If an observation has a missing or nonpositive value for the SIZE variable, PROC SURVEYSELECT excludes that observation from the sample selection. The procedure provides a log note that gives the number of observations omitted due to missing or nonpositive size measures.

PROC SURVEYSELECT treats missing STRATA variable values like any other STRATA variable value. The missing values form a separate stratum.

If a value of _NSIZE_ (or SampleSize) is missing in the SAMPSIZE= secondary input data set, then PROC SURVEYSELECT cannot select a sample from that stratum. Similarly, if other secondary data set variables have missing values for a stratum, a sample cannot be selected from that stratum. These variables include _NRATE_, _MINSIZE_, _MAXSIZE_, _CERTSIZE_, and _CERTP_. See the section Secondary Input Data Set for details.

If a value of _ALLOC_, _VAR_, or _COST_ is missing in the secondary input data set, PROC SURVEYSELECT cannot compute the sample allocation.