This term is an acronym for a procedure entitled Analysis Of Variance.
This procedure employs the statistic (F) to test the statistical significance of the differences among the obtained MEANS of two or more random samples from a given population.
More specifically, using the Central Limit Theorem, one calculates two estimates of a population variance.
(1) An estimate in which the s² of the obtained means of the several samples is multiplied by n (the size of the samples).
(2) An estimate that is calculated as the average (mean) of all the s² of the several samples.
The statistic (F) is formed as the ratio of (1) over (2). If this ratio is sufficiently larger than 1, the observed differences among the obtained means are described as being statistically significant.
