book

A Step-by-Step Approach to Using SAS for Factor Analysis and Structural Equation Modeling, Second Edition, 2nd Edition

by Psych. D. Psych. Norm O'Rourke, Ph. D. Larry Hatcher

March 2013

Intermediate to advanced

444 pages

13h 49m

English

SAS Institute

Read now

Unlock full access

About This Book
Acknowledgments from the First Edition
Chapter 1: Principal Component Analysis
Introduction: The Basics of Principal Component AnalysisA Variable Reduction ProcedureAn Illustration of Variable RedundancyWhat Is a Principal Component?Principal Component Analysis Is Not Factor AnalysisExample: Analysis of the Prosocial Orientation InventoryPreparing a Multiple-Item InstrumentNumber of Items per ComponentMinimal Sample Size RequirementsSAS Program and OutputWriting the SAS ProgramResults from the OutputSteps in Conducting Principal Component AnalysisStep 1: Initial Extraction of the ComponentsStep 2: Determining the Number of “Meaningful” Components to RetainStep 3: Rotation to a Final SolutionStep 4: Interpreting the Rotated SolutionStep 5: Creating Factor Scores or Factor-Based ScoresStep 6: Summarizing the Results in a TableStep 7: Preparing a Formal Description of the Results for a PaperAn Example with Three Retained ComponentsThe QuestionnaireWriting the ProgramResults of the Initial AnalysisResults of the Second AnalysisConclusionAppendix: Assumptions Underlying Principal Component AnalysisReferences
Chapter 2: Exploratory Factor Analysis
Introduction: When Is Exploratory Factor Analysis Appropriate?Introduction to the Common Factor ModelExample: Investment Model QuestionnaireThe Common Factor Model: Basic ConceptsExploratory Factor Analysis versus Principal Component AnalysisHow Factor Analysis Differs from Principal Component AnalysisHow Factor Analysis Is Similar to Principal Component AnalysisPreparing and Administering the Investment Model QuestionnaireWriting the Questionnaire ItemsNumber of Items per FactorMinimal Sample Size RequirementsSAS Program and Exploratory Factor Analysis ResultsWriting the SAS ProgramResults from the OutputSteps in Conducting Exploratory Factor AnalysisStep 1: Initial Extraction of the FactorsStep 2: Determining the Number of “Meaningful” Factors to RetainStep 3: Rotation to a Final SolutionStep 4: Interpreting the Rotated SolutionStep 5: Creating Factor Scores or Factor-Based ScoresStep 6: Summarizing the Results in a TableStep 7: Preparing a Formal Description of the Results for a PaperA More Complex Example: The Job Search Skills QuestionnaireThe SAS ProgramDetermining the Number of Factors to RetainA Two-Factor SolutionA Four-Factor SolutionConclusionAppendix: Assumptions Underlying Exploratory Factor AnalysisReferences
Chapter 3: Assessing Scale Reliability with Coefficient Alpha
Introduction: The Basics of Response ReliabilityExample of a Summated Rating ScaleTrue Scores and Measurement ErrorUnderlying Constructs versus Observed VariablesReliability DefinedTest-Retest ReliabilityInternal ConsistencyReliability as a Property of Responses to ScalesCoefficient AlphaFormulaWhen Will Coefficient Alpha Be High?Assessing Coefficient Alpha with PROC CORRGeneral FormA 4-Item ScaleHow Large Must a Reliability Coefficient Be to Be Considered Acceptable?A 3-Item ScaleSummarizing the ResultsSummarizing the Results in a TablePreparing a Formal Description of the Results for a PaperConclusionNotesReferences
Chapter 4: Path Analysis
