This book is about making valid inferences from scientific data when a meaningful analysis depends on a model. Our general objective is to provide scientists with a readable text giving practical advice for the analysis of empirical data under an information-theoretic paradigm. We first assume that an exciting scientific question has been carefully posed and relevant data have been collected, following a sound experimental design or probabilistic sampling program. Often, little can be salvaged if data collection has been seriously flawed or if the question was poorly posed (Hand 1994). We realize, of course, that these issues are never as ideal as one would like. However, proper attention must be placed on the collection of data (Chatfield 1991, 1995a). We stress inferences concerning understanding the structure and function of the biological system, estimators of relevant parameters, and valid measures of precision; we say somewhat less about formal prediction.
KeywordsModel Selection Biological Science True Model Candidate Model Approximate Model
Unable to display preview. Download preview PDF.