BookmarkSubscribeRSS Feed
bikashten
Fluorite | Level 6

Hi  All, 

I am trying to create the five non-overlapping groups from the original sample using proc surveyselect. I found this info in this webpage about how to create the independent groups from original sample frame (http://support.sas.com/kb/36/236.html). Statistically, I am wondering about this process: does this process qualify the survey sampling with the simple random sampling without replacement to create the five groups for statistical analysis?

 

OR

 

Do I need to take one group at each time using "simple random sampling (SRS)". The process is: select the first group of certain sample size from original sample frame using SRS method, select second group from "the remaining sample size frame" (excluding the observations in the first groups) and so on to create the five non-overlapping groups. 

 

Which methods are statistically correct from sampling point of view? 

2 REPLIES 2
ballardw
Super User

Depends on what you need. Does your sample frame have identified subpopulations that you want to sample from, such as might be indicated with a geography, product purchased, age or similar subpopulation definition?

If so then most likely you need 1) a single variable to identify which group the observation in your sample frame belongs to, 2) sort the data by that variable, 3) use a STRATA statement to indicate that variable defines groups in Proc Surveyselect.

 

Yes the referenced approach simulates an SRS and adds a variable they use called SampleId to indicate which group the selected observations belong in. Following the example logic it is a single pass. The key elements are specifying how many records you need at the first step (number of groups X number of records in each group) as the sample size in the Surveyselect SAMPSIZE option and then using the same number of records in the last data step in the

SampleID=ceil(_n_/10);

 

bikashten
Fluorite | Level 6
Hi Ballardw,
In this analysis, We don't have any identifiable sub-populations. Simply, we are trying to use the "SRS" method without any strata in proc surveyrselect to create the five non-overlapping groups. It is just a process about sampling procedures to create the non-overlapping groups.

Thank you,
Bikash

hackathon24-white-horiz.png

The 2025 SAS Hackathon Kicks Off on June 11!

Watch the live Hackathon Kickoff to get all the essential information about the SAS Hackathon—including how to join, how to participate, and expert tips for success.

YouTube LinkedIn

Mastering the WHERE Clause in PROC SQL

SAS' Charu Shankar shares her PROC SQL expertise by showing you how to master the WHERE clause using real winter weather data.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 911 views
  • 0 likes
  • 2 in conversation