n. in statistics, refers to the classic dilemma on how to compute and test for the equality (or the difference) between the means of two normally-distributed populations when one cannot make any assumptions, whatsoever, that their variances are equal. Named after W.V. Behrens and Ronald Fisher.
BEHRENS-FISHER PROBLEM: “Still an unsolved problem in statistics, the Behrens-Fisher Problem becomes even more difficult to solve because there are many different interpretations on how to go about it.”