`SAMPLINGUNIT | CLUSTER`

variables < / options > ;

The SAMPLINGUNIT statement names variables that identify the sampling units as groups of observations (clusters). The combinations of categories of SAMPLINGUNIT variables define the sampling units. If there is a STRATA statement, sampling units are nested within strata.

When you use a SAMPLINGUNIT statement to define units (clusters), PROC SURVEYSELECT selects a sample of these units by using the selection method and design parameters that you specify in the PROC SURVEYSELECT statement. If you do not use a SAMPLINGUNIT statement, then PROC SURVEYSELECT uses the input data set observations as sampling units by default.

The SAMPLINGUNIT variables are one or more variables in the DATA= input data set. These variables can be either character or numeric. The formatted values of the SAMPLINGUNIT variables determine
the SAMPLINGUNIT variable levels. Thus, you can use formats to group values into levels. See the FORMAT procedure in the
*Base SAS Procedures Guide* and the FORMAT statement and SAS formats in
*SAS Formats and Informats: Reference* for more information.

You can use a SAMPLINGUNIT statement with any equal probability selection method or PPS selection method. The SAMPLINGUNIT statement is not available for Poisson sampling (METHOD=POISSON).

If you specify the PPS option in the SAMPLINGUNIT statement and do not specify a SIZE statement, then the procedure computes sampling unit size as the number of observations in the sampling unit. If you specify a SIZE statement with a SAMPLINGUNIT statement, then the procedure computes sampling unit size by summing the size measures of all observations in the sampling unit.

By default, PROC SURVEYSELECT sorts the input data set by the SAMPLINGUNIT variables within strata before sample selection. This groups the observations into sampling units and orders the sampling units by the SAMPLINGUNIT variables. If you do not want the procedure to sort the input data set by the SAMPLINGUNIT variables, then specify the PRESORTED option in the SAMPLINGUNIT statement. By using the PRESORTED option, you can provide the order of the sampling units for systematic and sequential selection methods. The CONTROL statement is not available with the SAMPLINGUNIT statement.

Note that the SAMPLINGUNIT statement defines groups of observations (clusters) to use as sampling units, and PROC SURVEYSELECT selects a sample of these units. When you use a SAMPLINGUNIT statement, PROC SURVEYSELECT does not select samples of observations from within the sampling units (clusters). To select independent samples within groups, use the STRATA statement.

You can specify the following options in the SAMPLINGUNIT statement after a slash (/):