Principles and Practice of Structural Equation Modeling.pdf

1、Principles and Practice of Structural Equation ModelingMethodology in the Social SciencesDavid A.Kenny,Founding EditorTodd D.Little,Series EditorThis series provides applied researchers and students with analysis and research design books that emphasize the use of methods to answer research question

2、s.Rather than emphasizing statistical theory,each volume in the series illustrates when a technique should(and should not)be used and how the output from available software programs should(and should not)be interpreted.Common pitfalls as well as areas of further development are clearly articulated.R

3、ecent VolumesCONFIRMATORY FACTOR ANALYSIS FOR APPLIED RESEARCHTimothy A.BrownDYADIC DATA ANALYSISDavid A.Kenny,Deborah A.Kashy,and William L.CookMISSING DATA:A GENTLE INTRODuCTIONPatrick E.McKnight,Katherine M.McKnight,Souraya Sidani,and Aurelio Jos FigueredoMuLTILEVEL ANALYSIS FOR APPLIED RESEARCH:

4、ITS JuST REGRESSION!Robert BickelTHE THEORY AND PRACTICE OF ITEM RESPONSE THEORYR.J.de AyalaTHEORY CONSTRuCTION AND MODEL-BuILDING SKILLS:A PRACTICAL GuIDE FOR SOCIAL SCIENTISTSJames Jaccard and Jacob JacobyDIAGNOSTIC MEASuREMENT:THEORY,METHODS,AND APPLICATIONSAndr A.Rupp,Jonathan Templin,and Robert

5、 A.HensonAPPLIED MISSING DATA ANALYSISCraig K.EndersADVANCES IN CONFIGuRAL FREQuENCY ANALYSIS Alexander A.von Eye,Patrick Mair,and Eun-Young MunPRINCIPLES AND PRACTICE OF STRuCTuRAL EQuATION MODELING,Third EditionRex B.KlinePrinciples and Practice of Structural Equation ModelingTHIRD EDITIONRex B.Kl

6、ineSeries Editors Note by Todd D.LittleTHE GuILFORD PRESS New York London 2011 The Guilford PressA Division of Guilford Publications,Inc.72 Spring Street,New York,NY All rights reservedNo part of this book may be reproduced,translated,stored in a retrieval system,or transmitted,in any form or by any

7、 means,electronic,mechanical,photocopying,microfilming,recording,or otherwise,without written permission from the Publisher.Printed in the United States of AmericaThis book is printed on acid-free paper.Last digit is print number:9 8 7 6 5 4 3 2 1Library of Congress Cataloging-in-Publication DataKli

8、ne,Rex B.Principles and practice of structural equation modeling/Rex B.Kline.3rd ed.p.cm.(Methodology in the social sciences)Includes bibliographical references and index.ISBN 978-1-60623-877-6(hardcover)ISBN 978-1-60623-876-9(pbk.)1.Structural equation modeling.2.Social sciencesStatistical methods

9、Data processing.I.Title.QA278.K585 2011 519.53dc22 2010020226For my familyJoanna,Julia Anne,and Luke ChristopherGreat knowledge sees all in one.Small knowledge breaks down into the many.Chuang Tzu,fourth or third century bce,China (in Merton,1965,p.40)vii Series Editors NoteIt is a pleasure to write

10、 an introductory note for a book that is so popular you can sim-ply refer to it as“the Kline book”and everyone will know what you mean.Rex Kline is a quantitative expert with that rare ability to provide clear and accessible guidance on how to best use structural equation modeling(SEM)to answer crit

11、ical research ques-tions.It takes a very special author to overcome students fears and engage them in the principles and practice of SEM.In each edition of his book Kline has done just this,and with each edition it gets better and better!The literature on SEM is always evolving and being refined.To

12、keep up with this literature is a challenge even to the quantitative expert.Thankfully,we have Rex Kline to rely on.If you are a fan of the earlier editions,I think you will find the improvements to the third edition both welcome and enlightening.For example,based on the helpful feed-back of readers

13、 like you,Kline has reorganized Part II to model the phases and steps one follows in a typical analysis,from initial model specification,to identification consider-ations,to parameter estimation,to evaluating hypotheses,and,finally,to model respeci-fication.Pedagogically,he has also added useful exe

14、rcises with answers and informative topic boxes that cover key concepts,core techniques,and specialized issues in the world of SEM.He also elegantly addresses“troublesome”examples,which leads to discussions of how to handle known problems that arise in SEM analyses.If you have not looked at“the Klin

