Pseudoreplication

From Wikipedia, the free encyclopedia

Pseudoreplication is a source of error in the statistical inferences drawn from observational studies, primarily in the fields of biology and ecology. Briefly, the error consists of assigning an exaggerated estimate of the statistical significance of a set of measurements by treating the data as independent observations when they are in fact interdependent. The term "pseudoreplication" was popularized by Stuart H. Hurlbert in an influential monograph published in 1984.^[1]

[edit] Types

In his paper, Hurlbert distinguished three types of pseudoreplication.

• Simple pseudoreplication occurs when there are no true replications of the observations because repeated measurements of the same sample are treated as independent in the statistical analysis.

• Temporal pseudoreplication is similar to simple pseudoreplication, the difference being that the multiple samples from each experimental unit are not taken simultaneously, but over a period of time.

• Sacrificial pseudoreplication occurs when observational data are pooled prior to statistical analysis (thus depressing the calculated variance of the observations), or when the multiple samples or measurements taken from each experimental unit are treated as independent (thus "sacrificing" some of the sample variance by inflating the denominator).

[edit] References