Multilevel Models (MAD)

Kurs-Nr.: 080400 | Zeit: Di 14-16 | Raum: FNO 02/074 CIP Raum | Semester: WiSe 2016/2017


Registration in CampusOffice. The registration for this course starts on August 22th 2016.
Participants should have a basic understanding of linear regression models. Some experience with the statistical software R is required – nevertheless the course will start with a short introduction into R.

Kommentar Social scientists are often confronted with hierarchical structured data: An often stressed example are students, which are grouped into classes, classes belong to schools and those schools are influenced by national or federal regulations. Another example are hierarchical structured regional data like individuals from neighborhoods in cities and regions. Theoretical models in these settings often assume cross-level interactions between the individual level and higher levels. A common assumption is that the social composition of a school has an effect on the individual student performance or that the neighborhood context influences the individual probability of delinquent behavior.
Statistical models referred to as multilevel (linear) models, mixed-effects models, covariance component models or random-effects models have been proposed in the literature for this kind of data and are often rated superior to simple OLS regression.
The course will cover an introduction into practical application and interpretation of multilevel models using R, the discussion of statistical as well as theoretical limitations of these models and alternative methods. In addition, research examples from different fields will be discussed.
Please note that the course will be held in English.

Voraussetzungen für Studiennachweise / Modulprüfungen

Modulprüfung: active participation, completion of exercises in R and term paper
Studiennachweis: active participation, completion of exercises in R


  • Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge: Cambridge University Press.
  • Hox, J. (2002). Multilevel analysis. Mahwah, NJ [u.a.]: Erlbaum.
  • Kreft, I., & Leeuw, J. (2002). Introducing multilevel modeling. London [u.a.]: Sage.
  • Luke, D. A. (2004). Multilevel modeling. Thousand Oaks, Calif: Sage Publications.
  • Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: applications and data analysis methods (2nd ed). Thousand Oaks: Sage Publications.