PROC ANOVA can compute means of the dependent variables for any effect that appears on the righthand side in the MODEL statement.
You can use any number of MEANS statements, provided that they appear after the MODEL
statement. For example, suppose A
and B
each have two levels. Then, if you use the following statements
proc anova; class A B; model Y=A B A*B; means A B / tukey; means A*B; run;
means, standard deviations, and Tukey’s multiple comparison tests are produced for each level of the main effects A
and B
, and just the means and standard deviations for each of the four combinations of levels for A
*B
. Since multiple comparisons options apply only to main effects, the single MEANS statement
means A B A*B / tukey;
produces the same results.
Options are provided to perform multiple comparison tests for only main effects in the model. PROC ANOVA does not perform multiple comparison tests for interaction terms in the model; for multiple comparisons of interaction terms, see the LSMEANS statement in Chapter 45: The GLM Procedure.
Table 26.4 summarizes the options available in the MEANS statement.
Table 26.4: Options Available in the MEANS Statement
Option 
Description 

Perform multiple comparison tests 

Performs Bonferroni t tests of differences between means for all main effect means 

Performs Duncan’s multiple range test on all main effect means 

Performs Dunnett’s twotailed t test 

Performs Dunnett’s onetailed t test, testing if any treatment is significantly less than the control 

Performs Dunnett’s onetailed t test, testing if any treatment is significantly greater than the control 

Performs Gabriel’s multiplecomparison procedure on all main effect means 

Performs the RyanEinotGabrielWelsch multiple range test 

Performs Scheffé’s multiplecomparison procedure 

Performs pairwise t tests on differences between means with levels adjusted according to Sidak’s inequality 

Performs pairwise comparisons based on the studentized maximum modulus and Sidak’s uncorrelatedt inequality 

Performs the StudentNewmanKeuls multiple range test 

Performs pairwise t tests 

Performs Tukey’s studentized range test (HSD) 

Performs the WallerDuncan kratio t test 

Specify additional details for multiple comparison tests 

Specifies the level of significance for comparisons among the means. 

Presents results from options as confidence intervals 

Options as intervals for the mean of each level of the variables specified 

Specifies the error mean square used in the multiple comparisons 

Specifies the Type 1/Type 2 error seriousness ratio for the WallerDuncan test 

Presents results of options by listing the means in descending order and indicating nonsignificant subsets by line segments 

Prevents the means from being sorted into descending order 

Test for homogeneity of variances 

Requests a homogeneity of variance test 

Compensate for heterogeneous variances 

Requests the Welch (1951) varianceweighted oneway ANOVA 
Descriptions of these options follow. For a further discussion of these options, see the section Multiple Comparisons in Chapter 45: The GLM Procedure.