Doctoral theses supervised by Thomas Gautschi, Henning Hillmann, Frank Kalter, Florian Keusch, Irena Kogan and Frauke Kreuter, respectively, will be discussed.
Colloquium | |||||||
Keusch/ |
03.09.18 – 03.12.18 | Monday | 12:00 – 13:30 | tbc | |||
Kalter/ |
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | tbd | |||
Gautschi/ |
04.09.18 – 04.12.18 | Tuesday | 15:30 – 17:00 | tbc | |||
The course “Current Research Perspectives” introduces first year doctoral students to the theoretically informed research approaches and substantive research fields that build the strongholds of social science research in Mannheim. A series of talks provides first year doctoral students with an overview of current debates and ongoing research in the fields of psychology, political science and sociology. CDSS faculty members will present an overview of their research fields, report on prime examples of their current research, and provide an outlook on potential topics for future research. Doctoral students will have the opportunity to discuss the short talks with the respective lecturer during the remaining discussion time.
Homework assignment (3 pages)
Lecture | |||||||
04.09.18 – 11.09.18 | Tuesday | 13:45 – 17:00 | 211 in B6, 30–32 | Link | |||
12.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
26.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
It is increasingly important for modern social scientists to have a level of mathematical literacy, as mathematical research methods such as statistics and formal modelling have entered the main stream. This course is intended to provide an introduction to mathematical logic and rigour, and to some fundamental mathematical concepts that form the foundation of the modern subject. The course covers introductory set and function theory, including analysis of functions, and includes sections on both probability and linear algebra, which together are the basis of data analysis.
The exam is scheduled for 11 December at 9.30am in room 108 in B6, 30–32.
Basic readings:
Additional readings:
Lecture | |||||||
05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | 211 in B6, 30–32 | ||||
All researchers face similar challenges with core issues of research design. A research design is a plan that specifies how you are going to carry out a research project and, particularly, how to use evidence to answer your research question. The goal of this course is to jump-start students with their dissertation proposal. This course should help students to see the trade-offs involved in choosing a particular research design in their research projects. Consequently students are expected to develop own ideas about potential research questions and actively participate in those seminar-style meetings that are organized within this lecture course.
Workshop | |||||||
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | A 203 in B6, 23–25 | Link | |||
Questions of cause and effect are at the heart of social science. And yet, establishing credible causal effects in empirical analyses is a difficult enterprise. This course will introduce some of the key conceptual and methodological approaches to tackle the causal inference problem: the potential outcomes model of causal inference, experimental designs, matching and regression, instrumental variables, regression discontinuity designs as well as difference-in-differences and fixed effects.
Workshop | |||||||
06.09.18 – 11.10.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
not on 12 Oct | 07.09.18 – 19.10.18 | Friday | 10:15 – 13:30 | 310 in B6, 30–32 | |||
CSSR, TBCI, EAW, Literature Review, Dissertation Proposal
The goal of this course is to provide support and crucial feedback for second and third year CDSS PhD candidates in sociology on their ongoing dissertation project. In this workshop CDSS students are expected to play two roles. They should provide feedback to their peers as well as present their own work in order to receive feedback.
Workshop | |||||||
1st meeting to determine dates across the semester | 04.09.18 | Tuesday | 15:00 – 16:00 | C212 in A5, 6 entrance C | |||
19.10.18 | Friday | 11:00 – 12:00 | |||||
05.12.18 | Wednesday | 14:00 – 18:00 | |||||
Please refer to the MZES webpages for dates and times.
You should be familiar with the basics of regression models and maximum likelihood estimation. No previous knowledge of software for Bayesian inference is required. However, we will use R as a front-end to JAGS and for graphical displays. Resources to learn R basics are UCLA’s Stat Consulting Site as well as the official manuals at http: //www.r-project.org/.
This course is intended as an introduction to applied Bayesian statistics for social scientists. Bayesian analysis involves two key aspects – inference based on probability theory and estimation using stochastic simulation. The course will spend some time on the basic principles of both aspects (Where do Bayesian priors come from? How does MCMC work?) and then apply them to some of the workhorse models of the social sciences: linear and discrete outcome models as well as their hierarchical counterparts. We will also discuss practical issues of applied Bayesian analysis, such as MCMC convergence diagnostics as well as Bayesian model checking and parameter summary.
Course readings
Workshop | |||||||
18.10.18 – 06.12.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
02.11.18 – 07.12.18 | Friday | 10:15 – 13:30 | 211 | ||||
The course offers an overview and several hands-on experiences on some of the most relevant methods and tools developed in the field of natural language processing, which have been often adopted as basis for quantitative content analyses in social science research. Attention is dedicated to tasks such as collection building, topic modeling, sentiment analysis, text classification, clustering and scaling as well as to the application of methods such as latent Dirichlet allocations, word embeddings and entity linking. A brief introduction to practices such as web scraping and text pre-processing (e.g. tokenisation, part-of-speech tagging, lemmatisation and stemming) is also offered.
The programming language adopted is Python (and in particular the use of Jupyter Notebooks). No previous programming experience is needed.
Further information is available on Federico's web page.
A maximum of 10 ECTS can be obtained for the successful completion of this course.
