Although an value is computed for all SS in the analysis by using the residual MS as an error term, you can request additional tests that use other effects as error terms. You need a TEST statement when a nonstandard error structure (as in a split plot) exists.

Caution:The ANOVA procedure does not check any of the assumptions underlying the statistic. When you specify a TEST statement, you assume sole responsibility for the validity of the statistic produced. To help validate a test, you might want to use the GLM procedure with the RANDOM statement and inspect the expected mean squares. In the GLM procedure, you can also use the TEST option in the RANDOM statement.

You can use as many TEST statements as you want, provided that they appear after the MODEL statement.

You can specify the following terms in the TEST statement.

H=effects

specifies which effects in the preceding model are to be used as hypothesis (numerator) effects.

E=effect

specifies one, and only one, effect to use as the error (denominator) term. The E= specification is required.

The following example uses two TEST statements and is appropriate for analyzing a split-plot design.

   proc anova;
      class a b c;
      model y=a|b(a)|c;
      test h=a e=b(a);
      test h=c a*c e=b*c(a);
   run;