Introduction: The Basics of Path AnalysisSome Simple Path DiagramsImportant Terms Used in Path AnalysisWhy Perform Path Analysis with PROC CALIS versus PROC REG?Necessary Conditions for Path AnalysisOverview of the AnalysisSample Size Requirements for Path AnalysisStatistical Power and Sample SizeEffect SizesEstimating Sample Size RequirementsExample 1: A Path-Analytic Investigation of the Investment ModelOverview of the Rules for Performing Path AnalysisPreparing the Program FigureStep 1: Drawing the Basic ModelStep 2: Assigning Short Variable Names to Manifest VariablesStep 3: Identifying Covariances among Exogenous VariablesStep 4: Identifying Residual Terms for Endogenous VariablesStep 5: Identifying Variances to Be EstimatedStep 6: Identifying Covariances to Be EstimatedStep 7: Identifying the Path Coefficients to Be EstimatedStep 8: Verifying that the Model Is OveridentifiedPreparing the SAS ProgramOverviewThe DATA Input StepThe PROC CALIS StatementThe LINEQS StatementThe VARIANCE StatementThe COV StatementThe VAR StatementInterpreting the Results of the AnalysisMaking Sure That the SAS Output File “Looks Right”Assessing the Fit between Model and DataCharacteristics of an Ideal FitModifying the ModelProblems Associated with Model ModificationRecommendations for Modifying ModelsModifying the Present ModelPreparing a Formal Description of the Analysis and Results for a PaperPreparing Figures and TablesPreparing TextExample 2: Path Analysis of a Model Predicting Victim Reactions to Sexual HarassmentComparing Alternative ModelsThe SAS ProgramResults of the AnalysisConclusion: How to Learn More about Path AnalysisNoteReferences
Chapter 5: Developing Measurement Models with Confirmatory Factor Analysis
Introduction: A Two-Step Approach to Analyses with Latent VariablesA Model of the Determinants of Work PerformanceThe Manifest Variable ModelThe Latent Variable ModelBasic Concepts in Latent Variable AnalysesLatent Variables versus Manifest VariablesChoosing Indicator VariablesThe Confirmatory Factor Analytic ApproachThe Measurement Model versus the Structural ModelAdvantages of Covariance Structure AnalysesNecessary Conditions for Confirmatory Factor AnalysisSample Size Requirements for Confirmatory Factor Analysis and Structural Equation ModelingCalculation of Statistical PowerCalculation of Sample Size RequirementsExample: The Investment ModelThe Theoretical ModelResearch Method and Overview of the AnalysisTesting the Fit of the Measurement Model from the Investment Model StudyPreparing the Program FigurePreparing the SAS ProgramMaking Sure That the SAS Log and Output Files “Look Right”Assessing the Fit between Model and DataModifying the Measurement ModelEstimating the Revised Measurement ModelAssessing Reliability and Validity of Constructs and IndicatorsCharacteristics of an “Ideal Fit” for the Measurement ModelConclusion: On to Covariance Analyses with Latent Variables?References
Chapter 6: Structural Equation Modeling
Basic Concepts in Covariance Analyses with Latent VariablesAnalysis with Manifest Variables versus Latent VariablesA Two-Step Approach to Structural Equation ModelingThe Importance of Reading Chapters 4 and 5 FirstTesting the Fit of the Theoretical Model from the Investment Model StudyThe Rules for Structural Equation ModelingPreparing the Program FigurePreparing the SAS ProgramInterpreting the Results of the AnalysisCharacteristics of an “Ideal Fit” for the Theoretical ModelUsing Modification Indices to Modify the Present ModelPreparing a Formal Description of Results for a PaperFigures and TablesPreparing Text for the Results Section of the PaperAdditional Example: A SEM Predicting Victim Reactions to Sexual HarassmentConclusion: To Learn More about Latent Variable ModelsReferences
Appendix A.1: Introduction to SAS Programs, SAS Logs, and SAS Output
What Is SAS?Three Types of SAS FilesThe SAS ProgramThe SAS LogThe SAS Output FileSAS Customer SupportConclusionReference
Appendix A.2: Data Input
Introduction: Inputting Questionnaire Data versus Other Types of DataEntering Data: An Illustrative ExampleInputting Data Using the DATALINES StatementAdditional GuidelinesInputting String Variables with the Same Prefix and Different Numeric SuffixesInputting Character VariablesUsing Multiple Lines of Data for Each ParticipantCreating Decimal Places for Numeric VariablesInputting “Check All That Apply” Questions as Multiple VariablesInputting a Correlation or Covariance MatrixInputting a Correlation MatrixInputting a Covariance MatrixInputting Data Using the INFILE Statement Rather Than the DATALINES StatementConclusionReferences