6 ECTS written exam
4 ECTS coding exercise
Seminar | |||||||
not on 3, 24 and 31 October | 05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | C 108 Methods lab in A5, 6 entrance C | Link | ||
01.10.18 | Monday | 10:15 – 11:45 | 209 in B6, 30–32 | ||||
17.10.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
07.11.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
Sound understanding of linear regression models (OLS), knowledge in linear algebra and calculus, and being familiar with the statistical package Stata.
The main focus lies on the introduction to statistical models and estimators beyond linear regression useful to a social scientists. A good understanding of the classical linear regression model is a prerequisite and required for the further topics of the course. We will first discuss violations of the asymptotic properties of the linear regression model and ways to address these violations (heteroscedasticity, endogeneity, proxy variables, IV-estimator). The second part of the class is dedicated to rst the maximum likelihood estimator and second to generalized linear models (GLS) for binary choice decisions (Logit, Probit), ordinal dependent variables, and count data (Poisson, Negative Binomial). Classes will be accompanied by lab sessions to repeat and practice the topics from the classes. We will use the statistical package Stata.
Assessment type: written exam (90 min)
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 13:45 – 15:15 | 008.2 in B 6, 30–32 | ||||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 15:30 – 17:00 | C-108 in A5,6 entrance C | ||||
05.09.18 – 05.12.18 | Wednesday | 13:45 – 15:15 | C-108 in A5,6 entrance C |
Depending on teaching needs and room availabilities, lectures and tutorials may be swapped, such that two lectures or two tutorial take place on one day within the time slots allocated to the course, i.e. on Tuesdays, 8.30 – 11.45. Such swaps will be announced before the beginning of the semester.
Summary
This course gives an overview of data used in political science and their measurement properties. At the beginning of the course we will focus on survey data and traditional statistics and move to data science approaches for big data towards the end of the course. Topics covered include the Total Survey Error (TSE) framework, operationalizing research questions, guidelines for writing survey questions, testing questions with cognitive interviews and eye-tracking, sampling, coverage, and nonresponse of survey and big data, and data analytics approaches in data science.
Zu erbringende Prüfungsleistung
Credits for the lectures are obtained through a graded term paper. The term paper covers selected topics from the complete course and has to be submitted by ...tbc. Term papers can be written individually or in groups of max. 3 students/
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 08:30 – 10:00 | A 102 in B6, 23–25 entrance A | Link | |||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | C-108 in A 5, 6 entrance C | Link |
Basic knowledge of probability and statistics.
Machine Learning is about designing algorithms that are able to make predictions about data or extract knowledge from data. The aim of this module is to study algorithms, underlying concepts, and theoretic principles that allow for algorithms to automatically learn how to make predictions. The course focuses on selected “hot topics” and their applications, which include:
• Basics of machine learning and probability theory
• Probabilistic graphical models
• Inference and parameter estimation
• Neural networks
Methods: The course consists of a lecture accompanied by theoretical and practical exercises as well as case studies with real data. In the exercises, students will deepen the material discussed in the lecture, apply the methods in practice, and present the result.
Literature:
• K.P. Murphy. Machine Learning: A Probabilistic Perspective, The MIT Press, 2012
• D. Koller, N. Friedman. Probabilistic graphical models. The MIT Press, 2009
• Additional material and articles provided in lecture notes
Form of assessment: Oral or written examination, homework assignments
Workload Hours per semester: 56h (4 SWS)
Self-study per semester: 98 h
• 70 h: pre- and post lecture studying and revision
• 28 h: exam preparation
Lecture | |||||||
Lecture & Exercises | 04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | A 101 in B 6, 23–25 entrance A | Link | ||
Lecture & Exercises | 05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | A 101 in B 6, 23–25 entrance A | |||
In addition to a thorough understanding of the substantive field you are studying you need firm methodological and statistical knowledge in order to successfully conduct quantitative social research. This seminar will give you the opportunity to apply and expand your knowledge of social research by replicating published research findings.
The research that we are going to replicate was conducted with data from publicly available survey data like the European Social Survey (ESS), the International Social Survey Programme (ISSP) or the European Values Study (EVS). Data from surveys like these have several advantages: the surveys follow a repeated cross-section design, a research design particularly well suited to study social change; they are comparative surveys allowing you to compare data cross-nationally on a broad range of topics; the surveys follow rigorous methodological standards and, finally, data are available at no cost and can be downloaded from the web.
Replicating published research has the advantage that you are able to check your results against existing results. By trying to replicate previous research you learn where the original researcher has made tacit decisions not documented in the paper (e.g. defining the analysis sample, coding of variables, treatment of missing values). Replicating social research also trains you to judge the validity of research results.
In addition to these primarily pedagogical aspects replicating research is important from an epistemological point of view as well. Through replication of research by independent research groups biases in previous work can be discovered and findings can be validated (see Hendrick 1991, King 1995).
Participants should choose a published paper and try to replicate the findings reported in it using the same data. The results to be replicated often will be given in a table containing the outcome of a multivariate model. Please document each step in your attempt to replicate the findings, report and explain the decisions you had to make during data preparation and data analysis. If you fail to replicate the results please indicate possible explanations. Your paper should not exceed 10,000 words; please add your documented syntax in the appendix.
