sampling error

sampling error

[′samp·liŋ ‚er·ər]

(statistics)

That portion of the difference between the value of a statistic derived from observations and the value that it is supposed to estimate; attributed to the fact that samples represent only a portion of a population.

McGraw-Hill Dictionary of Scientific & Technical Terms, 6E, Copyright © 2003 by The McGraw-Hill Companies, Inc.

sampling error

the difference between the ‘true’ value of a characteristic within a population and the value estimated from a sample of that population. ‘Error’ occurs because no SAMPLE can be expected to exactly represent the parent population from which it was drawn. To minimize, and to be able to estimate, sampling error, it is necessary to ensure that the selection of the sample is RANDOM, and this is normally done by random numbers or systematic sampling. Sampling error is not the same as BIAS or systematic error, which may occur due to the process of data collection, but is nothing to do with the sample selection.

Collins Dictionary of Sociology, 3rd ed. © HarperCollins Publishers 2000