Название: Applied Linear Regression for Longitudinal dаta: With an Emphasis on Missing Observations
Автор: Frans E.S. Tan, Shahab Jolani
Издательство: CRC Press
Год: 2022
Страниц: 249
Язык: английский
Формат: pdf (true), epub
Размер: 10.1 MB

This book introduces best practices in longitudinal data analysis at intermediate level, with a minimum number of formulas without sacrificing depths. It meets the need to understand statistical concepts of longitudinal data analysis by visualizing important techniques instead of using abstract mathematical formulas. Different solutions such as multiple imputation are explained conceptually and consequences of missing observations are clarified using visualization techniques. Key features include the following:

• Provides datasets and examples online
• Gives state-of-the-art methods of dealing with missing observations in a non-technical way with a special focus on sensitivity analysis
• Conceptualises the analysis of comparative (experimental and observational) studies

Special attention is given to the analysis of longitudinal intervention and life-event studies, where the objective is to evaluate a treatment or life-event effect. Several statistical methods to deal with missing observations are presented, depending on the type of missing data mechanism and whether the dependent variable (outcome), the independent variables (covariates) or both are partly missing.

Chapter 1 introduces the scientific framework of linear regression analysis and the underlying theory of missing data methods. Chapter 2 starts with a brief review of standard linear regression model, and the notation and terminology of multilevel linear models are introduced. In addition, this chapter reviews simple and advanced methods for handling missing observations. The material in Chapter 3 and Chapter 4 forms the heart of multilevel analysis. Various examples are used to introduce random-effects and marginal models and to explain steps of model building in longitudinal data. Suggestions are given on how to deal with missing data problems when considering imputation strategies. Chapter 5 compares the analysis of covariance (ANCOVA) and gain-score approach in pre/post measurement designs. To address the problem of missing observations, sensitivity analysis via multiple imputation is demonstrated. Chapter 6 and Chapter 7 serve as case-studies to perform a full analysis on longitudinal data in observational and experimental studies, respectively.

This book can be used as a text for a 12-week course. We recommend starting with a brief introduction to the statistical software/packages that will be used during the course. The book’s companion website offers a manual to perform the analyses in SPSS and R. Although the theory is explained independently of any statistical package, SPSS and R were mainly used to produce tables and figures. We also provide data files in the SPSS system-file format that may be used (not necessarily in SPSS) when working on the assignments. An introduction in SPSS (or R) can be covered in 1 week. Chapter 1 and 2 can be taught in 2 weeks including a discussion of the assignments. Furthermore, each of the Chapter 3, Chapter 4, Chapter 5 and Chapter 6 can also be taught in about 2 weeks, including a repetition of the preceding chapters and a discussion of the assignments. Finally, Chapter 7 can be covered in 1 week.

Basic knowledge of linear regression analysis and testing theory would be an advantage for a smooth understanding of the topics of this book. In fact, we have used it in courses where participants already had some experience analysing data with regression methods.

It is the ideal companion for researcher and students in epidemiological, health, and social and behavioral sciences working with longitudinal studies without a mathematical background.