LS-Means Diffogram with PROC GLIMMIX

This example is taken from the section Graphics for LS-Mean Comparisons in Chapter 43: The GLIMMIX Procedure. The following statements create a SAS data set that contains measurements from an experiment that investigates how snapdragons grow in various soils:

data plants;
   input Type $ @;
   do Block = 1 to 3;
      input StemLength @;
Clarion   32.7 32.3 31.5
Clinton   32.1 29.7 29.1

   ... more lines ...   


The following statements run PROC GLIMMIX:

ods graphics on;

proc glimmix data=plants order=data plots=diffogram;
   class Block Type;
   model StemLength = Block Type;
   lsmeans Type;

The PLOTS=DIFFOGRAM option produces a diffogram, shown in Figure 21.10, that displays all of the pairwise least squares mean differences and indicates which are significant.

Figure 21.10: LS-Means Diffogram