Psychometric Methods

Larry R. Price, PhD, is Professor of Biostatistics in the Department of Psychiatry and Behavioral Sciences, University of Texas Health Science Center at San Antonio, Long School of Medicine. Previously, he served as a faculty statistician and psychometrician at Emory University School of Medicine (Department of Psychiatry and Behavioral Sciences), as Professor of Psychometrics and Statistics at Texas State University, and at the Psychological Corporation (now part of Pearson's Clinical Assessment Group). Dr. Price is a Fellow of the American Psychological Association, Division 5 (Evaluation, Measurement, and Statistics), and an Accredited Professional Statistician of the American Statistical Association.

“I like the book and it meshes well with what I plan to do in my course. I particularly like the generalizability theory, norms and test equating, scaling, and validation process chapters. It is very easy reading--I am planning to use the book next spring."--R. J. de Ayala, PhD, Chair of Educational Psychology, University of Nebraska–Lincoln"I encourage all psychologists and educators to read this marvelous book. I learned a lot from reading it. The key terms are very useful, as are the chapter summaries. Readers of all levels will find material relevant to them, including SPSS code and GfGc datasets on intelligence that will be quite useful in trying out the ideas. I give this book my highest recommendation and think it will be a great classroom text."--John J. McArdle, PhD, Department of Psychology, University of Southern California"This book is both comprehensive and accessible, laying the foundation for all the requisite skills needed to be both a successful consumer and producer of psychometrics. Scholars who are unfamiliar with measurement could easily teach themselves from this text, becoming quite proficient at psychometrics. There is excellent integration of quantitative statistics throughout, so that readers will be able not only to understand the psychometric concepts, but also to apply their knowledge. This is a useful text for a graduate-level Psychometric Methods or Measurement class."--Debbie L. Hahs-Vaughn, PhD, Department of Educational and Human Sciences, University of Central Florida"An encyclopedia of psychometric issues--a real 'must have' for anyone teaching Tests and Measurement or Research Methods, or directing student research projects. The book's high level of detail makes it invaluable for any professional who works with, creates, or analyzes psychometric material. The use of intelligence testing data throughout the chapters helps bring cohesiveness."--John Wallace, PhD, Department of Psychological Science, Ball State University "With vast expertise in psychometric instrument development, statistical applications, and research, Price has produced a theoretically informed, practical volume. Professionals in health-related fields will find this book extremely valuable for guidance in the development of rigorous instruments, such as patient-reported outcome measures. Featuring examples using a range of software, this text is ideal for graduate courses on measurement in schools of medicine, public health, nursing, or health professions."--Byron J. Gajewski, PhD, Department of Biostatistics, University of Kansas Medical Center-This book is suitable for emerging assessment professionals and practitioners who are interested in learning psychometrics but with little knowledge in statistics. It provides not only a theoretical foundation to the topics but also worked examples to highlight their practical applications. The syntax, output, and interpretations based on software programs like SPSS and BILOG-MG will help readers to bridge theory and methods with hands-on examples. This book would be a convenient toolbox for applied researchers who would like to conduct psychometric analyses, and it would also serve as a handbook for graduate students who study measurement and psychometrics.--Psychometrika, 3/29/2019ƒƒThis excellent book explores the basic concepts of psychometric knowledge and practice….This would be a great addition to the libraries of graduate students and researchers.--Doody's Review Service, 3/17/2017

