By Neil Wrigley
Categorical Data Analysis for Geographers and Environmental Scientists, originally published in 1985, provided the first account of the new integrated approaches to the analysis of categorical data designed specifically to meet the needs of the geographer and the environmental scientist. It is intended to be the logical sequel to the type of multivariate statistics course that most researchers in those fields will have encountered. As such, it is much more comprehensive in scope than other texts in the field. The book contains more than 40 empiric illustrations (from oil exploration to transport planning in cities), which are designed to form an integrated part of the text. These serve to link the theory to the practice of geographical and environmental science research.
Dr. Wrigley’s book was a milestone in data analysis in the spatial sciences. It provided an account of a revolution that has swept through an area of statistical methodology and that has transformed the practice of data analysis for social and environmental scientists.
The book is sensitive to the likely statistic/mathematic backgrounds of geographers and environmental scientists and is written in a fashion that should be accessible to all higher level undergraduate and postgraduate students, faculty and researchers in those areas.
"One of the liveliest areas of statistics during the past 15 years has been the analysis of categorical data, counts or frequencies of different classes. Historically a poor relation of the analysis of continuous data, the field has been unified by the development of related families of models (logit, logistic, log-linear, and so on), which supersede or subsume earlier approaches based on measures of association or chi-square testing, many of them rather isolated or ad hoc. As, among geographers, Neil Wrigley has been the leading advocate and exponent of these new methods, it is especially appropriate that he should produce a guidebook for his colleagues. Thorough, professional and beautifully laid out, his text leads the reader steadily through the new terrain, with many helpful remarks on confusing points like notation . . . . He can be congratulated on a major contribution to the geographical literature."
The Times Higher Education Supplement
"The book must certainly be regarded as a major contribution to quantitative analysis in geography and environmental science. It will undoubtedly appear daunting to many readers at first, but is worth persevering with and has the merit that it will bear repeated reading as the expertise of the analyst develops."
Transactions of the Institute of British Geographers
"As a postgraduate student attending a residential course on Data Collection and Analysis at Bristol University, I vividly recall being introduced to categorical data analysis by Wrigley. Over the years, this reading and research material has been sifted, sorted, and enlarged to create the contents of this substantial publication. Those who regard quantitative geography as passé should read this book. It describes recent developments in data analysis as a ‘second quantitative revolution’ which swept through the social sciences in the late 1970s and early 1980s. Although these claims may be overstated, there is little doubt that logit, logistic, and log-linear models have colonised many areas of geography and environmental science. In these areas, a basic understanding of these modelling techniques has become a vital part of the technical armoury used by researchers. This book will play a valuable role in the initiation ceremony for those already versed in normal regression who wish to move into this newer area. . . . For those who are convinced of the usefulness of sophisticated models, this book makes a very comprehensive ‘reference manual’, bringing together in one text many of the new developments in categorical data analysis, to which Wrigley has been a major contributor."
Environment and Planning A
"Wrigley has succeeded in producing a first-rate book which covers many of the major geographically relevant developments of categorical data analysis and presents them in an integrative, stimulating, and comprehensible way. The book should become both a standard text for courses on statistical methodology at the higher-level undergraduate and postgraduate level and a must for the researcher involved in the modelling of spatial choice behaviour and in analyzing categorical data."
Neil Wrigley Ph.D. (Cambridge), D.Sc. (Bristol) is Professor of Geography at the University of Southampton, UK (where he has been Head of Department 1992-95, 1999-2001). Previously, he was Professor and Head of the Department of City & Regional Planning, University of Wales, Cardiff, and before that Reader in Geography at the University of Bristol. He has held visiting appointments at several universities in the USA, Canada and Australia, an Erskine Fellowship in New Zealand, and was Senior Research Fellow at St Peter’s College, Oxford (1996-97). He is currently founding editor of the Journal of Economic Geography (Oxford University Press) and was previously (1988-93) editor of Transactions of the Institute of British Geographers.
Author of more than 130 academic papers in leading journals and 10 books and edited volumes, his research has focused on two main areas. During the 1990s, and currently, its focus has been on economic geography – with particular reference to the corporate restructuring, regulation and globalization of the retail industry. During the 1970s and 1980s, it focused on quantitative social science and spatial statistics – particularly the fields of categorical data analysis, discrete choice modelling, and longitudinal/panel data analysis. Categorical Data Analysis, first published by Longman in 1985 and now republished by The Blackburn Press, comes from that earlier period of his work when he was based at the University of Bristol and a member of the UK Economic and Social Research Council’s Research Methods and Resources Committee. The book has long been recognized as a landmark volume in quantitative geography and has been extensively cited and widely used across many disciplines as a research methods text for over 15 years. Although the methods it describes are now a standard part of the practice of data analysis in the social and environmental sciences (and are available via all the leading statistical software packages) its account of the revolution that swept through this area of statistic methodology retains its vitality and pertinence. Moreover, the clarity of its pedagogic approach and its successful student-oriented integration of empirical illustration and statistical discussion has ensured its continuing popularity as a research methods text.