In statistics, when analyzing collected data, the samples observed differ in such things as means and standard deviations, or proportions, from the population from which the sample is taken. This is sampling error and is controlled by ensuring that, as much as possible, the samples taken have no systematic characteristics and are a true random sample from all possible samples. If the observations are a true random sample, statistics can make probability estimates of the sampling error and allow the researcher to estimate what further experiments are necessary to minimize it. Used in reference to fitted models the term means experimental error which can contain both random error and systematic error. Used in reference to surveys it generally means a form of systematic error.
See more at Wikipedia.org...
a measure of the extent to which the chosen sample in a marketing research study can be expected to represent the total population on the characteristics being studied.