Methods of Meta-Analysis

Correcting Error and Bias in Research Findings

AvJohn E. Hunter,Frank L. Schmidt

Häftad, Engelska, 2004

3 244 kr

Beställningsvara. Skickas inom 3-6 vardagar. Fri frakt över 249 kr.

Fler format och utgåvor

E-bok

1 499 kr

Beskrivning

Click 'Additional Materials' for downloadable sample chapter

"Clearly written and compellingly argued, this book explains the procedures and benefits of correcting for measurement error and range restriction and details the methodological developments in meta-analysis over the last decade. No one should consider conducting a meta-analysis without first reading this book. It is essential reading for all scientists."
—Michael A. McDaniel, Virginia Commonwealth University

"A book that will certainly appeal not only to the students, but will also be a great reference source for the technically sophisticated professional. The breadth and depth of the coverage, not to mention the novelty and clarity of writing, makes this book a classic in the field. It covers (and at times introduces) many novel issues that will be in the forefront for some years to come—as such a must read for all meta-analysts."
—Vish C. Viswesvaran, Ph.D., Director of I/O Program, Florida International University

Meta-analysis is arguably the most important methodological innovation in the social and behavioral sciences in the last 25 years. Developed to offer researchers an informative account of which methods are most useful in integrating research findings across studies, this book will enable the reader to apply, as well as understand, meta-analytic methods. Rather than taking an encyclopedic approach, the authors have focused on carefully developing those techniques that are most applicable to social science research, and have given a general conceptual description of more complex and rarely-used techniques. Fully revised and updated, Methods of Meta-Analysis, Second Edition is the most comprehensive text on meta-analysis available today.

New to the Second Edition:

* An evaluation of fixed versus random effects models for meta-analysis
* New methods for correcting for indirect range restriction in meta-analysis
* New developments in corrections for measurement error
* A discussion of a new Windows-based program package for applying the meta-analysis methods presented in the book
* A presentation of the theories of data underlying different approaches to meta-analysis

Produktinformation

Utgivningsdatum:2004-06-03
Mått:177 x 254 x 36 mm
Vikt:1 170 g
Format:Häftad
Språk:Engelska
Antal sidor:616
Upplaga:2
Förlag:SAGE Publications
ISBN:9781412904797

Utforska kategorier

Sociologi inom Samhälle och politik

Mer om författaren

John E. (Jack) Hunter (1939--2002) was a professor in the Department of Psychology at Michigan State University. He received his Ph.D. in quantitative psychology from the University of Illinois. Jack coauthored four books and authored or coauthored over 200 articles and book chapters on a wide variety of methodological topics, including confirmatory and exploratory factor analysis, measurement theory and methods, statistics, and research methods. He also published numerous research articles on such substantive topics as intelligence, attitude change, the relationship between attitudes and behavior, validity generalization, differential validity/selection fairness, and selection utility. Much of his research on attitudes was in the field of communications, and the American Communications Association named a research award in his honor. Professor Hunter received the Distinguished Scientific Award for Contributions to Applied Psychology from the American Psychological Association (APA) (jointly with Frank Schmidt) and the Distinguished Scientific Contributions Award from the Society for Industrial/Organizational Psychology (SIOP) (also jointly with Frank Schmidt). He was a Fellow of APA, APS, and SIOP, and was a past president of the Midwestern Society for Multivariate Experimental Psychology. For the story of Jack’s life, see Schmidt (2003). Frank L. Schmidt is the Gary F. Fethke Leadership Professor Emeritus in the Department of Management and Organization in the Tippie College of Business at the University of Iowa. He received his Ph.D. in industrial/organizational psychology from Purdue University and has been on the faculties of Michigan State and George Washington Universities. He has authored or coauthored seven books and nearly 200 articles and book chapters on measurement, statistics, research methods, individual differences, and personnel selection. He headed a research program in the U.S. Office of Personnel Management in Washington, D.C., for 11 years, during which time he published numerous research studies in personnel psychology, primarily with John Hunter. Their research on the generalizability of employment selection method validities led to the development of the meta-analysis methods presented in this book. Professor Schmidt has received the Distinguished Scientific Award for Contributions to Applied Psychology from the American Psychological Association (APA) (jointly with John Hunter) and the Distinguished Scientific Contributions Award from the Society for Industrial/Organizational Psychology (SIOP) (also jointly with John Hunter). He has also received the Ingram Olkin Award and the Frederick Mosteller Award, both for contributions to meta-analysis methodology; the Scientific Award for Applications of Psychology from the Association for Psychological Science (APS); the Gold Medal Lifetime Achievement award from the APA Foundation; the Distinguished Career Award for Contributions to Human Resources, and the Distinguished Career Achievement Award for Contributions to Research Methods, both from the Academy of Management. He is a Fellow of the APA, the Association for Psychological Science, and SIOP, and is past president of Division 5 (Measurement, Statistics, & Evaluation) of the APA.

Recensioner i media

“Clearly written and compellingly argued, this book explains the procedures and benefits of correcting for measurement error and range restriction and details the methodological developments in meta-analysis over the last decade. No one should consider conducting a meta-analysis without first reading this book. It is essential reading for all scientists.”