15、e book,”or not in a while,I encourage you to take a look at this third edition.Kline provides an accurate and authoritative“translation”of the technical world of SEM for students and applied researchers alike.It is the Rosetta stone for understanding SEM and for showing substantive researchers how t

16、o use SEM in the conduct of their science.It strikes a tidy balance between the technical and the practical aspects of SEM so that you will be able to both clarify and expand your knowl-edge of the vast possibilities of SEM.It serves as a conduit for substantive researchers to stay connected to the

17、ever-changing field of SEM.Since the first edition,the books success is the consensus viewpoint of critical reviewers and researcherswho lean heavily on it.The second edition was a complete viii Series Editors Noteand thorough update to the best practice in the field and saw pedagogic changes that e

18、levated the second edition to a bonafide bestseller in the social and behavioral sciences.And the third edition is nothing short of remarkable in terms of its authoritative sum-mary of an ever-advancing field.The chapter dedicated to the use of different software packages(Chapter 4)is expanded.Cover

19、age of assessing the identification status of mea-surement models with correlated errors and complex indicators is updated in Chapter 6.Chapter 7 gives expanded coverage of estimation,including more specific information for analyzing models with categorical outcome variables.Chapter 12 expands cover

20、age of estimating interactive effects and multilevel SEMs.And the list goes on!You can see by the praise of the many reviewers of this latest edition that Rex Kline has managed to take“the Kline book”to another level of clarity and coverage.Todd D.Little University of Kansas Lawrence,Kansas ix Prefa

21、ce and AcknowledgmentsIts not often in life that you get three chances at something.Thus,it was a privilege for me to write the third edition of this book.This edition builds on the strengths of the second by presenting structural equation modeling(SEM)in a clear,accessible way for readers without e

22、xtensive quantitative backgrounds.Many new examples of the applica-tion of SEM to actual research problems are included in this edition,but,like the second edition,these examples come from a wide range of disciplines,including education,psychometrics,business,and psychology.I selected some of these

23、examples because there were technical problems in the analysis,such as when output from a computer program contains error messages.These“troublesome”examples give a context for dis-cussing how to handle various problems that can crop up in SEM analyses.That is,not all applications of SEM described i

24、n this book are picture perfect,but neither are actual research problems.There are many changes in this edition from the second edition,all intended to enhance the pedagogical presentation of SEM and cover recent developments in the field,especially concerning how structural equation modelsand the c

25、orresponding research hypothesesshould be tested.These changes are as follows:1.Part II of the third edition,about core SEM techniques,is now organized accord-ing to phases of the analysis,starting with model specification,going on to consider-ation of its identification status,next to estimation,an

26、d then to the testing of hypotheses and model respecification(Chapters 58).In contrast,the second edition covered this material on a more technique-by-technique basis.I think that the new organization corresponds more closely to how researchers usually proceed with an SEM analysis.It should also giv

27、e students a better view of the“big picture”concerning major issues that apply in most applications of SEM.2.There are now exercises with suggested answers for all chapters that introduce prerequisite statistical and measurement concepts(Part I)and also for all chapters in Part II about core techniq

28、ues.These exercises give students additional opportunities for x Preface and Acknowledgmentslearning about SEM by responding to questions that test their concept knowledge.Some exercises also involve the analysis of structural equation models with actual data sets(i.e.,learning by doing).All of thes

29、e features also support self-study of SEM;that is,they should help readers who wish to learn about SEM but are not participating in a formal course or seminar.3.Website support for this edition is even stronger than that of the second edition.For example,readers can freely download for every detaile

30、d example in Part II all syntax,data,and output files for each of three widely used SEM computer tools:EQS,LISREL,and Mplus.This allows readers to reproduce the analysis on their own computer using the corresponding computer tool.Even if the reader uses a different computer tool for SEM,all of these

31、 files can be opened with a standard text editor,such as Windows Note-pad.That is,the reader does not need to have EQS,LISREL,or Mplus installed on his or her computer in order to view the contents of these files.And for readers who already use one of three computer tools for SEM(e.g.,LISREL),it can

32、 be educational to view the results of the same analysis generated by a different computer tool(e.g.,Mplus).Other resources for readers may be found on the books website(described in Chapter 1),the address of which is presented on page 3.4.The chapter on hypothesis testing in SEM(Chapter 8)reflects

33、some of the most recent thinking in this area that is described by several different authors in a special issue on SEM in the journal Personality and Individual Differences(Vernon&Eysenck,2007).Briefly,there is a general consensus that(a)standard practices for evaluating models in SEM have been lax

