error

Definition:
In sampling surveys, error is defined as being of two kinds. (1) Statistical error (or sampling error) is the difference between the result actually achieved and that which would be achieved had data been collected from the entire survey population by the same methods as that used for the sample. Such error may be statistically estimated, and diminishes as sample size increases. (2) Non-statistical error (or BIAS) arises from many causes, such as failure to understand questions, failure to contact the correct sample, or to record the correct data. Error is often of great importance, but owing to the difficulty of measuring it by statistical methods, its effects are often neglected. Indeed, error is often assumed to be only statistical.

Cross-References:

Links:

Figures: