This term is used in statistics in its ordinary sense, but most frequently occurs in connection with samples from different populations which may or may not be identical. If the populations are identical they are said to be homogeneous, and by extension, the sample data are also said to be homogeneous. In a more restricted sense populations may be said to be homogeneous in respect of some of their constraints, e.g. k populations with identical means but different dispersions are homogeneous in their means.