Ten common misuses of statistics in agronomic research and reporting

Abstract

ABSTRACT Ten common misuses of statistics in agronomic research and reporting are discussed. Some of these are a result of changes in statistical philosophy over the years to which biologists in general, and agronomists in particular, have not responded in terms of their data analytic and interpretational techniques. Others have been created by an overdependence on computers without careful study of the basic data patterns or without careful consideration of the calculations which the computer is performing. The importance of planning experiments properly with a view toward subsequent analysis is emphasized. Careful, well‐controlled experimental technique is also recommended. Proper planning usually assures that logical comparisons can be made without resorting to the use of mechanical procedures such as multiple comparisons. The matter of misuse of multiple comparison procedures such as Duncan's Multiple Range Test is also discussed. It is pointed out that in cases where logical structure doesn't exist in the treatments (a rare event), the use of multiple comparison procedures is valid.