An important task in analyzing data with classification effects is to
estimate the typical response for each level of a given effect;
often, you also want to compare these estimates to determine
which levels are equivalent in terms of the response. You can perform
this task in two ways with the GLM procedure: with direct, arithmetic
group means; and with so-called least-squares means (LS-means).
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.