34、and,consequently,(b)researchers need to take a more rigorous,skeptical,and disciplined approach to hypothesis testing.How to do so is a major theme of Chapter 8 and indeed of the whole book.5.There is more coverage in this edition of two advanced topics in SEM:the estima-tion of interactive effects

35、of observed or latent variables and multilevel analysis(Chapter 12).Many developments have taken place recently in each of these areas,and more and more researchers are estimating models in which these types of effects are represented.Accordingly,the chapter on how to fool yourself with SEM(Chapter

36、13)is now expanded to include the failure to consider these types of effects,among other more prosaic ways to become irrational with SEM.6.Several chapters feature topic boxes about concepts,techniques,or specialized issues in the conduct of SEM.These boxes offer relatively short summaries of topics

37、 that complement or elaborate on the presentation in the main text.More advanced topics are covered in chapter appendices,which allows readers of various skill levels to get more out of the book.C.Deborah Laughton,Publisher,Methodology and Statistics,at The Guilford Press,has a special knack for giv

38、ing me exactly the type of feedback I need at precisely the right moment in the writing process.She collected reviews of the second edition and drafts for the third edition from a variety of scholars with differing backgrounds and levels of experience,from those just learning about SEM to renowned p

39、rofessors whose work is very widely known in their respective fields.C.Deborah sent these reviews Preface and Acknowledgments xito me without identifying their authors,and the content of the reviews was extremely helpful in the planning and writing of this edition.C.Deborah,thanks again for all your

40、 work and support.The names of the reviewers were revealed to me only after the writing was done,and their original comments were not associated with their names.A big thanks to all the persons listed next(in alphabetical order)who put in a lot of time and effort to communicate their thoughts about

41、the book in various stages of its writing;their comments and suggestions were invaluable:Alan C.Acock,Department of Human Development,Oregon State UniversityNoel A.Card,John and Doris Norton School of Family and Consumer Sciences,Division of Family Studies and Human Development,University of Arizona

42、David F.Gillespie,Department of Social Work,Washington University in St.LouisDebbie Hahs-Vaughn,College of Education,Department of Educational Research,Technology,and Leadership,University of Central FloridaLance Holbert,Department of Communications,Ohio State UniversityJacob Marszalek,School of Edu

43、cation,Research and Psychology,University of MissouriKansas CityRichard A.Posthuma,College of Business Administration,University of Texas at El PasoJames Schreiber,School of Education,Department of Foundations and Leadership,Duquesne UniversityGreg Welch,School of Education,Department of Psychology

44、and Research in Edu-cation,University of KansasCraig Wells,School of Education,Department of Educational Policy,Research,and Administration,University of Massachusetts at AmherstDuan Zhang,Morgridge College of Education,Quantitative Research Methods,University of DenverIt was a pleasure to work with

45、 the Methodology in the Social Sciences Series Editor at Guilford,Todd D.Little,in putting together the final version of this book.His com-ments were very helpful,and it was a pleasure to meet Todd when he visited Concordia University in Montral in November 2009.Betty Pessagno served as the copyedit

46、or for the original manuscript,and her work and suggested changes improved the clarity of the presentation.I also appreciate the efforts of the Guilford production editor,Wil-liam Meyer,in preparing the final version of this book.I asked Lesley Hayduk of the Department of Sociology at the University

47、 of Alberta to review a draft of Chapter 8 about hypothesis testing in SEM.Les has long advocated for a more rigorous approach to test-ing in SEM,and the rest of the field is catching up to this viewpoint.I was hoping that Less comments would give the final version of Chapter 8 more backbone,and I w

48、as not disappointed.Thanks,Les,for saying the kinds of things I needed to hear about this crucial topic.The most recent versions of computer tools for SEM were generously provided for me by Multivariate Software(EQS),Muthn and Muthn(Mplus),and Scientific Soft-xii Preface and Acknowledgmentsware Inte

49、rnational(LISREL).In particular,I wish to thank Linda Muthn and Peter Bentler for their comments on earlier drafts of descriptions of,respectively,Mplus and EQS.And once again,my heartfelt thanks to my wife,Joanna,and children,Julia and Luke,for all their love and support while writing this book.Rex

50、 B.Kline Montral rex.klineconcordia.ca xiii ContentsPart I.ConCePts and tools 1Introduction 3The Books Website 3Pedagogical Approach 4Getting Ready to Learn about SEM 5Characteristics of SEM 7Widespread Enthusiasm,but with a Cautionary Tale 13Family History and a Reminder about Context 15Extended La