MATCHING SAMPLING DESIGNS AND SIGNIFICANCE TESTS IN ENVIRONMENTAL-STUDIES

Traditional ANOVA models are often successfully employed in environmental fields. However, there is a need to carefully examine the basic design, the assumptions and the form of the analysis so that valid tests of significance and standard errors are used. Examples will show that wrong conclusions can be drawn if this is not done. Environmental studies usually deal with mixed models that contain some factors that are fixed (i.e. particular levels or effects are of interest) and others that are random (i.e. the levels of these factors are sampled from a larger population). The type of replication, and consequent choice of error, applied to testing any fixed effect depend on the underlying assumptions made. Examples illustrate that correct matching of designs, tests and standard erors will allow these traditional methods to be applied, using mixed-models software, to a wide range of studies where standard textbook replication cannot be used.