Qualitative Comparative Analysis (QCA) and related methods

Professor Dirk Berg-Schlosser, Philipps University Marburg.

This course provides an introduction and overview of systematic comparative analyses in political science where one often is confronted with a small number of highly complex cases, as for example in comparisons of Latin American political systems or particular policy areas. In this respect new approaches and software have been developed in recent years (“Tools for Small N Analysis”, TOSMANA, and “fuzzy set qualitative comparative analysis”, fs-qca) which are able to reduce this complexity and arrive at “configurational” solutions based on set theory and Boolean algebra which are more meaningful in this context than the usual broad-based statistical methods. Real-life, published applications are used throughout the course; participants are also encouraged to bring their own data, if available. Some basic quantitative or qualitative methodological training is probably useful to get more out of the course, but participants with little methodological training should find no major obstacles to follow.

Mixed methods design

Prof. Max Bergman, University of Basel.

This course deals with methodological triangulation and “mixed methods” as one way of overcoming the often deplored quantitative – qualitative divide in social research methods. We will study when, why, and how to combine and integrate qualitative and quantitative research methods. Our focus is primarily practical, which means that our main emphasis will be on exploring the possibilities and limits of mixing qualitative and quantitative methods. The course includes lectures, practical exercises, computer lab work, and assignments. In the first part of the course, more conventional mixed method designs will be covered, including sequential, convergent, and concurrent designs. The second part of the course will be dedicated to the exploration of a more advanced approach to integrating methods, particularly holistic designs and the immediate context during data collection and analysis. Intermediate knowledge in either qualitative or quantitative methods and basic skills in SPSS and word processing are assumed.

Analyzing cross-national data sets with multivariate techniques
Prof. Bruno Cautrès, Sciences Po, Paris.

One of the most important recent transformations in the field of comparative politics has been the expanding range of cross-national survey resources facilitating the systematic cross-national analysis of public opinion and socio-political attitudes around the globe. The globalization of the study of cross-national public opinion over successive decades as well as the recent developments of sophisticated statistical methods has stimulated the field of comparative politics in directions like the quantitative and comparative analysis of political values (new cultural values), social cleavages (decline of old politics) or attitudes to democracy and democratization (critical citizens thesis). This course will go in two main and complementary directions: a) how to use these comparative data sets? what are the designs of these studies and how does it make possible to control for national variations? how do they face the methodological questions of equivalence among countries ? b) how the major multivariate statistical techniques (multiple linear regression, logistic regression, factor analysis) can be used to analyze these cross-national surveys and to test for homogeneity of the relationship between variables across countries ? In other words, how to control for country effect thanks to multivariate methods ? The course will teach these multivariate methods in simple terms. The data sets used will be the World Values Studies and the Latinobarometer studies. The course combines the learning of fundamental points about the statistical techniques used to analyse such surveys (without going into too complex statistical things) and the learning of the practical problems encountered when using them. This is an introductory course even if exploring multivariate methods.

Behavioral games and strategies in politics
Prof. Rebecca Morton, New York University.

How do governments form coalitions after elections? When does a president or prime minister decide to delegate decision-making to others? Why would a democracy engage in war? Do compulsory voting laws result in a more informed electorate? How do runoff election requirements affect which candidates win? Is ethnic identity important in forming parties in new democracies? What determines when individuals decide to protest a dictatorship? These questions are just some of the many examples of political science questions that have been addressed using behavioral game theory. Game theory is the study of interactions between individuals in which the choices one person or group makes have impacts on the choices of others and vice-versa. In behavioral game theory we consider how these choices might be sometimes made under situations of limited information or limited rationality on the part of the actors. In the course we will explore both standard and behavioral game theory with applications to political science research questions such as those mentioned above. Students will also work on a project in which they read and study behavioral games applied to a political science question of their own interest.

Quantitative textual analysis
Dr. Iñaki Sagarzazu, Nuffield College, University of Oxford

This course is intended to review, understand, and apply content analysis methods which systematically extract information from texts for analysis for scientific purposes. The course will discuss the more traditional approaches; however, it is aimed at the most recent advances in quantitative content analysis that treat words as data that can later be analyzed using statistical tools. Lessons will consist of a mixture of theoretical grounding in approaches and techniques, with hands-on content analysis of real texts using analytic and statistical software. Throughout the length of the class the students will have ample opportunity to understand issues associated with textual analysis, such as research design, reliability, validity, generalizability, and subjectivity.

Confirmatory factor analysis and structural equation modeling: Cross cultural analysis with the European social survey (ESS)
Prof. Peter Schmidt, University of Giessen.

In this course the basic ideas and applications of confirmatory factor analysis and full structural equation models with latent variables and multiple indicators are dealt with. Special Emphasis is given to cross cultural applications. We use the AMOS program (version 19) with the graphical input. In a first step, the testing of measurement models and scale construction including multiple factor analysis, Multi-Trait Multi-Method (MTMM) models, and second order factor analysis are treated,using the Schwartz Value inventory. In addition, special emphasis is given to multiple group confirmatory factor analysis for cross-cultural comparison and comparative politics in general. In the second step, full structural equation models including the topics of missing values, mediation and moderation (interaction) are dealt with. Special emphasis is given also to the use of multiple group modeling as alternative to multilevel modeling for cross-national research.Data and examples are from the European Social Survey (ESS) and the International Social Survey Program (ISSP). Participants will have a chance to address their own research problems in this field.

Multiple regression analysis
Prof. Guy Whitten, Texas A&M University.

This course is designed to take students with a minimal background in statistics and mathematics and teach them the tools that they need to test their theories and produce presentations of their results for the top journals in Political Science, International Relations, Public Policy, and other related disciplines. Students are encouraged to bring their own data sets so that they can get hands-on experiences with applying the techniques covered in this course. Thus the emphasis is on making the transitions between theory, model specification, and result presentation as seamless as possible. The course is divided into three parts. The first part involves a thorough presentation of the logic and the central assumptions underlying the multivariate ordinary least squares regression model. The second part focuses on issues that researchers typically encounter as they attempt to test their theories in a regression framework. The third part focuses on application and extension of the concepts covered in the first two parts to time series data, pooled time series data, and models of limited dependent variables. This is a hands-on course, meaning that a major goal is to have students learn about techniques by putting them to work with statistical software. To facilitate this, we will have lectures on each topic followed by lab sessions supervised by a teaching assistant. In these lab sessions students will have the option of working with their own data or working with data provided by the instructor. The main statistical software program that we will use for the labs is Stata. When possible, exercises will also be available in R.

Case study methodology – small-n research designs - NEW ADDITION
Dr. Derek Beach, University of Aarhus, Denmark

The aim of this course is to provide students with a set of methodological tools that enable the use of case study methods. The course is designed for students in the early to mid-stages of the dissertation who have already defined their research question. A constant theme throughout the course will be on debating the strengths and limitations of different small-n methods, illustrating the types and scopes of inferences that are possible, and whether and how they can be nested into mixed-methods research designs. The course starts by introducing the debate on whether there is divide between quantitative, large-n and qualitative, small-n methods. This is followed by sessions on working with concepts, case selection principles in different case study methods, and discussions about causal inferences and causal relationships in small-n methods. The course then turns to individual case study methods, including cross-case, comparative research designs (Mill’s methods, structured-focused comparisons and typological theorization) and within-case methods (congruence/pattern-matching, and process tracing). Particular emphasis will be given to process tracing methods. In the third part of the course, we debate when and how different small-n methods can be nested into mixed-methods designs. The final exercise will be the production of a 5-10 page research design paper. Course prerequisites: each student should arrive at the summer school with a 2-3 page description of their research question and tentative research design which will be presented and discussed during the course.

Techniques for analyzing public opinion for political and economic decision makers
Prof. Clifford Young, IPSOS-Washington [Ph.D., University of Chicago]

[Please note that this is only a one-week course. It will run from January 30 to February 3]

Public opinion is sacrosanct in any democracy, but is public opinion a leading or lagging indicator? Indeed, it both determines who will govern as well as helps determine which policies will be more and less likely to succeed. At a political level, pundits and politicians often see public opinion as highly ephemeral—easily influenced when the right messages are pushed. At a public policy level, policy makers often focus on the most 'technically sweet' policy solution, without considering what public opinion thinks. Contrary to these beliefs, however, both practice and scientific evidence show that public opinion is a stable, measurable, and, ultimately, predictable phenomenon. The course will explore this issue both conceptually and in practice. First, it will include a review and discussion of relevant literature on the subject: what is public opinion? Second, the course will include concrete case-studies exploring the uses and misuse of public opinion and polling by political and policy stakeholders. Cases will include, for example, the 2005 Brazilian vote buying scandal, the Obama Election 2008, and the US healthcare reform 2009. The final objective is to develop a critical eye when analyzing public policy and political problems. There no special statistical or mathematical requisites for this course, previous knowledge of survey research techniques in general is helpful.

A Practical Introduction to Bayesian Statistical Modeling - NEW ADDITION
Prof. Simon Jackman, Stanford University

[Please note that this is only a one-week course. It will run from February 6 to February 10]

This course provides a practical introduction to Bayesian statistical inference, with an emphasis on applications in the social sciences. We will begin with a brief consideration of how Bayesian statistical inference differs from classical or frequentist inference. We will examine these differences in the context of simple, familiar statistical procedures and models: e.g., inference for proportions, regression, etc. But the bulk of the class will focus on simulation-based Bayesian inference. The explosion in desktop computing power through the 1990s and 2000s has made Bayesian approaches attractive for more complex models. Specifically, the set of algorithms known as Markov chain Monte Carlo (MCMC) allow researchers to tackle classes of problems that used to fall in the ‘‘too hard’’ basket. Today, MCMC is well and truly part of the statistical computing toolkit available to social scientists, and implemented in various forms in many different software packages (we will survey some of these, see below). We will examine how these algorithms make Bayesian inference feasible, their strengths and weaknesses, and some of the pitfalls to avoid when deploying MCMC algorithms. The applications to be considered in this part of the course include • generalized linear models for binary and ordinal data • multinomialchoicemodels • models for latent variables,e.g.,factor-analytical models ,including structural equation models; item-response models; • classification and clustering, both cross-sectionally and dynamically (i.e., change-point or ‘‘structural breaks’’) • dynamic latent state models (e.g., tracking public opinion over time), • hierarchical models of various flavors (appropriate to many forms of data in the social sciences)

Refresher Course in Mathematics and Statistics

Dates: January 23-27, 2011

Profs. Glauco Peres da Silva, Fernando Guarnieri and Lorena G. Barberia

This course is designed for students who require basic training in mathematical concepts and statistics, which are essential for understanding formal and quantitative analysis in political science research. It prepares students for the courses offered in the IPSA Summer School 2012.  The course will take place in the week preceding the commencement of the Summer School. It will cover topics including calculus, linear algebra, and probability theory. To complement lectures, students will apply the concepts taught in lectures to analyze problems using the mathematical and software packages commonly used in quantitative social science research including  Matlab, Stata, R, and SPSS.   The course presumes students have basic notions of arithmetic and algebra operations. 

The file below presents the outline of the course and the Professors vita.