Multilevel modelling 1: an introductionMethodology courses and philosophy of science


Course information

ECTS: 2.5 
Number of sessions: 4
Hours per session: 3
Course fee:

  • free for PhD candidates of the Graduate School
  • €475,- for non-members
  • consult our enrolment policy for more information

Contact:

Telephone: +31 (0)10 4082607 (Graduate School)


Session 1
March 20 (Wednesday) 2019
09:30-12:30
Location will be announced soon

Session 2
March 27 (Wednesday) 2019
09:30-12:30
Location will be announced soon

Session 3
April 3 (Wednesday) 2019
09:30-12:30
Location will be announced soon

Session 4
April 10 (Wednesday) 2019
09:30-12:30
Location will be announced soon


Aims and working method

In this course, PhD candidates will get an introduction into the theory of multilevel modelling, while focussing on two level multilevel models with a 'continuous' response variable. In addition, participants will learn how to run basic two-level model in R. 

Before each meeting, participants will have to (individually) follow the assigned parts of our Massive Open Online Course (MOOC) on Coursera.org.  During the meetings the theory presented in the MOOC will be discussed in more detail, and any remaining questions will be answered. In addition, participants will get to practice with multilevel analysis in R using both exercises and their own data.
 


Learning objectives

The objective of this course is to get participants acquainted with multilevel models. These models are often used for the analysis of ‘hierarchical’ data, in which observations are nested within higher level units (e.g. repeated measures nested within individuals, or pupils nested within schools).

In this type of data causes of outcomes (e.g. the performance of pupils in schools) are located both at the level of the individual (e.g., own and parental resources), and at a higher, contextual, level shared by some of the individuals (e.g. characteristics of the class and of the teacher).

Because of this, the assumption of 'independent observations' is violated with hierarchical data, but multilevel modelling can easily account for that. Moreover, multilevel modelling can easily deal with missing data (in most circumstances).


Session descriptions

Session 1: 
Introduction to multilevel modelling

Preparation

Session 2:
The Basic Two-Level Regression Model and the HLM program.

Preparation

  • Read Chapter 2:  The basic two-level regression model: introduction” Hox, J. (2002) Multilevel Analysis. Techniques and Applications. Mahwah: Lawrence Erlbaum Associates, Inc., Publishers. Available online (pdf).
  • Bring your laptop to class.

Session 3:
Longitudinal data

Preparation:

  • Prepare questions on your own research.
  • Bring your laptop to class.

Session 4:
Methodological and statistical issue and own research

Preparation:

  • Before class, send in question about your own research. You will receive personal feedback during class.
  • Bring your laptop to class.

     


About the instructor

Joran Jongerling is an Assistant Professor at the EUR Department of Pedagogical and Education Sciences, where he convenes courses in research methodology, SPSS skills and applied multivariable data analysis. In his research he applies and analyses principles from Bayesian statistics, Multilevel analysis, and Structural Equation Modelling.