Innehållsförteckning

Preface to 2nd EditionPreface to 1st EditionAcknowledgements Introduction to Meta-AnalysisIntegration Research Findings Across StudiesGeneral problem and an exampleProblems with statistical significance testsIs statistical power the solution?Confidence intervalsMeta-analysisRole of meta-analysis in the behavioral and social sciencesRole of meta-analysis in theory developmentIncreasing use of meta-analysisMeta-analysis in industrial-organizational psychologyWider impact of meta-analysis on psychologyImpact of meta-analysis outside psychologyMeta-analysis and social policyMeta-analysis and theories of dataConclusionsStudy Artifacts and Their Impact on Study Outcomes Study ArtifactsSampling error, statistical power, and the interpretation of research literaturesWhen and how to cumulateUndercorrection for artifacts in the corrected standard deviationCoding study characteristics and capitalization on sampling error in moderator analysisA look ahead in the bookMeta-Analysis of CorrelationsMeta-Analysis of Correlations Corrected Individually for ArtifactsIntroduction and OverviewBare bones meta-analysis: Correcting for sampling error onlyArtifacts other than sampling errorMultiple simultaneous artifactsMeta-analysis of individually corrected correlationsA worked example: Indirect range restrictionSummary of meta-analysis correcting each correlation individuallyExercise 1: Bare bones meta-analysisExercise 2: Meta-analysis correcting each correlation individuallyMeta-Analysis of Correlations Using Artifact DistributionsFull artifact distribution meta-analysisAccuracy of corrections for artifactsMixed meta-analysis: Partial artifact information in individual studiesSummary of artifact distribution of meta-analysis of correlationsExercise: Artifact distribution meta-analysisTechnical Questions in Meta-Analysis of Correlationsr versus : Which should be used?r vs. regression slopes and intercepts in meta-analysisTechnical factors that cause overestimation of Fixed and random models in meta-analysisCredibility vs. confidence intervals in meta-analysisComputing confidence intervals in meta-analysisRange Restriction in meta-analysis: New technical analysisCriticisms of meta-analysis procedures for correlationsMeta-Analysis of Experimental Effects and Other Dichotomous Comparisons Treatment Effects: Experimental Artifacts and Their ImpactQuantification of the treatment effect: The d statistic and the point-biserial correlationSampling error in d values: IllustrationsError of measurement in the dependent variableError of measurement in the treatment variableVariation across studies in treatment strengthRange variation on the dependent variableDichotomization of the dependent variableImperfect construct validity in the dependent variableImperfect construct validity in the treatment variableBias in the effect size (d statistic)Recording, computational, and transcriptional errorsMultiple artifacts and correctionsMeta-Analysis Methods for d ValuesEffect size indices: d and rAn Alternative to d: Glass′ d Sampling error in the d statisticCumulation and correction of the variance for sampling errorAnalysis of moderator variablesCorrecting d values for measurement error in the dependent variableMeasurement error in the independent variable in experimentsOther artifacts and their effectsCorrecting for multiple artifactsSummary of meta-analysis of d valuesExercise: Meta-Analysis of d-ValuesTechnical Questions in Meta-Analysis of d ValuesAlternative experimental designsWithin-subjects experimental designsMeta-analysis and the within-subjects designStatistical power in the two designsThreats to internal and external validityBias in observed d valuesUse of multiple regression in moderation analysis of d valuesGeneral Issues in Meta-AnalysisGeneral Technical Issues in Meta-analysisFixed effects versus random effects models in meta-analysisSecond order sampling error: General principlesDetecting moderators not hypothesized a prioriSecond order meta-analysisLarge N studies and meta-analysisSecond order sampling error: Technical treatmentThe detection of moderator variables: SummaryHierarchical analysis of moderator variablesExercise: Second order meta-analysisCumulation of Findings Within StudiesFully replicated designsConceptual replicationsConceptual replications and confirmatory factor analysisConceptual replications: A alternative approachAnalysis of subgroupsSummaryMethods of Integrating Findings Across Studies and Related SoftwareThe traditional narrative procedureThe traditional voting methodCumulation of p-values across studiesStatistically correct vote counting proceduresMeta-analysis of research studiesUnresolved problems in meta-analysisSummary of methods of integrating studiesComputer programs for meta-analysisLocating, Evaluating, and Coding StudiesConducting a thorough literature searchWhat to do about studies with "methodological weaknesses"Coding studies in meta-analysisWhat to include in the meta-analysis reportInformation needed in reports of primary studiesAppendix: Coding sheet for validity studiesAvailability and Source Bias in Meta-AnalysisSome evidence on biasEffects of methodological quality on mean effect sizes from different sourcesMultiple hypotheses and other considerations in availability biasMethods for detecting availability biasMethods for correcting for availability biasSummary of Psychometric Meta-AnalysisMeta-analysis methods and theories of dataWhat is the ultimate purpose of meta-analysis?Appendix: Windows Based Meta-Analysis Software PackageReferencesAuthor IndexSubject IndexAbout the Authors