The Link to the Book: https://hsiutingyu.github.io/analyzing-change/

Analyzing Change

Longitudinal, Intensive Longitudinal, and Dynamic Data Analysis for the Social Sciences 

Introduction and Description

Repeated-measures data now span an enormous range of designs, from two-wave pre-post studies and multi-wave panel surveys to daily diaries, ecological momentary assessment, ambulatory physiology, and passive sensing, and the methodological literature addressing them has fragmented accordingly. A researcher has typically needed to consult one source for growth models, another for diary and experience-sampling designs, another for longitudinal structural equation modeling, and scattered articles and tutorials for dynamic, idiographic, and network methods, with no single volume carrying a reader from a paired-samples comparison to a regime-switching state-space model under one coherent conceptual thread, one consistent notation, and one set of shared datasets and working code. Analyzing Change is written to close that gap for a psychology and social-science audience.

The book is organized around a single recurring conceptual axis: the decomposition of between-person differences from within-person change and dynamics, together with the closely related questions of timescale, measurement, and the data-generating process a given model assumes. Every method is introduced by asking what question about change it answers, what it assumes about the underlying process, and what information it borrows across persons. This thread is what connects repeated-measures ANOVA, which conflates the between- and within-person levels, to multilevel and latent growth models, which separate them; the cross-lagged panel model to its random-intercept correction, where the distinction becomes decisive for causal interpretation; and the dynamic structural equation, multilevel vector-autoregression, and network families, where within-person dynamics themselves become the target of inference rather than a nuisance to be averaged away.

Thirty-six chapters, organized into eight parts, carry this thread from foundations through classical methods, the multilevel and structural-equation-modeling frameworks, intensive longitudinal and dynamic models, and a set of special topics, before closing with four applications chapters and a capstone chapter that consolidates method selection, reproducibility, and reporting into a single working system. Four appendices provide reference material: an R primer and package registry, the mathematical and statistical foundations the main text presumes, a catalogue of every dataset the book uses together with its known-truth generating parameters, and a software crosswalk across R, Mplus, SAS, Stata, and Python. Nearly every worked example in the book is built on a simulated dataset whose true data-generating parameters are known and stored with the data, so that a method is shown to recover the truth that produced it rather than merely to produce a plausible-looking number, a design choice explained at length in Appendix C.

The intended reader is a graduate student or researcher in psychology or a neighboring social or behavioral science, such as education, family science, organizational behavior, communication, or public health, who has had one graduate course in regression or ANOVA and is comfortable with basic probability but need not already know matrix algebra, likelihood theory, Bayesian estimation, or R. Advanced undergraduates in quantitative courses and applied statisticians entering psychology are a secondary audience. R is the backbone of every worked example, with Mplus sidebars where Mplus is the field standard, and Appendix D gives a brief crosswalk for readers working primarily in SAS, Stata, or Python.

Table of Contents

Part I. Foundations: Change, Designs, and Data (Chapters 1-6)

Chapter 1. Studying Change: Why Within-Person Processes Require Their Own Methods

Chapter 2. Longitudinal Research Designs: From Panel Surveys to EMA, Diary, and Ambulatory Assessment

Chapter 3. Measurement Across Time and Timescales

Chapter 4. Planning Longitudinal Studies: Power, Precision, and Design Trade-Offs

Chapter 5. Preparing and Managing Repeated Measures Data

Chapter 6. Missing Data and Attrition in Longitudinal Research

Part II. Describing and Visualizing Change (Chapters 7-9)

Chapter 7. Variance Decomposition and Descriptive Statistics for Repeated Measures

Chapter 8. Visualizing Change I: Trajectories, Panels, and Comparative Displays

Chapter 9. Visualizing Change II: Intensive Time Series, Dynamics, and Interactive Graphics

Part III. Classical Methods: The ANOVA and Regression Traditions (Chapters 10-12)

Chapter 10. Analyzing Two Occasions: Paired Comparisons, Change Scores, and Reliable Change

Chapter 11. Repeated Measures ANOVA and MANOVA

Chapter 12. Marginal Models and Generalized Estimating Equations

Part IV. The Multilevel Modeling Framework (Chapters 13-17)

Chapter 13. Linear Mixed-Effects Models: Foundations

Chapter 14. Growth Curve Modeling in the Multilevel Framework

Chapter 15. Generalized Linear Mixed Models for Categorical and Count Outcomes

Chapter 16. Modeling Within-Person Variability: Mixed-Effects Location-Scale Models

Chapter 17. Bayesian Estimation for Longitudinal Models

Part V. The Structural Equation Modeling Framework (Chapters 18-22)

Chapter 18. Longitudinal Factor Analysis and Measurement Invariance Across Time

Chapter 19. Latent Growth Curve Models

Chapter 20. Latent Change Score and Advanced Growth Models

Chapter 21. Cross-Lagged Panel Designs: CLPM, RI-CLPM, and Causal Inference with Panel Data

Chapter 22. Growth Mixture, Latent Class Growth, and Latent Transition Models

Part VI. Intensive Longitudinal Data: Dynamic and Idiographic Models (Chapters 23-28)

Chapter 23. Multilevel Models for Diary and Intensive Longitudinal Data

Chapter 24. Time Series Analysis for Single Subjects

Chapter 25. Vector Autoregression, Multilevel VAR, and Dynamic SEM

Chapter 26. Dynamic Factor Analysis and State-Space Models

Chapter 27. Continuous-Time Models

Chapter 28. Network Approaches to Intensive Longitudinal Data

Part VII. Special Topics and Extensions (Chapters 29-32)

Chapter 29. Survival and Event-History Analysis

Chapter 30. Flexible Curves: GAMMs, TVEM, and Functional Data Approaches

Chapter 31. Machine Learning and Data-Driven Exploration for Longitudinal Data

Chapter 32. Nonlinear Dynamics: Regime Switching, Differential Equations, and Early Warning Signals

Part VIII. Applications and Workflow (Chapters 33-36)

Chapter 33. Applications in Clinical Psychology and Mental Health

Chapter 34. Applications in Developmental and Educational Psychology

Chapter 35. Applications in Social, Personality, Organizational, and Health Psychology

Chapter 36. The Complete Workflow: Method Selection, Reproducibility, and Reporting

Appendices

Appendix A. R Primer for Longitudinal Analysis

Appendix B. Mathematical and Statistical Foundations

Appendix C. Book Datasets and Codebooks

Appendix D. Software Crosswalk: R, Mplus, SAS, Stata, and Python