Practical multilevel modeling using R / Francis L. Huang, University of Missouri.

Format:

Book

Author/Creator:

Huang, Francis L., author.

Series:

Advanced quantitative techniques in the social sciences ; 15.

Advanced Quantitative Techniques in the Social Sciences ; 15

Language:

English

Subjects (All):

Multilevel models (Statistics).

R (Computer program language).

Physical Description:

1 online resource : illustrations

Edition:

1st ed.

Place of Publication:

Thousand Oaks : SAGE Publications, Inc., [2023]

Summary:

This book provides students with a step-by-step guide for running their own multilevel analyses. Detailed examples illustrate the conceptual and statistical issues that multilevel modeling addresses in a way that is clear and relevant to students in applied disciplines. Clearly annotated R syntax illustrates how multilevel modeling (MLM) can be used, and real-world examples show why and how modeling decisions can affect results. The accompanying website includes R code and the dataset used in the book.

Contents:

Cover

PRAISE FOR THIS BOOK

Title Page

Advanced Quantitative Techniques in the Social Sciences

Copyright Page

Dedication Page

BRIEF CONTENTS

DETAILED CONTENTS

Acknowledgments

Preface

About the Author

Chapter 1. Introduction

1.1 Why Bother With Multilevel Modeling?

1.2 Why Another MLM Book?

1.3 Using R for MLM

Chapter 2. The Unconditional Means Model

2.1 Understanding MLM Notation

2.2 Fitting an Unconditional/Null Model

2.2.1 Computing the Intraclass Correlation Coefficient

2.2.2 Understanding the ICC Further

2.3 Summary

Chapter 3. Adding Predictors to a Random Intercept Model

3.1 Adding a Level-1 Predictor

3.1.1 How Much Variance Is Explained at Level One?

3.2 Creating and Adding Level-2 Predictors

3.2.1 How Much Variance Is Explained at Level Two?

3.2.2 What About an Overall R2?

3.2.3 Adding Categorical Predictors at Level Two

3.2.3.1 Changing the Reference Group of a Factor

3.3 Revisiting the Need for Multilevel Models

3.3.1 Comparing MLM and OLS Regression Results

3.3.2 Can We Really Ignore Clustering If the ICC Is Low?

3.3.2.1 Design Effects

3.4 Summary

Chapter 4. Investigating Cross-Level Interactions and Random Slope Models

4.1 Testing for Cross-Level Interactions

4.1.1 Using a Likelihood Ratio Test (LRT) for Fixed Effects

4.2 Investigating the Presence of Random Slopes

4.2.1 Using a Modified LRT for Random Effects

4.3 Revisiting the Random Intercept

4.4 When Should Random Slopes Be Included?

4.5 Dealing With Modeling Issues

4.6 Summary

Chapter 5. Understanding Growth Models

5.1 Introduction

5.2 Exploring and Reshaping the Data

5.2.1 Plotting Using the lattice Package

5.2.1.1 Computing Means by Groups

5.3 Specifying the Multilevel Growth Model

5.3.1 The Unconditional Growth Model.

5.3.2 Adding a Random Slope

5.3.3 Adding a Person-Level Predictor

5.3.4 An Alternative Approach: Using Robust Standard Errors

5.4 Summary

Chapter 6. Centering in Multilevel Models

6.1 What Is Centering?

6.2 Types of Centering

6.3 Understanding Different Effects Related to Centering

6.3.1 The Total Effect

6.3.2 The Within-Group Effect

6.3.2.1 The Fixed Effects Approach

6.3.2.2 Including the Group Mean Approach

6.3.3 The Between-Group Effect

6.3.4 The Compositional/Contextual Effect

6.4 Respecifying the Models Using MLM

6.5 Centering Binary Variables

6.6 Which Type of Centering to Use?

6.7 Summary

Chapter 7. Multilevel Modeling Diagnostics

7.1 Why Conduct Regression Diagnostics?

7.2 Residual Diagnostics

7.2.1 Spotting Nonlinear Relationships

7.2.2 Detecting Outliers

7.2.3 Assessing Normality

7.2.4 Understanding Influential Data

7.2.5 Assessing Issues Related to Homoskedasticity

7.2.5.1 Using the H Statistic

7.2.5.2 Using Robust Standard Errors

7.3 Multicollinearity

7.4 Summary

Chapter 8. Multilevel Logistic Regression Models

8.1 Introduction

8.2 Fitting a Multilevel Logistic Regression Model

8.2.1 Understanding Our Data

8.2.2 Getting the ICC

8.2.3 Adding Predictors of Interest

8.2.4 Obtaining an R2 Measure

8.3 Dealing With Nonconvergence Issues

8.4 Beyond Binary Outcomes

8.5 Summary

Chapter 9. Modeling Data Structures With Three (or More) Levels

9.1 Specifying a Three-Level Model

9.1.1 Estimating the ICC

9.1.2 Specifying the Model of Interest

9.1.3 Alternative Model Syntax for Multiple Levels

9.2 What If a Level Is Ignored?

9.2.1 What Happens If the Intermediate Level Is Ignored?

9.2.2 What Happens If the Higher Level Is Ignored?

9.2.3 Comparison of Output If a Level Is Ignored.

9.3 Including Random Slopes in a Three-Level Model

9.4 Do You Really Need a Three-Level Model?

9.5 Summary

Chapter 10. Missing Data in Multilevel Models

10.1 Introduction

10.1.1 Types of Missing Data

10.1.2 How Much and Which Data Are Missing?

10.2 Inspecting Our Data

10.3 From Imputation to Pooling Results

10.3.1 Getting Ready to Impute

10.3.2 Imputing the Data

10.3.3 Analyzing the Imputed Datasets

10.3.4 Pooling Results

10.3.5 Checking for Convergence

10.4 Other Options for Dealing With Missing Data

10.5 Summary

Chapter 11. Basic Power Analyses for Multilevel Models

11.1 Why Conduct a Power Analysis?

11.1.1 Approaches to Power Analyses

11.2 Elements Needed for a Power Analysis

11.2.1 The Significance Level

11.2.2 The Level of Power

11.2.3 Specifying an Effect Size

11.2.4 The Sample Size

11.3 Example of a Single-Level Power Analysis

11.3.1 Using Base R

11.3.2 Using PowerUp!

11.4 Accounting for Clustering in a Power Analysis

11.4.1 The Role of the Intraclass Correlation Coefficient and Design Effect

11.4.2 Conducting a Multilevel Power Analysis Using PowerUp!

11.4.3 Conducting a Multilevel Power Analysis Using PowerUpR

11.4.4 Writing Up a Power Analysis

11.5 Other Software/Websites for Power Analysis

11.6 Summary

Chapter 12. Alternatives to Multilevel Models

12.1 Introduction

12.2 Level-1 Variables of Interest: Estimating Fixed Effect (FE) Models

12.2.1 Computing Cluster Robust Standard Errors

12.2.2 Using lm_robust

12.3 Adding Level-2 Predictors: Beyond FE Models

12.4 Using the Generalized Estimating Equations (GEE) Approach

12.4.1 The Working Correlation Matrix

12.4.2 Using geeglm

12.5 Some Limitations

12.6 Summary

Glossary

References

Index.

Notes:

Includes bibliographical references and index.

Description based on publisher supplied metadata and other sources.

ISBN:

1-0718-4616-7

1-0718-4613-2

1-0718-4614-0

1-0718-4615-9

9781071846162

OCLC:

1372399266

Publisher Number:

276872