I. Fundamentals 1. Introduction 1.1. A Scientist in Training 1.2. Questions of Whether, If, How, and When 1.3. Conditional Process Analysis 1.4. Correlation, Causality, and Statistical Modeling 1.5. Statistical and Conceptual Diagrams, and Antecedent and Consequent Variables 1.6. Statistical Software 1.7. Overview of This Book 1.8. Chapter Summary 2. Fundamentals of Linear Regression Analysis 2.1. Correlation and Prediction 2.2. The Simple Linear Regression Model 2.3. Alternative Explanations for Association 2.4. Multiple Linear Regression 2.5. Measures of Model Fit 2.6. Statistical Inference 2.7. Multicategorical Antecedent Variables 2.8. Assumptions for Interpretation and Statistical Inference 2.9. Chapter Summary II. Mediation Analysis 3. The Simple Mediation Model 3.1. The Simple Mediation Model 3.2. Estimation of the Direct, Indirect, and Total Effects of X 3.3. Example with Dichotomous X: The Influence of Presumed Media Influence 3.4. Statistical Inference 3.5. An Example with Continuous X: Economic Stress among Small-Business Owners 3.6. Chapter Summary 4. Causal Steps, Confounding, and Causal Order 4.1. What about Baron and Kenny? 4.2. Confounding and Causal Order 4.3. Effect Size 4.4. Statistical Power 4.5. Multiple Xs or Ys: Analyze Separately or Simultaneously? 4.6. Chapter Summary 5. More Than One Mediator 5.1. The Parallel Multiple Mediator Model 5.2. Example Using the Presumed Media Influence Study 5.3. Statistical Inference 5.4. The Serial Multiple Mediator Model 5.5. Models With Parallel and Serial Mediation Properties 5.6. Complementarity and Competition among Mediators 5.7. Chapter Summary 6. Mediation Analysis with a Multicategorical Antecedent X 6.1. Relative Total, Direct, and Indirect Effects 6.2. An Example: Sex Discrimination in the Workplace 6.3. Using a Different Group Coding System 6.4. Some Miscellaneous Issues 6.5. Chapter Summary III. Moderation Analysis 7. Fundamentals of Moderation Analysis 7.1. Conditional and Unconditional Effects 7.2. An Example: Climate Change Disasters and Humanitarianism 7.3. Visualizing Moderation 7.4. Probing an Interaction 7.5. The Difference between Testing for Moderation and Probing It 7.6. Artificial Categorization and Subgroups Analysis 7.7. Chapter Summary 8. Extending the Fundamentals of Moderation Analysis 8.1. Moderation with a Dichotomous Moderator 8.2. Interaction between Two Quantitative Variables 8.3. Hierarchical versus Simultaneous Entry 8.4. The Equivalence between Moderated Regression Analysis and a 2 x 2 Factorial Analysis of Variance 8.5. Chapter Summary 9. Some Myths and Further Extensions of Moderation Analysis 9.1. Truths and Myths about Mean Centering 9.2. The Estimation and Interpretation of Standardized Regression Coefficients in a Moderation Analysis 9.3. A Caution on Manual Centering and Standardization 9.4. More than One Moderator 9.5. Comparing Conditional Effects 9.6. Chapter Summary 10. Multicategorical Focal Antecedents and Moderators 10.1. Moderation of the Effect of a Multicategorical Antecedent Variable 10.2. An Example from the Sex Discrimination in the Work Place Study 10.3. Visualizing the Model 10.4. Probing the Interaction 10.5. When the Moderator is Multicategorical 10.6. Using a Different Coding System 10.7. Chapter Summary IV. Conditional Process Analysis 11. Fundamentals of Conditional Process Analysis 11.1. Examples of Conditional Process Models in the Literature 11.2. Conditional Direct and Indirect Effects 11.3. Example: Hiding Your Feelings from Your Work Team 11.4. Estimation of a Conditional Process Model using PROCESS 11.5. Quantifying and Visualizing (Conditional) Indirect and Direct Effects 11.6. Statistical Inference 11.7. Chapter Summary 12. Further Examples of Conditional Process Analysis 12.1. Revisiting the Disaster Framing Study 12.2. Moderation of the Direct and Indirect Effects in a Conditional Process Model 12.3. Statistical Inference 12.4. Mediated Moderation 12.5. Chapter Summary 13. Conditional Process Analysis with a Multicategorical Antecedent 13.1. Revisiting Sexual Discrimination in the Work Place 13.2. Looking at the Components of the Indirect Effect of X 13.3. Relative Conditional Indirect Effects 13.4. Testing and Probing Moderation of Mediation 13.5. Relative Conditional Direct Effects 13.6. Putting It All Together 13.7. Chapter Summary V. Miscellanea 14. Miscellaneous Topics and Some Frequently Asked Questions 14.1. A Strategy for Approaching a Conditional Process Analysis 14.2. How Do I Write about This? 14.3. Should I Use Structural Equation Modeling Instead of Regression Analysis? 14.4. The Pitfalls of Subgroups Analysis 14.5. Can a Variable Simultaneously Mediate and Moderate Another Variable's Effect? 14.6. Interaction between X and M in Mediation Analysis 14.7. Repeated Measures Designs 14.8. Dichotomous, Ordinal, Count, and Survival Outcomes 14.9. Chapter Summary Appendices Appendix A. Using PROCESS Appendix B. Constructing and Customizing Models in PROCESS Appendix C. Monte Carlo Confidence Intervals in SPSS and SAS References About the Author
