Fit Group and By

vince_faller

Is Fit Group supposed to actually force by graphs inside it to respect by columns?  It seems like the graphs are just being duplicated with a header that says Fit Group age=12, sex=M.  

 

names default to here(1);
dt = open("$SAMPLE_DATA\Big Class.jmp");
fg = dt << Fit Group(
	dt << OneWay( Y( :weight, :height ), X( :name ) ), 
	<<{Arrange in Rows(2)},
	By(:age, :Sex)
); //I like the way this looks but it doesn't seem to actually respect the By in the graphs

//this one respects the by but puts the groupings in each graph which is pretty ugly
//and doesn't arrange in rows
g1 = dt << Oneway( Y( :weight, :height ), X( :name ), By(:age, :Sex) )

jules

Hi @vince_faller,

You can include the By statement in both the call to Oneway() and Fit Group() and JMP will arrange the output accordingly. Is this what you were looking for? 

 

names default to here(1);
dt = open("$SAMPLE_DATA\Big Class.jmp");
g2 = dt << Fit Group( Oneway( Y( :weight, :height ), X( :name ), By(:age, :Sex) ), << Arrange in Rows(2), By(:age, :Sex));

You mentioned that you may have up to 10 By variables, which means the titles on the "Oneway Analysis" outline boxes will get pretty unruly. You could remove them altogether (which will also hide the Red Triangle menu) with the following code: 

 ( g2 << XPath("//OutlineBox[@helpKey='Oneway Report']") ) << {Set Title( "" )};

 

 

 

Or, you could rename them all with just a space to retain the red triangle menu but have a cleaner view: 

( g2 << XPath("//OutlineBox[@helpKey='Oneway Report']") ) << {Set Title( " " )};

 

 

 

I hope this helps!

@jules