An Easy and Convenient Method for Constructing Contrasts
SESUG 2009: The Proceedings of the Southeast SAS Users Group
In many experiment design and observational study settings the researcher is not interested in the omnibus tests of hypotheses for "main effects" or "interaction effects." When confronted with this situation, SAS™software provides an easy and convenient way to test the "testable" hypotheses of interest via the CONTRAST and ESTIMATE statements. This paper will demonstrate an easy way to construct interpretable and testable hypotheses.
Vogel, Robert L..
"An Easy and Convenient Method for Constructing Contrasts."
SESUG 2009: The Proceedings of the Southeast SAS Users Group, 1 (5).