Royce Carroll is a Professor in Comparative Politics at the University of Essex. His research focuses on representation and legislative politics, as well as methods to analyse political choices, attitudes, preferences and ideology. He has previously taught at Rice University. He is co-author of the scaling method textbook Analyzing spatial models of choice and judgment with R (2014), as well as many articles on related topics.
Course Content: This course focuses on methods to discover and understand latent patterns in data and is especially suited to students with projects using survey data and other forms of relational data. The course introduces students to measurement theory and methods of ideal point estimation, as well as related scaling techniques, including item response theory. The first part of the course will provide an overview of the foundations of these techniques and introduce students to the most common methods used in political science. The course will demonstrate how to interpret latent dimensions of data via a variety of ideal point estimation and scaling methods using the open-source programming language R. The course will also discuss a range of applications these methods to social science studies of relational and perception data derived from elite behaviour and surveys. The course concludes with discussions of recent advances in the field and practical advice for those seeking to use such methods.
The course first covers how to analyse data from scales, focusing on surveys that ask respondents to place themselves and / or stimuli on issue or attribute scales. The course begins with the Aldrich-McKelvey scaling and ’basic space’ methods to generate bias adjusted and latent spatial data from survey responses. The course next examines similarities and dissimilarities data and covers multidimensional scaling (MDS) with a focus on the SMACOF optimization method implemented in R as well as Bayesian applications to metric MDS. Next, the course covers unfolding analysis of rating scale data such as feeling thermometers in which respondents report their favourability scale toward stimuli such as politicians or social groups. Finally, the course provides extensive overview of the unfolding of binary choice data, such as legislative roll call voting. Here we will cover Poole and Rosenthal’s W-NOMINATE and Poole’s Optimal Classification unfolding method, as well as Bayesian analysis of binary and ordinal choice data using Item Response Theory (IRT) implemented by Jackman’s pscl and Martin and Quinn’s MCMCpack.
This course will enable students to derive latent spatial preference information and/or dimensional structure from various types of choice and judgement data, which is applicable to a wide range of social science applications. Consumers of research based on these methods will also benefit from a deeper understanding of this type of methodology, its potential and its limitations.
Text(s): Analyzing spatial models of choice and judgment with R., 1st Edition
Author(s): Armstrong, D. A., Bakker, R., Carroll, R., Hare, C., Poole, K. T., & Rosenthal, H.; This book will be provided by the Summer School on arrival as part of the course material.CRC Press. ISBN-13: 9781466517158
Course Objectives: Students will learn to use various computational methods to generate relational measures of ideology and preferences and understand the latent dimensional properties of social science data. As these techniques are fundamental parts of much recent work in political science in particular, students will be able to better understand and produce this type of research
Course Prerequisites: : The course is designed to be accessible to social science graduate students of all backgrounds. However, students familiar with the R programming environment will find it easier to adapt to course content and assignments, so it is recommended to complete some preparation in the basic structure of R. In addition, basic the course assumes basic familiarity with general statistics.
Background knowledge required
OLS = e
Maximum Likelihood = e
R = m
e = elementary, m = moderate, s = strong