04 Jun module 02: multilevel analysis
- module 2: Multilevel analysis
09/09/2019 - 13/09/2019
9:30 am - 11:00 am
Mon 9/09: 9.30 am - 12.45 pm | S8 room 2.036
Tue 10/09: 9.30 am - 12.45 pm | S9 room 3.1
Wed 11/09: 9.30 am - 12.45 pm | S9 room 3.1
Thu 12/09: 9.30 am - 12.45 pm | S9 room 3.1
Fri 13/09: 9.30 am - 11.00 am | S9 room 3.1
In this course you will learn the basics of multilevel analysis to analyze complex, nested data. In many situations data have a nested structure: observations are clustered within a (naturally occurring) entity. In for example, psychological research employees may be nested within organizations, in neuroscience multiple cells may be harvested from the same mouse, and in longitudinal studies repeated measurements are nested within persons. The appropriate analysis technique for this kind of data is multilevel modeling: using multilevel analysis prevents spurious significant results, ensures optimal statistical power and enables one to answer new types of research questions. In this course you will learn the theoretical basics of multilevel modeling and you will learn to analyze data sets with the computer programs SPSS and HLM (and R, if preferred). Synonyms for multilevel analysis are hierarchical linear model, random effects model, and mixed linear model. This course aims at introducing participants to multilevel analysis in a very hands-on style: every lecture is followed by a computer lab to practice analyzing data using multilevel analysis. Examples in the computer lab are mainly from the social sciences.
Day 1: Introduction to multilevel analysis and the basic two-level regression model.
Day 2: Lab: Introduction to multilevel analysis and the basic two-level regression model.
Day 3: Analysing longitudinal data and contextual effects using a multilevel framework.
Day 4: Lab: Analysing longitudinal data and contextual effects using a multilevel framework.
Day 5: Analysing dichotomous and ordinal data using multilevel analysis.
To take this course, you should have a good working knowledge of the principles and practice of multiple linear regression, as well as elementary statistical inference. We will use SPSS and HLM and, if you prefer, R/Rstudio. Some familiarity with SPSS is desirable.
!!! Bring your own laptop, with at least SPSS and HLM installed on it !!! A student version of HLM7 can be downloaded for free at http://www.ssicentral.com/hlm/student.html
- Multilevel analysis: Techniques and applications (2018). Joop J. Hox, Mirjam Moerbeek and Rens van de Schoot. Third edition, Routledge. ISBN: 978-1-138-12136-2.
- Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: a new look at an old issue. Psychological methods, 12(2), 121.
Prof. Emmeke Aarts is an assistant professor based at the department of methodology and statistics at Utrecht University. After a research master Methodology and Statistics (cum laude) at Utrecht University, she obtained her PhD at the Center for Neurogenomics and Cognitive Research at the VU University Amsterdam, combining statistics with Neuroscience. After working at the Max Planck Institute for Molecular Genetics in Berlin and at TNO Life Sciences in Zeist, She returned to Utrecht University. Here, she combines teaching with developing new statistical methods in the area of intense longitudinal data, in close collaborations with applied researchers from various fields. Her areas of expertise include multilevel analysis, hidden Markov models, Bayesian estimation methods and machine learning methods.
Venue: University of Ghent
Venue Phone: +32 9 264 52 39
Venue Website: https://www.flames-statistics.com/summer-school/venue/Address: