Chapter 3: Introduction to Analysis-of-Variance Procedures

Overview

This chapter reviews the SAS/STAT software procedures that are used for analysis of variance: GLM, ANOVA, CATMOD, MIXED, NESTED, NPAR1WAY, TRANSREG, TTEST, and VARCOMP. Also discussed are SAS/STAT and SAS/QC software procedures for constructing analysis of variance designs: PLAN, FACTEX, and OPTEX.

The flagship analysis-of-variance procedure is the GLM procedure, which handles most standard problems. The following are descriptions of PROC GLM and other procedures that are used for more specialized situations:

ANOVA	performs analysis of variance, multivariate analysis of variance, and repeated measures analysis of variance for balanced designs. PROC ANOVA also performs several multiple comparison tests.
CATMOD	fits linear models and performs analysis of variance and repeated measures analysis of variance for categorical responses.
GENMOD	fits generalized linear models and performs analysis of variance in the generalized linear models framework. The methods are particularly suited for discrete response outcomes .
GLM	performs analysis of variance, regression, analysis of covariance, repeated measures analysis, and multivariate analysis of variance. PROC GLM produces several diagnostic measures, performs tests for random effects, provides contrasts and estimates for customized hypothesis tests, performs several multiple comparison tests, and provides tests for means adjusted for covariates.
MIXED	performs mixed-model analysis of variance and repeated measures analysis of variance via covariance structure modeling. Using likelihood -based or method-of-moment estimates, PROC MIXED constructs statistical tests and intervals, allows customized contrasts and estimates, and computes empirical Bayes predictions .
NESTED	performs analysis of variance and analysis of covariance for purely nested random models.
NPAR1WAY	performs nonparametric one-way analysis of rank scores.
TTEST	compares the means of two groups of observations.
TRANSREG	fits univariate and multivariate linear models, optionally with spline and other nonlinear transformations.
VARCOMP	estimates variance components for random or mixed models.

The following section presents an overview of some of the fundamental features of analysis of variance. Subsequent sections describe how this analysis is performed with procedures in SAS/STAT software. For more detail, see the chapters for the individual procedures. Additional sources are described in the References section on page 67.