Short Communications: Use of linear orthogonal contrasts in analysis of environmental data

Abstract

Environmental studies often involve the analysis of many samples from a limited geographic region. The goal of these studies frequently is to compare samples from different areas within the region to determine relative levels of contamination or adverse effects. Analysis of variance (ANOVA) followed by multiple comparison tests is an often-used data analysis technique in these situations. However, ANOVA followed by the use of linear orthogonal contrasts is a more appropriate method of data analysis if obvious groups of samples for comparison can be identified a priori. The utility of the technique is demonstrated in an analysis of the growth (dry weight) of Chironomus tentans larvae during 10-d sediment toxicity assays of sediments from western Lake Erie and the Maumee River at Toledo, Ohio. The use of linear orthogonal contrasts revealed several pertinent aspects of the data set which were not apparent after ANOVA followed by multiple range tests.