Select Random Samples in SAS with PROC SURVEYSELECT in SAS

by finnstats

Select Random Samples in SAS with PROC SURVEYSELECT in SAS, you can utilize PROC SURVEYSELECT to draw random samples from a dataset efficiently.

Select Random Samples in SAS

Below are three common methods for using this procedure:

Example 1: Selecting a Simple Random Sample

To select a simple random sample from a dataset, you can use the following syntax:

proc surveyselect data=my_data
    out=my_sample
    method=srs     /* Utilize simple random sampling */
    n=5            /* Select a total of 5 observations */
    seed=1;        /* Set a seed for reproducibility */
run;

In this example, 5 random observations are selected from the complete dataset.

Example 2: Selecting a Stratified Random Sample

To conduct stratified random sampling, use the following code:

proc surveyselect data=my_data
    out=my_sample
    method=srs               /* Use simple random sampling within strata */
    n=2                      /* Select 2 observations from each stratum */
    seed=1;                  /* Set a seed for reproducibility */
    strata grouping_var;     /* Specify the variable for stratification */
run;

This example selects 2 random observations from each unique stratum defined by grouping_var.

Example 3: Selecting a Clustered Random Sample

For clustered random sampling, utilize this syntax:

proc surveyselect data=my_data
    out=my_sample
    n=2                      /* Select a total of 2 clusters */
    seed=1;                  /* Set a seed for reproducibility */
    cluster grouping_var;     /* Specify the variable for clustering */
run;

This example selects 2 random clusters from the dataset, including all observations from each selected cluster.

Practical Implementation with a Basketball Dataset

Let’s say we have a dataset with information about basketball players across different teams:

/* Create dataset */
data my_data;
    input team $ points;
    datalines;
A 12
A 14
A 22
A 35
A 40
B 12
B 10
B 29
B 33
C 40
C 25
C 11
C 10
C 15
;
run;

/* View dataset */
proc print data=my_data;
run;

Example 1: Simple Random Sample

To select a simple random sample of 5 observations from the dataset, use:

proc surveyselect data=my_data
    out=my_sample
    method=srs
    n=5
    seed=1;
run;

/* View sample */
proc print data=my_sample;
run;

The output will show 5 randomly selected observations from the dataset.

Example 2: Stratified Random Sample

For stratified sampling, where 2 observations are chosen from each team, the code would be:

proc surveyselect data=my_data
    out=my_sample
    method=srs
    n=2
    seed=1
    strata team;           /* Stratification is done by the 'team' variable */
run;

/* View sample */
proc print data=my_sample;
run;

The resulting sample will consist of 2 observations from each basketball team.

Example 3: Clustered Random Sample

To perform clustered sampling using teams as clusters, use the following code:

proc surveyselect data=my_data
    out=my_sample
    n=2
    seed=1
    cluster team;         /* Clustering based on the 'team' variable */
run;

/* View sample */
proc print data=my_sample;
run;

In this case, every observation from the selected clusters (teams) will be included in the sample.

Summary

Using PROC SURVEYSELECT, you can easily obtain random samples from a dataset in SAS through various methods, including simple random, stratified, and clustered sampling.

By selecting random samples, you can effectively analyze data while maintaining the integrity of the underlying population.

SAS Archives » FINNSTATS

Select Random Samples in SAS with PROC SURVEYSELECT in SAS