• 1. Introduction1.1 Psychological Measurement and Tests1.2 Tests and Samples of Behavior1.3 Types of Tests1.4 Origin of Psychometrics1.5 Definition of Measurement1.6 Measuring Behavior1.7 Psychometrics and Its Importance to Research and Practice1.8 Organization of This BookKey Terms and Definitions2. Measurement and Statistical Concepts2.1 Introduction2.2 Numbers and Measurement2.3 Properties of Measurement in Relation to Numbers2.4 Levels of Measurement2.5 Contemporary View on the Levels of Measurement and Scaling2.6 Statistical Foundations for Psychometrics2.7 Variables, Frequency Distributions, and Scores2.8 Summation or Sigma Notation2.9 Shape, Central Tendency, and Variability of Score Distributions2.10 Correlation, Covariance, and Regression2.11 SummaryKey Terms and Definitions3. Criterion, Content, and Construct Validity3.1 Introduction3.2 Criterion Validity3.3 Essential Elements of a High-Quality Criterion3.4 Statistical Estimation of Criterion Validity3.5 Correction for Attenuation3.6 Limitations to Using the Correction for Attenuation3.7 Estimating Criterion Validity with Multiple Predictors: Partial Correlation3.8 Estimating Criterion Validity with Multiple Predictors: Higher-Order Partial Correlation3.9 Coefficient of Multiple Determination and Multiple Correlation3.10 Estimating Criterion Validity with More Than One Predictor: Multiple Linear Regression3.11 Regression Analysis for Estimating Criterion Validity: Development of the Regression Equation3.12 Unstandardized Regression Equation for Multiple Regression3.13 Testing the Regression Equation for Significance3.14 Partial Regression Slopes3.15 Standardized Regression Equation3.16 Predictive Accuracy of a Regression Analysis3.17 Predictor Subset Selection in Regression3.18 SummaryKey Terms and Definitions4. Statistical Aspects of the Validation Process4.1 Techniques for Classification and Selection4.2 Discriminant Analysis4.3 Multiple-Group Discriminant Analysis4.4 Logistic Regression4.5 Logistic Multiple Discriminant Analysis: Multinomial Logistic Regression4.6 Model Fit in Logistic Regression4.7 Content Validity4.8 Limitations of the Content Validity Model4.9 Construct Validity4.10 Establishing Evidence of Construct Validity4.11 Correlational Evidence of Construct Validity4.12 Group Differentiation Studies of Construct Validity4.13 Factor Analysis and Construct Validity4.14 Multitrait–Multimethod Studies4.15 Generalizability Theory and Construct Validity4.16 Summary and ConclusionsKey Terms and Definitions5. Scaling5.1 Introduction5.2 A Brief History of Scaling5.3 Psychophysical versus Psychological Scaling5.4 Why Scaling Models Are Important5.5 Types of Scaling Models5.6 Stimulus-Centered Scaling5.7 Thurstone’s Law of Comparative Judgment5.8 Response-Centered Scaling5.9 Scaling Models Involving Order5.10 Guttman Scaling5.11 The Unfolding Technique5.12 Subject-Centered Scaling5.13 Data Organization and Missing Data5.14 Incomplete and Missing Data5.15 Summary and ConclusionsKey Terms and Definitions6. Test Development6.1 Introduction6.2 Guidelines for Test and Instrument Development6.3 Item Analysis6.4 Item Difficulty6.5 Item Discrimination6.6 Point–Biserial Correlation6.7 Biserial Correlation6.8 Phi Coefficient6.9 Tetrachoric Correlation6.10 Item Reliability and Validity6.11 Standard Setting6.12 Standard-Setting Approaches6.13 The Nedelsky Method6.14 The Ebel Method6.15 The Angoff Method and Modifications6.16 The Bookmark Method6.17 Summary and ConclusionsKey Terms and Definitions7. Reliability7.1 Introduction7.2 Conceptual Overview7.3 The True Score Model7.4 Probability Theory, True Score Model, and Random Variables7.5 Properties and Assumptions of the True Score Model7.6 True Score Equivalence, Essential True Score Equivalence, and Congeneric Tests7.7 Relationship between Observed and True Scores7.8 The Reliability Index and Its Relationship to the Reliability Coefficient7.9 Summarizing the Ways to Conceptualize Reliability7.10 Reliability of a Composite7.11 Coefficient of Reliability: Methods of Estimation Based on Two Occasions7.12 Methods Based on a Single Testing Occasion7.13 Estimating Coefficient Alpha: Computer Programs and Example Data7.14 Reliability of Composite Scores Based on Coefficient Alpha7.15 Reliability Estimation Using the Analysis of Variance Method7.16 Reliability of Difference Scores7.17 Application of the Reliability of Difference Scores7.18 Errors of Measurement and Confidence Intervals7.19 Standard Error of Measurement7.20 Standard Error of Prediction7.21 Summarizing and Reporting Reliability Information7.22 Summary and ConclusionsKey Terms and Definitions8. Generalizability Theory8.1 Introduction8.2 Purpose of Generalizability Theory8.3 Facets of Measurement and Universe Scores8.4 How Generalizability Theory Extends Classical Test Theory8.5 Generalizability Theory and Analysis of Variance8.6 General Steps in Conducting a Generalizability Theory Analysis8.7 Statistical Model for Generalizability Theory8.8 Design 1: Single-Facet Person by Item Analysis8.9 Proportion of Variance for the p x i Design8.10 Generalizability Coefficient and CTT Reliability8.11 Design 2: Single-Facet Crossed Design with Multiple Raters8.12 Design 3: Single-Facet Design with the Same Raters on Multiple Occasions8.13 Design 4: Single-Facet Nested Design with Multiple Raters8.14 Design 5: Single-Facet Design Multiple Raters Rating on Two Occasions8.15 Standard Errors of Measurement: Designs 1–58.16 Two-Facet Designs8.17 Summary and ConclusionsKey Terms and Definitions9. Factor Analysis9.1 Introduction9.2 Brief History9.3 Applied Example with GfGc Data9.4 Estimating Factors and Factor Loadings9.5 Factor Rotation9.6 Correlated Factors and Simple Structure9.7 The Factor Analysis Model, Communality, and Uniqueness9.8 Components, Eigenvalues, and Eigenvectors9.9 Distinction between Principal Components Analysis and Factor Analysis9.10 Confirmatory Factor Analysis9.11 Confirmatory Factor Analysis and Structural Equation Modeling9.12 Conducting Factor Analysis: Common Errors to Avoid9.13 Summary and ConclusionsKey Terms and Definitions10. Item Response Theory10.1 Introduction10.2 How IRT Differs from CTT10.3 Introduction to IRT10.4 Strong True Sco