Product Description

Analysis of Ordinal Categorical Data Alan Agresti Statistical Science Now has its first coordinated manual of methods for analyzing ordered categorical data. This book discusses specialized models that, unlike standard methods underlying nominal categorical data, efficiently use the information on ordering. It begins with an introduction to basic descriptive and inferential methods for categorical data, and then gives thorough coverage of the most current developments, such as loglinear and logit models for ordinal data. Special emphasis is placed on interpretation and application of methods and contains an integrated comparison of the available strategies for analyzing ordinal data. This is a case study work with illuminating examples taken from across the wide spectrum of ordinal categorical applications. 1984 (0 471-89055-3) 287 pp. Regression Diagnostics Identifying Influential Data and Sources of Collinearity David A. Belsley, Edwin Kuh and Roy E. Welsch This book provides the practicing statistician and econometrician with new tools for assessing the quality and reliability of regression estimates. Diagnostic techniques are developed that aid in the systematic location of data points that are either unusual or inordinately influential; measure the presence and intensity of collinear relations among the regression data and help to identify the variables involved in each; and pinpoint the estimated coefficients that are potentially most adversely affected. The primary emphasis of these contributions is on diagnostics, but suggestions for remedial action are given and illustrated. 1980 (0 471-05856-4) 292 pp. Applied Regression Analysis Second Edition Norman Draper and Harry Smith Featuring a significant expansion of material reflecting recent advances, here is a complete and up-to-date introduction to the fundamentals of regression analysis, focusing on understanding the latest concepts and applications of these methods. The authors thoroughly explore the fitting and checking of both linear and nonlinear regression models, using small or large data sets and pocket or high-speed computing equipment. Features added to this Second Edition include the practical implications of linear regression; the Durbin-Watson test for serial correlation; families of transformations; inverse, ridge, latent root and robust regression; and nonlinear growth models. Includes many new exercises and worked examples. 1981 (0 471-02995-5) 709 pp.

Product Details

• Amazon Sales Rank: #497022 in Books
• Published on: 1989-04-28
• Original language: English
• Number of items: 1
• Binding: Hardcover
• 514 pages

From the Publisher
A comprehensive introduction to the general structure equation systems--commonly known as the LISREL model--used for quantitative research in the social sciences. Unified approach presents path analysis, recursive and nonrecursive models, classical econometrics, and confirmatory factor analysis as special cases of a general model. Also discusses application of these techniques to empirical examples, including some LISREL and EQS programs.

From the Inside Flap
Statistical modeling and its associated terminology have seen tremendous change over the past ten years. Lisrel, covariance structures, latent variables, multiple indicators, and path models are now common phrases used in the analysis of statistical data. The structural equation models associated with these terms are changing researchers’ perspectives on statistical modeling and closing the gap between the way social scientists think substantively and the way they analyze data. In short, these models encompass and extend regression, econometric, and factor analysis procedures. Structural Equations with Latent Variables is a comprehensive treatment of the general structural equation system better known as the Lisrel model. The book serves three purposes. First, it demonstrates the generality of this model. Rather than treating path analysis, recursive and nonrecursive models, classical econometrics, and confirmatory factor analysis as unique, they are treated as special cases of a common model. The second purpose is to emphasize the application of these techniques. Empirical examples appear throughout. Several chapters contain some of the Lisrel or EQS programs the author used to obtain the results for the empirical examples. Finally, the book explores the crucial role played by substantive expertise in most stages of the modeling process. Specifically, the book is arranged as follows: After an introductory overview in Chapter 1, Chapter 2 introduces several methodological tools, while Chapter 3 addresses causality. The regression/econometric models for observed variables are the subject of Chapter 4. In Chapter 5, the consequences of random measurement error in the observed variable model are explained. Once it is recognized that variables are measured with error, the relationship between the error-free variable and the observed variable needs to be examined. Chapter 6 does this. Chapter 7 treats confirmatory factor analysis which enables the estimation of measurement models such as those in Chapter 6. Completing this work in Chapters 8 and 9 is the general structural equation model with latent variables. Chapter 8 focuses on the basics, while Chapter 9 treats more advanced topics, including distribution-free estimators and categorical observed variables. Structural equation models can be presented two ways—from the general model to simpler models,or from simpler models to the general model. Structural Equations with Latent Variables develops from the latter strategy, starting with the regression/econometric and factor analysis models and presenting them from the perspective of the general model. Structural Equations with Latent Variables fills the gap existing in the treatment of this subject between introductory texts and specialized papers. It provides social scientists, market researchers, applied statisticians, other analysts, and graduate students with a thorough examination of Lisrel/structural equation models. At the same time it presents new material on measurement reliability and validity, overall fit indices, model identification, and other topics.

