Simple random sampling

From Citizendium
Revision as of 22:40, 27 February 2009 by imported>Gene Shackman
Jump to navigation Jump to search

Simple random sampling (SRS) is the easiest way to take a sample from some population. The main idea is that everyone has exactly the same chance of being in the sample. Theoretically, then, the sample should be a good representation of the population.

Here is a simple example of getting a simple random sample. Suppose there is a club made up of one thousand people. If there is a complete membership list, the easiest way to get a simple random sample is to list all of the members, put a random number next to each member, then sort all of the members by the random numbers. The members are now in random order. To get a simple random sample, just take the first set of members, however many are desired.

Simple random sampling is simple, but in some cases, particularly where the sample is small, the result may not be representative, just due to the luck of the draw. Other methods may often be preferred.

Resources

Probability Sampling William M.K. Trochim, Research Methods Knowledge Base? 2006.

Simple Random Sampling The National EMSC Data Analysis Resource Center. Collecting Data. 2006.