The SURVEYSELECT Procedure

Allocation Output Data Set

When you specify the NOSAMPLE option in the STRATA statement, PROC SURVEYSELECT allocates the total sample size among the strata but does not select the sample. In this case, the OUT= data set contains the allocated sample sizes.

You can specify the name of the allocation output data set with the OUT= option in the PROC SURVEYSELECT statement. If you omit the OUT= option, the data set is named DATAn, where n is the smallest integer that makes the name unique.

The allocation output data set contains one observation for each stratum. The data set can include the following variables:

  • STRATA variables, which you specify in the STRATA statement.

  • Total, which is the total number of sampling units in the stratum.

  • Variance, which is the stratum variance. This variable is included if you specify the VAR , VAR=(values) , or VAR=SAS-data-set option for the ALLOC=OPTIMAL , ALLOC=NEYMAN , or MARGIN= allocation option.

  • Cost, which is the stratum cost. This variable is included if you specify the COST , COST=(values) , or COST=SAS-data-set option for ALLOC=OPTIMAL .

  • AllocProportion, which is the target allocation proportion (the proportion of the total sample size to allocate to the stratum). PROC SURVEYSELECT computes this proportion by using the allocation method that you specify.

  • SampleSize, which is the sample size allocated to the stratum.

  • ActualProportion, which is the actual proportion allocated to the stratum. The value of ActualProportion equals the allocated stratum sample size divided by the total sample size. This value can differ from the target AllocProportion because of rounding and other restrictions. For more information, see the section Sample Size Allocation.