Random Sample

Aerial view of a small group of people isolated within a larger groupA random sample is a subset of a larger group. Researchers use random sampling techniques to ensure that each member of the larger group has an equal likelihood of inclusion in the smaller study group.

Methods of Random Sampling

Researchers use a variety of methods to select random samples, including complex, computer-generated algorithms. In smaller studies, researchers might assign numbers to subjects and pick them randomly or use other methods to randomly select members of the sample group. These methods are slightly less reliable and might result in the overrepresentation of certain group members in the smaller study group.

Random samples are designed to be sufficiently large to obtain useful information. A researcher studying the political beliefs of Americans would not glean much information if his or her random sample consisted of one person. The appropriate size for a random sample depends upon the larger group; very large groups require larger random samples.

Random Samples in Science

Random sampling reduces, but does not eliminate, the risk of statistical biases and errors. A researcher studying political affiliations could pick and choose participants that meet his or her hypothesis, but random sampling prevents this. A properly selected sample of a sufficiently large size is typically representative of the larger group, which may be too big to study. Studies using random samples are therefore able to make predictions about behavior and estimates about the beliefs, inclinations, or motivations of the larger group.

Random Samples in Popular Culture

Many of the polls and studies people rely on to get information about their society are the results of random sampling. The most predictive presidential election polls, for example, use random samples of likely voters, voters in swing states, or undecided voters to gain information about potential election outcomes. When samples are not random, they may not be representative of the larger population, compromising the predictive value of the poll.

Reference:

  1. Colman, A. M. (2006). Oxford dictionary of psychology. New York, NY: Oxford University Press.

Last Updated: 08-20-2015