StATS: Selecting randomly from a list (July 18, 2006)

A common task in research is to randomly select a subgroup from a list. For example, you have the names of 26 patients, Alpha, Bravo, Charlie, Delta, etc. Suppose you want to select a small number of patients in this list, but you want to do it randomly. There are several approaches that work, but the simplest is to arrange your list in a systematic order, attach a list of random numbers and then sort your list by those random numbers. The figure below shows how this process works.

If you needed five randomly selected patients, they would be Xray, Sierra, Charlie, Oscar, and Hotel. If you found out after the fact that one of these five was ineligible, then you would just go to the next name on your list (Mike).

This sort of procedure is used commonly in audits. You don't want to review 100% of the records, so you randomly select a subset for audit.

The Office of Inspector General at the United States Department of Health and Human Services provides some free software, RAT-STATS, for just this situation.

RAT-STATS is a package of statistical software tools designed to assist the user in selecting random samples and evaluating the audit results. The goal behind RAT-STATS was to develop valuable analytical tools that could be easily used by auditors.

It looks like a well written program.

Related material on these web pages

This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Randomized trials.