MULTIPLE COMPARISONS IN CROSS-SECTIONALS DESIGNS WITH DEPENDENT DATA

The assumption of independence of observations may seem like a reazona ble assumption in examining data cross-sectional using experimental de signs of single-factor. The assumption of independence, however is sel dom verified by the investigators. By this reason we pretend to examin e the relationship between true Types I and II error probabilities und er the effects of departures from independence assumptions on hypothes is testing in six multiple comparison procedures.