AbstractThe approximations implicit in Cochran's approximation to the Behrens‐Fisher test are examined. This test is often used in evaluating ground water quality problems with temporal and spatial variabilities. Concern is focused on the impact of the number of samples and the magnitude of the variances at the sampling locations. Recommendations are presented for utilization of the t‐test and Cochran's approximation, because Cochran's test is an approximation formula. Indications are provided for where the results may be giving answers different from those inten
展开▼