Introduction to Hierarchical Models

Introduction to Hierarchical Models

Description

The course covers statistical models that account for observed hierarchies.  Such hierarchical structures provide great opportunities for research.  The course introduces the statistical tools for dealing with these hierarchies.  The course begins with multilevel linear models, then moves on to multilevel models with dichotomous outcomes, and finally covers multilevel generalized linear models.  In each case, the course focuses on how to specify, fit, and check multilevel models in R.

Required Background

The course assumes a basic knowledge of hypothesis testing, the linear regression model, and probability theory.  A basic familiarity with R is helpful.

Readings

Required text

Gelman, Andrew & Jennifer Hill (2007).  Data Analysis Using Regression and Multilevel/Hierarchical Models.  Cambridge: Cambridge University Press.

Background reading

Gelman, Andrew, Boris Shor, Joseph Bafumi & David Park (2007).  Rich State, Poor State, Red State, Blue State: What’s the Matter with Connecticut?  Quarterly Journal of Political Science.  345--367.   Available online at http://www.stat.columbia.edu/~gelman/research/published/rb_qjps.pdf

Keele, Luke (2006) An Introduction to R.  Available at: http://www.personal.psu.edu/ljk20/RIntro.pdf

See the course outline

Instructor

Andreas Murr
Departmental Lecturer in Quantitative Methods in Political Science
Department of Politics and International Relations
University of Oxford
andreas.murr@politics.ox.ac.uk
http://www.politics.ox.ac.uk/academic-faculty/andreas-murr.html

Andreas is Lecturer in Quantitative Methods in Political Science at the Department of Politics and International Relations at the University of Oxford, Associate Member at Nuffield College, Non-Stipendiary Lecturer at Lincoln College, and part of the Oxford Q-Step Centre.  He is member of the editorial board of Electoral Studies, co-editor of The Plot and co-convenor of the Political Methodology Specialist Group of the Political Studies Association.

Andreas specializes in quantitative methods, particularly in Bayesian statistics and hierarchical models. His substantive research focuses on electoral behaviour, including models of decision making and election forecasting.  His research focuses on electoral behaviour, in particular on election forecasting.  Part of this work has been published or is forthcoming in Electoral Studies, International Journal of Forecasting, Political Analysis, and Research & Politics.