What do you think?
Rate this book
Structural Equations with Latent Variables
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.
- GenresPsychology
530 pages, Kindle Edition
First published April 28, 1989
2 people are currently reading
66 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 3 of 3 reviews
September 27, 2020
I mean... what can I say...
it was helpful but I feel like I’ll need it more for consultation than for a single in-depth read.
I appreciated how the author tried to make it as accessible as possible to everyone, and the appendixes were particularly helpful.
it was helpful but I feel like I’ll need it more for consultation than for a single in-depth read.
I appreciated how the author tried to make it as accessible as possible to everyone, and the appendixes were particularly helpful.
August 30, 2009
Structural equation modeling (SEM) is a common tool for psychologists and social scientists. Many individuals know how to use software packages that produce results, but few understand the underlying mathematics involved in the analytic technique. Ken Bollen's book is the most rigorous treatment of structural equation modeling, even 19 years after it was written. Whereas other books may be more accessible, they do not illustrate the inner workings of the technique. Although some may argue that such understanding is not warranted, SEM techniques will produce numerous errors (e.g., under-identification, non-positive definite models, Heywood cases, etc.) that may sometimes be avoided by simply thinking about the mathematics involved. Jim Steiger, the former president of the Society for Multivariate Experimental Psychology (and an individual who has come up with statistics -- namely, the Root Mean Squared Error of Approximation, or RMSEA) has argued that one should not teach SEM until he or she has read and comprehended Bollen's book.
In addition to simply describing the technique especially well (as if that weren't enough), the book addresses issues in philosophy of science by way of statistical explanation. For example, classic definitions of causality have established three necessary but not sufficient conditions (1. cause precedes effect, 2. correlation, and 3. lack of potential alternative explanations). While (2) has long been taken for granted, Bollen explains mathematically why it is not necessary (although in all fairness, perhaps in the example he describes, (3) was not met). Still, I applaud the text writer for illuminating this point and changing the way I think about causality.
In terms of my earlier comment about accessibility, I used this text as suggested reading for a course on SEM. One of my brightest students who is now doing a Ph.D. in quantitative psychology and focusing on dynamic factor analysis (a particular usage of SEM) began the semester determined to read all suggested texts and articles. She quickly stated that she hated Bollen. However, after grasping SEM enough to use it for many purposes, the book seems more accessible. While there is almost nothing in the way of calculus in the book, there is a great deal of matrix algebra that occurs and, although there is a brief review/introduction to matrix algebra in the appendix, if the reader does not have such a background, the book may be hard going.
Furthermore, many recent developments in SEM are not covered. For example, the latent variable growth curve model is not discussed. Because the field has advanced, Bollen (and Patrick Curran, a colleague at UNC in psychology -- Bollen is in sociology) have also written a book on that very topic.
Additionally, I cannot suggest this text for the purpose of understanding the software packages for SEM per se; there are better texts (see Raykov & Marcoulides, for example) for this purpose. However, once the mathematics are understood, understanding the software is fairly easy.
I would suggest this book for use in a second course in SEM, or in a first course for someone who is very mathematically inclined. Furthermore, it is great for social scientists and statisticians who want to understand the elegance of this statistical technique. Perhaps in the future, Bollen will revise the text to address the minor issues of this wonderful book.
In addition to simply describing the technique especially well (as if that weren't enough), the book addresses issues in philosophy of science by way of statistical explanation. For example, classic definitions of causality have established three necessary but not sufficient conditions (1. cause precedes effect, 2. correlation, and 3. lack of potential alternative explanations). While (2) has long been taken for granted, Bollen explains mathematically why it is not necessary (although in all fairness, perhaps in the example he describes, (3) was not met). Still, I applaud the text writer for illuminating this point and changing the way I think about causality.
In terms of my earlier comment about accessibility, I used this text as suggested reading for a course on SEM. One of my brightest students who is now doing a Ph.D. in quantitative psychology and focusing on dynamic factor analysis (a particular usage of SEM) began the semester determined to read all suggested texts and articles. She quickly stated that she hated Bollen. However, after grasping SEM enough to use it for many purposes, the book seems more accessible. While there is almost nothing in the way of calculus in the book, there is a great deal of matrix algebra that occurs and, although there is a brief review/introduction to matrix algebra in the appendix, if the reader does not have such a background, the book may be hard going.
Furthermore, many recent developments in SEM are not covered. For example, the latent variable growth curve model is not discussed. Because the field has advanced, Bollen (and Patrick Curran, a colleague at UNC in psychology -- Bollen is in sociology) have also written a book on that very topic.
Additionally, I cannot suggest this text for the purpose of understanding the software packages for SEM per se; there are better texts (see Raykov & Marcoulides, for example) for this purpose. However, once the mathematics are understood, understanding the software is fairly easy.
I would suggest this book for use in a second course in SEM, or in a first course for someone who is very mathematically inclined. Furthermore, it is great for social scientists and statisticians who want to understand the elegance of this statistical technique. Perhaps in the future, Bollen will revise the text to address the minor issues of this wonderful book.
Read
April 17, 2017good
Displaying 1 - 3 of 3 reviews