This book has been written for the researcher who uses computers to analyse field data on plant and animal communities and their environment. The book originates from a post-graduate course, held at the Wageningen Agricultural University in 1983, 1984 and 1985, for biologists, geographers, agronomists and landscape architects active in nature management, water management, environmental impact assessment and landscape planning. We have included topics that are useful in a wide range of ecological field studies: regression, to model the relation between a particular species and environmental variables, and to detect whether a species can be an indicator for a particular environmental variable; calibration, to infer about the environment from (indicator) species; ordination and cluster analysis, to summarize data on communities of species; and spatial analysis, to model and display the spatial variation in the environment and the communities studied.

A major aim of the book is to bridge the gap between exploratory methods, e.g. ordination and cluster analysis, and methods that allow statistical hypotheses to be tested, such as regression analysis and analysis of variance. This is important because environmental impact studies, for example, require careful design and analysis that is directed to detecting the effect of the impact variable. The by now traditional methods of ordination and cluster analysis are simply not sufficient; they may fail to detect the effect of the impact variable because of natural sources of variation.