Multilevel Modeling

2025/2026

Учебный год

ENG

Обучение ведется на английском языке

3

Кредиты

Кто читает:

Департамент социологии

Статус:

Курс по выбору

Когда читается:

1-й курс, 3 модуль

Преподаватели

Иванюшина Валерия Александровна

Титкова Вера Викторовна

Course Syllabus

Abstract

Students will be introduced to the basics of hierarchical multilevel modeling. We will consider the applications of HMM and how to implement it depending on the needs of the research. The course includes a discussion of fixed-effects, random-effects, and multilevel interaction effects models. Attention is paid to both the theoretical foundations of analysis and practical skills of working with multilevel data, including model specification, parameter estimation, interpretation of results, and model adequacy testing. Students learn to use specialized software tools for analyzing multilevel data in R (lme4, nlme, plm, lmerTest, MuMIn, sjPlot and others). The application of multilevel analysis to different fields of research is discussed using the example of empirical articles. During the implementation of independent projects, students participate in applying the acquired knowledge in practice.The course is intended for students and researchers in various fields, including sociology, economics, education, healthcare, and others where data is hierarchical in nature.

Learning Objectives

• Upon completion of the course, students will be able to: - Understand the concepts and principles of multilevel modeling. - Apply various methods of multi-level data analysis to study and model complex hierarchical structures. - Evaluate the influence of factors at different levels of the hierarchy. - Interpret the results of multilevel models. - Apply the acquired knowledge and skills in research and practical activities in various fields where data has a hierarchical structure.

Expected Learning Outcomes

• Know the requirements and limitations of least squares regression. Know what a multi-level data structure is
• To understand the differences between models with random intercept and random slope
• Know how different types of centering differ. Understand what is moderation (interaction effect)
• Have an understanding of synthetic data simulation, the goals and capabilities of the simulation method. Be able to create datasets with specified characteristics
• Understand the difference between fixed and random models; requirements and limitations of each model type
• Understand the advantages and limitations of multi-level models. Be able to make an informed choice of the right model, taking into account the specifics of the data.

Course Contents

• Introduction to Multilevel analysis
• Types of multilevel models
• Centering variables. Interaction effects (moderation)
• Project presentations
• Multilevel models with a time component
• Evaluation of the model. Diagnostics of model quality.
• Model selection. Mistakes and misconceptions. Myths about multilevel models

Assessment Elements

• Hometasks
• Project 1
• Project 2
• Test

Interim Assessment

• 2025/2026 3rd module
  0.4 * Hometasks + 0.2 * Project 1 + 0.2 * Project 2 + 0.2 * Test

Bibliography

Recommended Core Bibliography

• A practical guide to using panel data, Longhi, S., 2015
• Data analysis using regression and multilevel/hierarchical models, Gelman, A., 2009
• Multilevel analysis : techniques and applications, Hox, J. J., 2017

Recommended Additional Bibliography

• Handbook of advanced multilevel analysis, , 2011
• Multilevel analysis : an introduction to basic and advanced multilevel modeling, Snijders, T. A. B., 2012
• Panel data: theory and applications, , 2010
• Time-series forecasting, Chatfield, C., 2000

Authors

• IVANYUSHINA VALERIYA ALEKSANDROVNA
• Titkova Vera Viktorovna