Appendix A.3: Working with Variables and Observations in SAS Datasets
Introduction: Manipulating, Subsetting, Concatenating, and Merging DataPlacement of Data-Manipulation and Data-Subsetting StatementsImmediately Following the INPUT StatementImmediately after Creating a New DatasetThe INFILE Statement versus the DATALINES StatementData ManipulationCreating Duplicate Variables with New Variable NamesDuplicating Variables versus Renaming VariablesCreating New Variables from Existing VariablesPriority of Operators in Compound ExpressionsRecoding Reversed VariablesUsing IF-THEN Control StatementsUsing ELSE StatementsUsing the Conditional Statements AND and ORWorking with Character VariablesUsing the IN OperatorData SubsettingUsing a Simple Subsetting StatementUsing Comparison OperatorsEliminating Observations with Missing Data for Some VariablesA More Comprehensive ExampleConcatenating and Merging DatasetsConcatenating DatasetsMerging DatasetsConclusionReference
Appendix A.4: Exploring Data with PROC MEANS, PROC FREQ, PROC PRINT, and PROC UNIVARIATE
Introduction: Why Perform Simple Descriptive Analyses?Example: An Abridged Volunteerism SurveyComputing Descriptive Statistics with PROC MEANSThe PROC MEANS StatementThe VAR StatementReviewing the OutputCreating Frequency Tables with PROC FREQThe PROC FREQ and TABLES StatementsReviewing the OutputPrinting Raw Data with PROC PRINTTesting for Normality with PROC UNIVARIATEWhy Test for Normality?Departures from NormalityGeneral Form for PROC UNIVARIATEResults for an Approximately Normal DistributionResults for a Distribution with an OutlierUnderstanding the Stem-Leaf PlotResults for Distributions Demonstrating SkewnessConclusionReference
Appendix A.5: Preparing Scattergrams and Computing Correlations
Introduction: When Are Pearson Correlations Appropriate?Interpreting the CoefficientLinear versus Nonlinear RelationshipsProducing Scattergrams with PROC PLOTComputing Pearson Correlations with PROC CORRComputing a Single Correlation CoefficientDetermining Sample SizeComputing All Possible Correlations for a Set of VariablesComputing Correlations between Subsets of VariablesOptions Used with PROC CORRAppendix: Assumptions Underlying the Pearson Correlation CoefficientReferences
Appendix B: Datasets
Dataset from Chapter 1: Principal Component AnalysisDatasets from Chapter 2: Exploratory Factor AnalysisDataset from Chapter 3: Assessing Scale Reliability with Coefficient Alpha
Appendix C: Critical Values for the Chi-Square Distribution
Index

Content preview from A Step-by-Step Approach to Using SAS for Factor Analysis and Structural Equation Modeling, Second Edition, 2nd Edition

Appendix A.1: Introduction to SAS Programs, SAS Logs, and SAS Output

What Is SAS?

Three Types of SAS Files

What Is SAS?

SAS is a modular, integrated, and hardware-independent system of statistical software. It is a particularly powerful tool for social scientists because it allows us to easily perform a myriad of statistical analyses that may be required in the course of conducting research. SAS is sufficiently comprehensive to perform the most sophisticated multivariate analyses (i.e., multiple dependent variables), but is also easy to use so that undergraduates can perform basic analyses after only a short period of instruction.

In a sense, SAS may be ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Analysis of Clinical Trials Using SAS, 2nd Edition

Publisher Resources

ISBN: 9781599942308

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

A Step-by-Step Approach to Using SAS for Factor Analysis and Structural Equation Modeling, Second Edition, 2nd Edition

by Psych. D. Psych. Norm O'Rourke, Ph. D. Larry Hatcher