Andrew F. Hayes, PhD, is Professor of Quantitative Psychology at The Ohio State University. His research and writing on data analysis has been published widely, and he is the author of Introduction to Mediation, Moderation, and Conditional Process Analysis, Second Edition, and Statistical Methods for Communication Science, as well as coauthor, with Richard B. Darlington, of Regression Analysis and Linear Models. Dr. Hayes teaches data analysis, primarily at the graduate level, and frequently conducts workshops on statistical analysis throughout the world. His website is www.afhayes.com.
"This second edition is a welcome addition to advanced regression books that can be used in doctoral courses in the social sciences or by social science researchers. Hayes maintains his usual level of clarity while adding coverage of such important topics as multicategorical variables for mediation, moderation, and conditional process models. Enhanced presentation of tabular materials, coupled with new plots, add to the reader's understanding of analyses. Incorporation of R syntax at points in the book is great, as many researchers turn to R for its open access and improved graphics capabilities. I loved the first edition for my first-year doctoral course, and will use the second edition in its place."--Ramona L. Paetzold, DBA, Department of Management, Texas A&M University "Since I began using the first edition of this text in my graduate statistics classes in 2014, the number of theses and dissertations that include mediation and/or moderation analysis in our department has increased dramatically. Valuable new material in the second edition includes 13 new models, including models with categorical variables and models with both parallel and serial mediation, as well as the recently developed index of moderated mediation. My copy of the first edition is filled with my annotations on the examples of PROCESS output--in the second edition, Hayes has provided useful annotations of his own. I highly recommend this book for statistics classes that include OLS mediation and moderation. It is also a terrific resource for researchers wishing to keep up with advances in moderation and mediation analysis."--Karl L. Wuensch, PhD, Department of Psychology, East Carolina University "This book provides clear instruction that is accessible to graduate students while also useful to seasoned researchers looking to expand their skills for more complex regression-based analyses. The second edition provides increased clarity in interpreting PROCESS output and documents PROCESS v3, which allows for great flexibility in analyzing models. Other useful developments in the second edition include chapters on multicategorical variables, incorporation of the index of moderated mediation, and the appendix of instructions on how to customize PROCESS for models not covered by the templates. Hayes's approach is cutting edge in both philosophy and pragmatics. I've used the first edition extensively as a course text as well as in my own research, and am excited to move to the second edition."--Elizabeth J. Kiel, PhD, Department of Psychology, Miami University "Using lucid prose and abundant, worked-through examples, Hayes walks readers through the promise and potential pitfalls of two of the most essential--yet convoluted--tasks in social science research. Novices will find this book to be a thorough, accessible description of ordinary least squares regression and a smart tutorial on mediation and moderation, but it is also much more. Any seasoned researcher who has slogged through the arcane computation and agonizing decision making related to the estimation and interpretation of direct and indirect effects, or the visualization and presentation of interactions, will find this volume (with the accompanying PROCESS macro) to be a veritable Swiss Army knife, and will return to it time and time again."--Jeffrey G. Parker, PhD, Department of Psychology, University of Alabama "This text is a wonderful combination of traditional mediation and moderation using regression and extensions into more complex variations. Coverage is clear and thorough--perfect for intermediate to advanced regression learners. Updates in the second edition include a new chapter with answers to many very important and common questions, which will be extremely helpful to learners. I can't wait to use this second edition with my students."--Jocelyn H. Bolin, PhD, Department of Educational Psychology, Ball State University