Statistical Computing Environments for Social Research

The nature of statistics has changed from classical notions of hypothesis testing toward graphical and exploratory data analysis that exploits the flexibility of interactive computing and graphical displays. With contributions from some of the leading researchers in the social sciences and statistics, Statistical Computing Environments for Social Research describes seven statistical computing environments--APL2STAT, GAUSS, Lisp-Stat, Mathematica, S, SAS/IML, and Stata--that can be used effectively in graphical and exploratory modeling. These statistical computing environments, in contrast to a standard statistical package, provide programming tools for building other statistical applications. Programmability, flexible data structures, and--in the case of some of the computing environments--graphical interfaces and object-oriented programming permit researchers to take advantage of emerging statistical methodologies. Three additional chapters, describing the Axis, R-code, and ViSta statistical packages, demonstrate how researchers have extended one of the computing environments--Lisp-Stat--to produce significant statistical applications employing graphical interfaces to statistical software. To illustrate the capabilities of the seven statistical computing environments, each contributor uses the same data set to perform three computing tasks: robust regression, bootstrap resampling, and kernel-density estimation. The same data are analyzed in the chapters on Axis, R-code, and ViSta packages. The chapters in Statistical Computing Environments for Social Research illustrate important ideas and techniques in modern data analysis and statistical computing, ideas and techniques thatreaders will be able to apply in the more effective analysis of their own data.

Booknews

Describes seven statistical computing environments that can be used in graphical and exploratory modeling, demonstrating how researchers have extended one of the computing environments, Lisp-Stat, to produce statistical applications employing graphical interfaces to statistical software. Chapters on the seven packages use the same data set to perform computing tasks including robust regression and bootstrap resampling. Paper edition (unseen), $24.95. Annotation c. Book News, Inc., Portland, OR (booknews.com)

John Fox is Professor of Sociology at McMaster University in Hamilton, Ontario, Canada. He was previously Professor of Sociology and of Mathematics and Statistics at York University in Toronto, where he also directed the Statistical Consulting Service at the Institute for Social Research. Professor Fox earned a Ph.D. in Sociology from the University of Michigan in 1972. He has delivered numerous lectures and workshops on statistical topics, at such places as the summer program of the Inter-University Consortium for Political and Social Research and the annual meetings of the American Sociological Association. His recent and current work includes research on statistical methods (for example, work on three-dimensional statistical graphs) and on Canadian society (for example, a study of political polls in the 1995 Quebec sovereignty referendum). He is author of many articles, in such journals as Sociological Methodology, The Journal of Computational and Graphical Statistics, The Journal of the American Statistical Association, The Canadian Review of Sociology and Anthropology, and The Canadian Journal of Sociology. He has written several other books, including Applied Regression Analysis, Linear Models, and Related Methods (Sage, 1997), Nonparametric Simple Regression (Sage, 2000), and Multiple and Generalized Nonparametric Regression (Sage, 2000).