The following statements are available in the SURVEYMEANS procedure:
PROC SURVEYMEANS <options> <statistic-keywords>;
BY variables;
CLASS variables;
CLUSTER variables;
DOMAIN variables <variable*variable variable*variable*variable …> </ option>;
POSTSTRATA variables / PSTOTAL= <option>;
POSTSTRATA variables / PSPCT= <option>;
RATIO <'label'> variables / variables;
REPWEIGHTS variables </ options>;
STRATA variables </ option>;
VAR variables;
WEIGHT variable;
The PROC SURVEYMEANS statement invokes the procedure. It optionally names the input data sets, specifies statistics for the procedure to compute, and specifies the variance estimation method. The PROC SURVEYMEANS statement is required.
The VAR statement identifies the variables to be analyzed. The CLASS statement identifies numeric variables that are to be analyzed as categorical variables. The STRATA statement lists the variables that form the strata in a stratified sample design. The CLUSTER statement specifies cluster identification variables in a clustered sample design. The DOMAIN statement lists the variables that define domains for subpopulation analysis. The RATIO statement requests ratio analysis for means or proportions of analysis variables. The WEIGHT statement names the sampling weight variable. The POSTSTRATA statement lists the variables that are used to form poststrata for poststratification. The REPWEIGHTS statement names replicate weight variables for BRR or jackknife variance estimation. You can use a BY statement with PROC SURVEYMEANS to obtain separate analyses for groups defined by the BY variables.
All statements can appear multiple times except the PROC SURVEYMEANS statement, POSTSTRATA statement, and WEIGHT statement, which can appear only once.
The rest of this section gives detailed syntax information for the BY, CLASS, CLUSTER, DOMAIN, POSTSTRATA, RATIO, REPWEIGHTS, STRATA, VAR, and WEIGHT statements in alphabetical order after the description of the PROC SURVEYMEANS statement.