• Understand and explain the methods introduced in the course
• Apply the methods to analyse a well-defined empirical research problem
• Select methods that are appropriate for analyzing particular types of variables and problems
• Interpret the results of the quantitative analysis appropriately and communicate the results in way that is comprehensible to a wider audience
• Implement a quantitative, empirical analysis using the statistics program Stata.
• Discuss and evaluate the strengths and weaknesses of the methods

The purpose of this course is to introduce students to quantitative methods at an intermediary and advanced level and to enable participants to apply the methods to analyse their own data and research problems. The course will increase the level of proficiency and knowledge of quantitative methods participants have acquired in courses corresponding to Statistics and Research Methods (IBP BSC). The course will emphasise the application of methods to empirical research problems, and provides an introduction to a series of methods and techniques – as well as their underlying assumptions – that are relevant for analysing quantitative data in comparative and international political economy, business research, and political and social science more broadly. The course will also emphasise the ability of students to present, interpret, and explain the results of their quantitative analysis in a comprehensible manner that is accessible to a wider audience. The course is organized in two parts. The first part deals with models for cross-sectional data. This part starts by recapitulating basic methods for analyzing cross-sectional data, and also contains an introductory module for the software program Stata. This is done through two brush-up sessions on OLS regression – the work horse model of quantitative social science research – including assumptions tests and the use of multiplicative interaction terms in OLS regressions. This is followed by two sessions on categorical dependent variables – introducing logit and ordered logit regressions. The course moves on to more advanced methods for cross-sectional data analysis, which introduces participants to propensity score matching and instrumental variables regression, both of which are becoming increasingly popular for causal analysis of quantitative data. The second part of the course deals with models for analyzing panel and multilevel data, which offer a distinct set of possibilities and problems in quantitative data analysis. In a panel, the data have a cross-sectional and a time-series dimension (e.g., a set of firms or countries observed over time). With multilevel data, one set of units (level-1) is nested within another set of units (level-2), e.g. individuals nested in firms or municipalities, meaning that information from multiple levels can enter the analysis. Throughout the course, the strengths and weaknesses of the particular methods will be discussed, just as advice for presenting and disseminating the results of quantitative analysis will be provided.

Teaching consists of a mix of lectures, group work, and practical exercises. Each session will consist of a lecture followed by group work and/or practical exercises. Data for use in the exercises are provided to the students. Students are also expected to prepare a certain number of exercises before class. The course is aimed at students at the IBP program (or similar), and is particularly useful for students who want to use quantitative methods in their upcoming BA projects and during their studies on the Master program. The software program used in the course is Stata, which can be accessed on https://e-campus.dk/it Students are encouraged to familiarize themselves with Stata before the course starts.

The course plan looks as follows

Part I: Models for cross-sectional data: 1) OLS brush-up and introduction to Stata I; 2) Interaction terms and introduction to Stata II; 3) Exercises using Stata; 4) Models for categorical dependent variables I: Logit regression; 5) Models for categorical dependent variables II: Ordered logit regression; 6) Propensity score matching; 7) Instrumental variables regression. Part II: Models for panel and multilevel data: 8) Panel data models I: Fixed and random effects; 9) Panel data models II: Dynamic models; 10) Multilevel data models I: Models for continuous dependent variables; 11) Multilevel data models II: Models for categorical dependent variables; 12) Presenting and disseminating quantitative analyses + problems and pitfalls with quantitative methods

Gelman, Andrew and Jennifer Hill (2006). Data Analysis Using Regression and

Multilevel/Hierarchical Models. New York: Cambridge University Press.

Sønderskov, Kim (2014). Stata – A Practical Introduction. Copenhagen. Hans Reitzels Forlag.

Wooldridge, Jeffrey (2009). Introductory Econometrics: A modern Approach.

Selection of journal articles