the Oneway platform, or by simply holding down the CNTL key when you request the t-test, and JMP will automatically display the analysis for all columns.

All comments

txnelson

steo45

im quite new to jump i dont know the tricks.

How do you change the preferences ? and thanks this seems to be interactive enough. but i have seen a lot of script written looping columns for same analysis. thought it was handy to do it in 1 click.

txnelson

File=>Preferences=>Platforms=>Oneway

As a new user.....concentrating on using the interactive methods will get you up and running much faster than trying to learn to script. The scripting really requires you to understand the interactive methods, because what the scripts are doing, are calling the interactive platforms to get the work done. Specifying 200 or 300 columns to run a Oneway analysis on them is a very few clicks. JMP is really geared to handle a large number of columns(thousands), and millions of rows.

Contrary to R, Matlab, SAS, etc. JMP requires no programming to be very productive with it. The scripting is a great add on, but if you first do what you want interactively, you can have JMP provide you with the script from the platform(s) you used.