Literally, denotes "without replacement" sampling with equal probabilities of selection within strata at the first stage. For all SUDAAN procedures, you can specify the sample design on the PROC statement, using the DESIGN options.
The NEST, TOTCNT, and (if included) SAMCNT statements must all list the same number of variables (one per sampling stage).
On the NEST statement follow the following guidelines :
Name the variables whose values identify the stratification and sampling stages.
List the NEST variables in order by stratification and sampling stages.
Use numeric values for the data record values that identify the sampling stages, as long as each distinct unit of a/the sampling stage is identified by a unique value.
Include variables identifying those substrata on the NEST statement if you used stratification at subsequent sampling stages.
Use the TOTCNT statement for DESIGN=WOR in order to compute finite population correction factors. Name the variables containing the population (sampling universe) counts for each strata or unit at each stage on the TOTCNT statement. List them in order by the corresponding variables on the NEST statement.
Include the SAMCNT statement, if you wish, to name in order those variables that contain sample counts for each stratum or unit at each stage. The variables are listed in order by the corresponding variables on the NEST and TOTCNT statements. SUDAAN verifies the counts given in the SAMCNT variables from the data.
The simplest NEST statement specification for the WOR design is:
NEST STRATA PSU;
where STRATA defines the stratification variable, and PSU defines the primary cluster variable. To define the entire population as a single stratum, use the SUDAAN keyword _ONE_ in place of STRATA. For Design=WOR, you will often specify additional stages of stratification and clustering on the NEST statement.