From the Back Cover
Analysis of Ordinal Categorical Data Alan Agresti Statistical Science Now has its first coordinated manual of methods for analyzing ordered categorical data. This book discusses specialized models that, unlike standard methods underlying nominal categorical data, efficiently use the information on ordering. It begins with an introduction to basic descriptive and inferential methods for categorical data, and then gives thorough coverage of the most current developments, such as loglinear and logit models for ordinal data. Special emphasis is placed on interpretation and application of methods and contains an integrated comparison of the available strategies for analyzing ordinal data. This is a case study work with illuminating examples taken from across the wide spectrum of ordinal categorical applications. 1984 (0 471-89055-3) 287 pp. Regression Diagnostics Identifying Influential Data and Sources of Collinearity David A. Belsley, Edwin Kuh and Roy E. Welsch This book provides the practicing statistician and econometrician with new tools for assessing the quality and reliability of regression estimates. Diagnostic techniques are developed that aid in the systematic location of data points that are either unusual or inordinately influential; measure the presence and intensity of collinear relations among the regression data and help to identify the variables involved in each; and pinpoint the estimated coefficients that are potentially most adversely affected. The primary emphasis of these contributions is on diagnostics, but suggestions for remedial action are given and illustrated. 1980 (0 471-05856-4) 292 pp. Applied Regression Analysis Second Edition Norman Draper and Harry Smith Featuring a significant expansion of material reflecting recent advances, here is a complete and up-to-date introduction to the fundamentals of regression analysis, focusing on understanding the latest concepts and applications of these methods. The authors thoroughly explore the fitting and checking of both linear and nonlinear regression models, using small or large data sets and pocket or high-speed computing equipment. Features added to this Second Edition include the practical implications of linear regression; the Durbin-Watson test for serial correlation; families of transformations; inverse, ridge, latent root and robust regression; and nonlinear growth models. Includes many new exercises and worked examples. 1981 (0 471-02995-5) 709 pp.

Customer Reviews

detailed coverage of an every growing topic in applied statistics
The software Lisrel was developed to model and analyze data using structural equation models which involve the introduction of latent variables. Although this topic has historically been most commonly used in the social sciences including psychology and sociology, it is finding a wide range of applications as statisticians encounter more and more problems where it is appropriate to use latent variables.

Bollen provides a thorough treatment of the topic that has advanced some since the publication of the book . This is still the best source for a detailed account of the methods. Bengt Meuthen at UCLA was one of the pioneers of the methodology and his books and papers provide good additional sources for the reader who wants to understand the theory and the software tools.

Bollen still has what it takes
Bollen is comprehensive and thorough, sometimes even to a fault. This work is definitely for the advanced academic who works with structural equations on a regular basis - certainly not for the faint of heart.

Although the field has made advances in the thirteen or so years since this book was first published, I found that Bollen has weathered well, is still suitable for instruction and practice, and should remain so for many years to come.

Bollen excels at a very difficult task
This book sticks out like a beautiful flower in a bed of weeds.

There is a dearth of really good books available to sociologists interested in deepening their knowledge of statistics in general, and structural equation modelling in particular. Most of the books I encounter are either too simple or too difficult. There is either too little, or too much detail. I suppose that in general writers cater to two types of statistics users: the statisticians, and those applied users who prefer the "black box" approach in running their regressions. A good example of a bad statistics books for sociologists: the often-cited econometrics textbooks that read like dictionaries by providing lists of formulas rather than integrating knowledge by carefully explaining general concepts.

And then there is Bollen's 1989 SEM book. Almost a class by itself. He achieves a balance between comprehensiveness and comprehensibility ideal for the sociologist. Bollen excels at the task of explaining statistics. If someone is thinking of taking a first class in SEM, I would recommend getting a copy of the syllabus. If the syllabus doesn't include this book, I would recommend that the person consider dropping the class and reading this book if at all possible.

Someone needs to convince Bollen to write books about every statistics topic he has mastered, and then to learn other topics and write books on those. Bollen is a resource that should not be missed.