Seminar | |||||||
bi-weekly | 07.09.18 – 30.11.18 | Friday | 10:15 – 13:30 | 309 in B6, 30–32 | Link | ||
The course introduces students to quantitative methods in political science. During the first half of the course, we will focus on linear regression models. The topics covered include discussions of the mathematical bases for such models, their estimation and interpretation, model assumptions and techniques for addressing violations of those assumptions, and topics related to model specification and functional forms. During the second half of the course, students will be introduced to likelihood as a theory of inference, including models for binary and count data.
The main goals of this course are to develop sound critical judgment about quantitative studies of political problems, to understand the logic of statistical inference, to recognize and understand the basics of the linear regression model, to develop the skills necessary to work with datasets to perform basic quantitative analyses, and to provide a basis of knowledge for more advanced statistical methods.
In the accompanying course “Tutorial Multivariate Analyses” students will develop the necessary expertise in using statistical software to conduct quantitative research in political science.
Graded assignments include homeworks, a mid-term exam and data analysis projects.
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 08:30 – 10:00 | B244 in A5,6 entrance B | Link | |||
Tutorial | |||||||
Neunhoeffer, Marcel | 06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | Room C -108 (PC Lab) in A5,6, entrance C | |||
Sternberg, Sebastian | 10.09.18 – 03.12.18 | Monday | 12:00 – 13:30 | Room C -108 (PC Lab) in A5,6, entrance C |
In der Vorlesung sollen grundlegende Konzepte zur Behandlung von linearen und nichtlinearen Optimierungsaufgaben erarbeitet werden. Ziel ist es, Lösungsverfahren algorithmisch umzusetzen und nebenbei die Algorithmen in einen sicheren mathematischen Rahmen einzubetten. Das Hauptaugenmerk liegt hierbei auf folgenden Gebieten:
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 426154:15 – 11:45 | B 243 in A 5, 6 entrance B | ||||
05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | B 144 in A 5, 6 entrance B | ||||
Tutorial | |||||||
06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | C 013 in A5, 6 entrance C | ||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | C 012 in A5, 6 entrance C | ||||
07.09.18 – 07.12.18 | Friday | 12:00 – 13:30 | C 015 in A5, 6 entrance C |
tbd
Lecture | |||||||
Lecture & Exercises | 06.09.18 – 06.12.18 | Thursday | 10:15 – 13:30 | C -109 PC Pool in A 5, 6 entrance C | Link | ||
The software R is a computer programming language designed for statistical analysis and graphics. The first part of the course deals with a basic introduction to R, i.e. data handling, basic statistical analyses, the creation of graphics, and linear modeling including test for specially designed hypotheses. In the second part we use R as a programming language for cognitive modeling. We will simulate data based on mathematical models of cognitive functions and analyze these data with maximum likelihood parameter estimation techniques. At the end, I will introduce some advanced techniques, for example the creation of statistical reports with R.
The software package R is free and available on all major platforms (www.r-project.org). I also recommend the free and platform independent Software RStudio as a comfortable IDE for R (www.rstudio.com). A basic introduction to R can be found under:
http://cran.r-project.org/doc/manuals/r-release/R-intro.pdf.
Literature will be given during the course
Course achievement – regular participation in the course; non-graded test
Academic assessment – graded homework
Seminar | |||||||
biweekly | 07.09.18 – 30.11.18 | Friday | 13:45 – 17:00 | EO 162, CIP-Pool | Link | ||
Knowledge of multiple linear regression and some familiarity with generalised linear models, to the level of Freedman (2005, see below) or equivalent. Familiarity with notions of research design in the social sciences and the statistical package Stata or R.
If you need to review material on regression models, please consult this excellent textbook: Freedman, David. 2005. Statistical Models: Theory and Practice. Cambridge University Press.
This course teaches the fundamental concepts behind the estimation of causal effects, including potential obstacles to causal inference, faulty measurement, spuriousness, specification errors, and other problems that can lead to inappropriate causal inferences. We will discuss the benefits and the difficulties of randomization in survey research in the first half of the class. The focus of the second half is on the design of observational studies. Real-world examples will be discussed, with an emphasis on examples from survey methodology. Students will come away with an understanding of how to estimate causal effects in both randomized and observational settings, with a particular focus on careful design of both types of studies.
There will be several assignments/
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | B 243 in A5, 6 entrance B | Link | |||
Tutorial | |||||||
13.09.18 – 06.12.18 | Thursday | 15:30 – 17:00 | A 102 in B6, 23–25 entrance A |
Further SMiP courses open to CDSS doctoral students are:
Hierarchical Bayesian Models with Computer Applications (Instructor: Jeffrey Rouder, Date: 01.10. (10:00 – 18:00) and 02.10.2018 (09:00 – 17:00), Location: Mannheim)
Analysis of Response Time (Instructor: Trisha van Zandt, Date: 16.10. (10:00 – 18:00) and 17.10. (09:00 – 17:00), Location: Mannheim)
An Introduction to modern R, Statistical Modeling, and Mixed Models (Instructor: Henrik Singmann, Date: 15.01. and 16.01.2019, Location: Freiburg)
This course gives an advanced overview of standard multivariate methods and current developments in multivariate modeling. The topics include general and generalized linear models, structural equation models, multilevel modeling and specific models for the analysis of time-related effects. In the morning sessions, we will provide the formal foundations and theoretical background of the model classes for continuous and discrete observations. The afternoon sessions focus on empirical applications and hands-on exercises in data analysis.
The goals are to bring the PhD candidates to a common level of statistical knowledge and data analytic skills and to set the stage for the more specialized topics in the Foundations 2 course and workshops in the following semesters.
24.10.2018: 11:00 – 14:00 in C-107 in A5, 6 15:00 – 18:00 in Schloss, EO 162
25.10.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
08.11.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
09.11.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
06.12.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
07.12.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
11.01.2019: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
Further details can be found on the web page of the RTG 'Statistical Modeling in Psychology'
In the course, an introduction in structural equation modeling with software Mplus and R is provided. The students are introduced in the usage of the software and fundamentals of structural equation modeling. The topics such as path analysis, confirmatory factor analysis, measurement invariance and structural regression models are covered. The students work with existing survey data and conduct analyses for the selected research questions. They present both, statistical bases and results of their analyses.
Required examinations: oral presentation and written paper (5000 words)
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | EO 259 (Schloss, Ehrenhof Ost) | Link | |||
The objective of this course is to provide students with the basics of formal modeling in political science. The course has some breadth in coverage in the sense that it provides a graduate-level introduction and overview to di erent areas in game theory. It is also narrow in the sense that the emphasis is not on application and model testing but getting trained in reading and writing down formal models. At the conceptual level the course will cover the following topics: normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signaling games, preferences and individual choices, basics of decision theory and social choice. At the substantial level, we will use these concepts to study, as examples, candidate competition, political lobbying, and war and deterrence.
Literature
tba
Assessment: mid-term and final exam
Tutorial
This tutorial accompanies the graduate-level introductory lecture in game theory. Its main objective is to practice solution concepts for static and dynamic games of complete and incomplete information.The contents are centered around the material covered in the lecture. Thus, the following key areas will be discussed: preferences and individual choices, decision theory, normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signalling games. At the substantial level, we will use these concepts to study, for instance, candidate competition, political lobbying, and war and deterrence. Students are required to submit weekly problem sets. Moreover, active participation in class discussions is expected.
Lecture | |||||||
03.09.18 – 03.12.18 | Monday | 10:15 – 11:45 | B 317 in A5,6 entrance B | Link | |||
Tutorial | |||||||
05.09.18 – 05.12.18 | Wednesday | 17:15 – 18:45 | B 317 in A5,6 entrance B |
OpenSesame is a free, open-source, and cross-platform software for creating laboratory experiments. Many standard tasks can be implemented in OpenSesame via drag and drop using its graphical user interface. In addition, complex tasks can be realized through the underlying programming language Python. The goal of the workshop is to provide an introduction to both approaches. In doing so, the workshop involves both structured input from the instructor as well as a number of practical exercises so that participants can directly explore the features of OpenSesame. Besides, the workshop will introduce plug-ins that extend OpenSesame for specific purposes, e.g., the psynteract plug-ins that implement real-time interactions between participants (as required in many economic games), and the mousetrap plug-ins that implement mouse-tracking during decision tasks (a method that is becoming increasingly popular in the cognitive sciences to measure preference development). Additional topics will be covered depending on the preferences of the workshop participants. No prior knowledge of the software or Python is required.
As an assignment, participants will create their own experiment based on the requirements discussed in the workshop.
Software:
OpenSesame can be downloaded for free under http://osdoc.cogsci.nl/index.html, where you can also find an extensive documentation.
Literature:
Mathôt, S., Schreij, D., & Theeuwes, J. (2012). OpenSesame: An open-source, graphical experiment builder for the social sciences. Behavior Research Methods, 44(2), 314–324. https://dx.doi.org/10.3758/s13428-011-0168-7
Workshop | |||||||
12.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP Pool | Link | |||
13.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
26.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP-Pool | ||||
27.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
This seminar has a particular focus on research activities in social psychology. Unlike seminars that concentrate on one core thematic topic, this seminar will address a selected variety of different research topics in current social psychology. In each seminar session we will have a presentation either by participating students or by members of the social psychology group. Each presentation will address a current research topic in social psychology. The seminar provides the opportunity to actively discuss methodological, theoretical, and applied implications of the presented research. . A particular focus will rest on the discussion of general methodological aspects. Participants need to have passed at least one other social psychology seminar in the master program or, alternatively, they have to be enrolled in one other social psychology seminar during the same semester (for international visiting students these preconditions may differ, please contact the program manager psychology).
Literature: Will be announced in the seminar
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | 309 in B6, 30–32 | Link | |||
The aim of this course is to introduce students to major strands in comparative-historical analysis in the social sciences. Topics include (but are not limited to) the formation of nation-states in early modern Europe; how such early modern states contrast with the social order of feudal society; the causes and consequences of revolutions; how conflicts among political elites shape the course of modern states; how climate, geography, cultural developments, politics and overseas trade come together to shape societies at large; but also how ordinary people in the past saw and imagined the world around them. We will consider these and other topics by drawing upon classic works in large-scale comparative studies, the Annales School, micro-history, historical anthropology, and studies of historical networks. One caveat: students planning to take this class should be prepared to consider empirical examples that come from rather exotic societies, remote in place and time from our contemporary society.
Do all assigned readings ahead of class
· Contribute creatively to class discussions
· Write a term paper that draws upon/
Seminar | |||||||
06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | B 143 in A 5, 6 entrance B | Link | |||
Experts shape everyday life with arcane knowledge that the public has no full understanding of, and consequentially cannot control. While many still regard doctors, lawyers and even obscure scientists highly for their services, this trust rapidly erodes in other groups. This course uncovers organizational arrangements of expert knowledge and asks how arcane expertise shapes public and private life. We investigate this relationship in three problem areas: (1) Free professions and bureaucratic occupations, including law and economics, as well as the EU and IMF; (2) health and science, including medicine, mental health, autism and AIDS; and (3) technology and the Internet, including power plants, programmers and data science. Across these empirical settings we analyze different processes by which abstract knowledge gains lay salience: Informal relationships between mentors and students, or doctors and patients; formal organizations, ranging from labs and firms to governments and NGOs; and occupations, which regulate medical doctors, architects and others; and we finally ask whether expertise could unfold systematically outside of specific relationships and institutionalized boundaries, such as in open discourse and arguments. On the basis of these processes we aim to understand today’s transition from bureaucratic to technological contexts of knowledge production and application.
Academic assessment: Homework assignment (max. 4000 words)
Seminar | |||||||
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | B 317 in A 5, 6 entrance B | Link | |||
The course “Current Research Perspectives” introduces first year doctoral students to the theoretically informed research approaches and substantive research fields that build the strongholds of social science research in Mannheim. A series of talks provides first year doctoral students with an overview of current debates and ongoing research in the fields of psychology, political science and sociology. CDSS faculty members will present an overview of their research fields, report on prime examples of their current research, and provide an outlook on potential topics for future research. Doctoral students will have the opportunity to discuss the short talks with the respective lecturer during the remaining discussion time.
Homework assignment (3 pages)
Lecture | |||||||
04.09.18 – 11.09.18 | Tuesday | 13:45 – 17:00 | 211 in B6, 30–32 | Link | |||
12.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
26.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
It is increasingly important for modern social scientists to have a level of mathematical literacy, as mathematical research methods such as statistics and formal modelling have entered the main stream. This course is intended to provide an introduction to mathematical logic and rigour, and to some fundamental mathematical concepts that form the foundation of the modern subject. The course covers introductory set and function theory, including analysis of functions, and includes sections on both probability and linear algebra, which together are the basis of data analysis.
The exam is scheduled for 11 December at 9.30am in room 108 in B6, 30–32.
Basic readings:
Additional readings:
Lecture | |||||||
05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | 211 in B6, 30–32 | ||||
All researchers face similar challenges with core issues of research design. A research design is a plan that specifies how you are going to carry out a research project and, particularly, how to use evidence to answer your research question. The goal of this course is to jump-start students with their dissertation proposal. This course should help students to see the trade-offs involved in choosing a particular research design in their research projects. Consequently students are expected to develop own ideas about potential research questions and actively participate in those seminar-style meetings that are organized within this lecture course.
Workshop | |||||||
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | A 203 in B6, 23–25 | Link | |||
Questions of cause and effect are at the heart of social science. And yet, establishing credible causal effects in empirical analyses is a difficult enterprise. This course will introduce some of the key conceptual and methodological approaches to tackle the causal inference problem: the potential outcomes model of causal inference, experimental designs, matching and regression, instrumental variables, regression discontinuity designs as well as difference-in-differences and fixed effects.
Workshop | |||||||
06.09.18 – 11.10.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
not on 12 Oct | 07.09.18 – 19.10.18 | Friday | 10:15 – 13:30 | 310 in B6, 30–32 | |||
Participation is mandatory for second and third year CDSS students of Political Science.
Participation is recommended for first year CDSS and visiting PhD students, as well as for later CDSS PhD candidates, but to no credit.
Other young researchers in the social sciences affiliated with the University of Mannheim (incl. MZES and SFB 884) are also invited to attend the talks.
The goal of this course is to provide support and crucial feedback for second and third year CDSS students on their ongoing dissertation project. In this workshop CDSS students are expected to play two roles. They should provide feedback to their peers as well as present their own work in order to receive feedback.
In order to receive useful feedback, participants will circulate their paper and two related published pieces of research one week before their talk.
Workshop | |||||||
05.09.18 – 05.12.18 | Wednesday | 12:00 – 13:30 | 211 in B6, 30–32 | Link | |||
Please refer to the MZES webpages for dates and times.
CSSR, TBCI, Dissertation Proposal
Attending the Seminar Series on the Political Economy of Reforms is a possible alternative to attending the MZES B colloquium. Please refer to the SFB 884 website for dates and times.
You should be familiar with the basics of regression models and maximum likelihood estimation. No previous knowledge of software for Bayesian inference is required. However, we will use R as a front-end to JAGS and for graphical displays. Resources to learn R basics are UCLA’s Stat Consulting Site as well as the official manuals at http: //www.r-project.org/.
This course is intended as an introduction to applied Bayesian statistics for social scientists. Bayesian analysis involves two key aspects – inference based on probability theory and estimation using stochastic simulation. The course will spend some time on the basic principles of both aspects (Where do Bayesian priors come from? How does MCMC work?) and then apply them to some of the workhorse models of the social sciences: linear and discrete outcome models as well as their hierarchical counterparts. We will also discuss practical issues of applied Bayesian analysis, such as MCMC convergence diagnostics as well as Bayesian model checking and parameter summary.
Course readings
Workshop | |||||||
18.10.18 – 06.12.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
02.11.18 – 07.12.18 | Friday | 10:15 – 13:30 | 211 | ||||
The course offers an overview and several hands-on experiences on some of the most relevant methods and tools developed in the field of natural language processing, which have been often adopted as basis for quantitative content analyses in social science research. Attention is dedicated to tasks such as collection building, topic modeling, sentiment analysis, text classification, clustering and scaling as well as to the application of methods such as latent Dirichlet allocations, word embeddings and entity linking. A brief introduction to practices such as web scraping and text pre-processing (e.g. tokenisation, part-of-speech tagging, lemmatisation and stemming) is also offered.
The programming language adopted is Python (and in particular the use of Jupyter Notebooks). No previous programming experience is needed.
Further information is available on Federico's web page.
A maximum of 10 ECTS can be obtained for the successful completion of this course.
6 ECTS written exam
4 ECTS coding exercise
Seminar | |||||||
not on 3, 24 and 31 October | 05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | C 108 Methods lab in A5, 6 entrance C | Link | ||
01.10.18 | Monday | 10:15 – 11:45 | 209 in B6, 30–32 | ||||
17.10.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
07.11.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
Sound understanding of linear regression models (OLS), knowledge in linear algebra and calculus, and being familiar with the statistical package Stata.
The main focus lies on the introduction to statistical models and estimators beyond linear regression useful to a social scientists. A good understanding of the classical linear regression model is a prerequisite and required for the further topics of the course. We will first discuss violations of the asymptotic properties of the linear regression model and ways to address these violations (heteroscedasticity, endogeneity, proxy variables, IV-estimator). The second part of the class is dedicated to rst the maximum likelihood estimator and second to generalized linear models (GLS) for binary choice decisions (Logit, Probit), ordinal dependent variables, and count data (Poisson, Negative Binomial). Classes will be accompanied by lab sessions to repeat and practice the topics from the classes. We will use the statistical package Stata.
Assessment type: written exam (90 min)
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 13:45 – 15:15 | 008.2 in B 6, 30–32 | ||||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 15:30 – 17:00 | C-108 in A5,6 entrance C | ||||
05.09.18 – 05.12.18 | Wednesday | 13:45 – 15:15 | C-108 in A5,6 entrance C |
Depending on teaching needs and room availabilities, lectures and tutorials may be swapped, such that two lectures or two tutorial take place on one day within the time slots allocated to the course, i.e. on Tuesdays, 8.30 – 11.45. Such swaps will be announced before the beginning of the semester.
Summary
This course gives an overview of data used in political science and their measurement properties. At the beginning of the course we will focus on survey data and traditional statistics and move to data science approaches for big data towards the end of the course. Topics covered include the Total Survey Error (TSE) framework, operationalizing research questions, guidelines for writing survey questions, testing questions with cognitive interviews and eye-tracking, sampling, coverage, and nonresponse of survey and big data, and data analytics approaches in data science.
Zu erbringende Prüfungsleistung
Credits for the lectures are obtained through a graded term paper. The term paper covers selected topics from the complete course and has to be submitted by ...tbc. Term papers can be written individually or in groups of max. 3 students/
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 08:30 – 10:00 | A 102 in B6, 23–25 entrance A | Link | |||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | C-108 in A 5, 6 entrance C | Link |
Basic knowledge of probability and statistics.
Machine Learning is about designing algorithms that are able to make predictions about data or extract knowledge from data. The aim of this module is to study algorithms, underlying concepts, and theoretic principles that allow for algorithms to automatically learn how to make predictions. The course focuses on selected “hot topics” and their applications, which include:
• Basics of machine learning and probability theory
• Probabilistic graphical models
• Inference and parameter estimation
• Neural networks
Methods: The course consists of a lecture accompanied by theoretical and practical exercises as well as case studies with real data. In the exercises, students will deepen the material discussed in the lecture, apply the methods in practice, and present the result.
Literature:
• K.P. Murphy. Machine Learning: A Probabilistic Perspective, The MIT Press, 2012
• D. Koller, N. Friedman. Probabilistic graphical models. The MIT Press, 2009
• Additional material and articles provided in lecture notes
Form of assessment: Oral or written examination, homework assignments
Workload Hours per semester: 56h (4 SWS)
Self-study per semester: 98 h
• 70 h: pre- and post lecture studying and revision
• 28 h: exam preparation
Lecture | |||||||
Lecture & Exercises | 04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | A 101 in B 6, 23–25 entrance A | Link | ||
Lecture & Exercises | 05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | A 101 in B 6, 23–25 entrance A | |||
The course introduces students to quantitative methods in political science. During the first half of the course, we will focus on linear regression models. The topics covered include discussions of the mathematical bases for such models, their estimation and interpretation, model assumptions and techniques for addressing violations of those assumptions, and topics related to model specification and functional forms. During the second half of the course, students will be introduced to likelihood as a theory of inference, including models for binary and count data.
The main goals of this course are to develop sound critical judgment about quantitative studies of political problems, to understand the logic of statistical inference, to recognize and understand the basics of the linear regression model, to develop the skills necessary to work with datasets to perform basic quantitative analyses, and to provide a basis of knowledge for more advanced statistical methods.
In the accompanying course “Tutorial Multivariate Analyses” students will develop the necessary expertise in using statistical software to conduct quantitative research in political science.
Graded assignments include homeworks, a mid-term exam and data analysis projects.
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 08:30 – 10:00 | B244 in A5,6 entrance B | Link | |||
Tutorial | |||||||
Neunhoeffer, Marcel | 06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | Room C -108 (PC Lab) in A5,6, entrance C | |||
Sternberg, Sebastian | 10.09.18 – 03.12.18 | Monday | 12:00 – 13:30 | Room C -108 (PC Lab) in A5,6, entrance C |
In der Vorlesung sollen grundlegende Konzepte zur Behandlung von linearen und nichtlinearen Optimierungsaufgaben erarbeitet werden. Ziel ist es, Lösungsverfahren algorithmisch umzusetzen und nebenbei die Algorithmen in einen sicheren mathematischen Rahmen einzubetten. Das Hauptaugenmerk liegt hierbei auf folgenden Gebieten:
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 426154:15 – 11:45 | B 243 in A 5, 6 entrance B | ||||
05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | B 144 in A 5, 6 entrance B | ||||
Tutorial | |||||||
06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | C 013 in A5, 6 entrance C | ||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | C 012 in A5, 6 entrance C | ||||
07.09.18 – 07.12.18 | Friday | 12:00 – 13:30 | C 015 in A5, 6 entrance C |
tbd
Lecture | |||||||
Lecture & Exercises | 06.09.18 – 06.12.18 | Thursday | 10:15 – 13:30 | C -109 PC Pool in A 5, 6 entrance C | Link | ||
The software R is a computer programming language designed for statistical analysis and graphics. The first part of the course deals with a basic introduction to R, i.e. data handling, basic statistical analyses, the creation of graphics, and linear modeling including test for specially designed hypotheses. In the second part we use R as a programming language for cognitive modeling. We will simulate data based on mathematical models of cognitive functions and analyze these data with maximum likelihood parameter estimation techniques. At the end, I will introduce some advanced techniques, for example the creation of statistical reports with R.
The software package R is free and available on all major platforms (www.r-project.org). I also recommend the free and platform independent Software RStudio as a comfortable IDE for R (www.rstudio.com). A basic introduction to R can be found under:
http://cran.r-project.org/doc/manuals/r-release/R-intro.pdf.
Literature will be given during the course
Course achievement – regular participation in the course; non-graded test
Academic assessment – graded homework
Seminar | |||||||
biweekly | 07.09.18 – 30.11.18 | Friday | 13:45 – 17:00 | EO 162, CIP-Pool | Link | ||
Knowledge of multiple linear regression and some familiarity with generalised linear models, to the level of Freedman (2005, see below) or equivalent. Familiarity with notions of research design in the social sciences and the statistical package Stata or R.
If you need to review material on regression models, please consult this excellent textbook: Freedman, David. 2005. Statistical Models: Theory and Practice. Cambridge University Press.
This course teaches the fundamental concepts behind the estimation of causal effects, including potential obstacles to causal inference, faulty measurement, spuriousness, specification errors, and other problems that can lead to inappropriate causal inferences. We will discuss the benefits and the difficulties of randomization in survey research in the first half of the class. The focus of the second half is on the design of observational studies. Real-world examples will be discussed, with an emphasis on examples from survey methodology. Students will come away with an understanding of how to estimate causal effects in both randomized and observational settings, with a particular focus on careful design of both types of studies.
There will be several assignments/
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | B 243 in A5, 6 entrance B | Link | |||
Tutorial | |||||||
13.09.18 – 06.12.18 | Thursday | 15:30 – 17:00 | A 102 in B6, 23–25 entrance A |
Further SMiP courses open to CDSS doctoral students are:
Hierarchical Bayesian Models with Computer Applications (Instructor: Jeffrey Rouder, Date: 01.10. (10:00 – 18:00) and 02.10.2018 (09:00 – 17:00), Location: Mannheim)
Analysis of Response Time (Instructor: Trisha van Zandt, Date: 16.10. (10:00 – 18:00) and 17.10. (09:00 – 17:00), Location: Mannheim)
An Introduction to modern R, Statistical Modeling, and Mixed Models (Instructor: Henrik Singmann, Date: 15.01. and 16.01.2019, Location: Freiburg)
This course gives an advanced overview of standard multivariate methods and current developments in multivariate modeling. The topics include general and generalized linear models, structural equation models, multilevel modeling and specific models for the analysis of time-related effects. In the morning sessions, we will provide the formal foundations and theoretical background of the model classes for continuous and discrete observations. The afternoon sessions focus on empirical applications and hands-on exercises in data analysis.
The goals are to bring the PhD candidates to a common level of statistical knowledge and data analytic skills and to set the stage for the more specialized topics in the Foundations 2 course and workshops in the following semesters.
24.10.2018: 11:00 – 14:00 in C-107 in A5, 6 15:00 – 18:00 in Schloss, EO 162
25.10.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
08.11.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
09.11.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
06.12.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
07.12.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
11.01.2019: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
Further details can be found on the web page of the RTG 'Statistical Modeling in Psychology'
In the course, an introduction in structural equation modeling with software Mplus and R is provided. The students are introduced in the usage of the software and fundamentals of structural equation modeling. The topics such as path analysis, confirmatory factor analysis, measurement invariance and structural regression models are covered. The students work with existing survey data and conduct analyses for the selected research questions. They present both, statistical bases and results of their analyses.
Required examinations: oral presentation and written paper (5000 words)
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | EO 259 (Schloss, Ehrenhof Ost) | Link | |||
The objective of this course is to provide students with the basics of formal modeling in political science. The course has some breadth in coverage in the sense that it provides a graduate-level introduction and overview to di erent areas in game theory. It is also narrow in the sense that the emphasis is not on application and model testing but getting trained in reading and writing down formal models. At the conceptual level the course will cover the following topics: normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signaling games, preferences and individual choices, basics of decision theory and social choice. At the substantial level, we will use these concepts to study, as examples, candidate competition, political lobbying, and war and deterrence.
Literature
tba
Assessment: mid-term and final exam
Tutorial
This tutorial accompanies the graduate-level introductory lecture in game theory. Its main objective is to practice solution concepts for static and dynamic games of complete and incomplete information.The contents are centered around the material covered in the lecture. Thus, the following key areas will be discussed: preferences and individual choices, decision theory, normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signalling games. At the substantial level, we will use these concepts to study, for instance, candidate competition, political lobbying, and war and deterrence. Students are required to submit weekly problem sets. Moreover, active participation in class discussions is expected.
Lecture | |||||||
03.09.18 – 03.12.18 | Monday | 10:15 – 11:45 | B 317 in A5,6 entrance B | Link | |||
Tutorial | |||||||
05.09.18 – 05.12.18 | Wednesday | 17:15 – 18:45 | B 317 in A5,6 entrance B |
OpenSesame is a free, open-source, and cross-platform software for creating laboratory experiments. Many standard tasks can be implemented in OpenSesame via drag and drop using its graphical user interface. In addition, complex tasks can be realized through the underlying programming language Python. The goal of the workshop is to provide an introduction to both approaches. In doing so, the workshop involves both structured input from the instructor as well as a number of practical exercises so that participants can directly explore the features of OpenSesame. Besides, the workshop will introduce plug-ins that extend OpenSesame for specific purposes, e.g., the psynteract plug-ins that implement real-time interactions between participants (as required in many economic games), and the mousetrap plug-ins that implement mouse-tracking during decision tasks (a method that is becoming increasingly popular in the cognitive sciences to measure preference development). Additional topics will be covered depending on the preferences of the workshop participants. No prior knowledge of the software or Python is required.
As an assignment, participants will create their own experiment based on the requirements discussed in the workshop.
Software:
OpenSesame can be downloaded for free under http://osdoc.cogsci.nl/index.html, where you can also find an extensive documentation.
Literature:
Mathôt, S., Schreij, D., & Theeuwes, J. (2012). OpenSesame: An open-source, graphical experiment builder for the social sciences. Behavior Research Methods, 44(2), 314–324. https://dx.doi.org/10.3758/s13428-011-0168-7
Workshop | |||||||
12.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP Pool | Link | |||
13.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
26.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP-Pool | ||||
27.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
This seminar has a particular focus on research activities in social psychology. Unlike seminars that concentrate on one core thematic topic, this seminar will address a selected variety of different research topics in current social psychology. In each seminar session we will have a presentation either by participating students or by members of the social psychology group. Each presentation will address a current research topic in social psychology. The seminar provides the opportunity to actively discuss methodological, theoretical, and applied implications of the presented research. . A particular focus will rest on the discussion of general methodological aspects. Participants need to have passed at least one other social psychology seminar in the master program or, alternatively, they have to be enrolled in one other social psychology seminar during the same semester (for international visiting students these preconditions may differ, please contact the program manager psychology).
Literature: Will be announced in the seminar
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | 309 in B6, 30–32 | Link | |||
Elections are the central focus of political activity in democracies. The characteristics of politics, parties and electoral systems are fundamental to the outcome of elections, which differ across and within countries. To better understand elections we need to study them comparatively, therefore this course focuses on comparative research on elections. The course focuses on the context in which elections are fought and how this affects electoral outcomes. A number of contextual effects of electoral behaviour will be covered, such as institutional configurations, election campaigns, the strategies of political parties and the importance of events in understanding the dynamics of electoral outcomes. We will consider competing theoretical and empirical explanations of the electoral process in democratic as well as partially democratic and even non-democratic countries.
Literature:
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | A305 in B6, 23–25 | Link | |||
Political behavior takes place in context. This statement is a truism and implies several challenges at the same time. Context is a multidimensional concept comprising – inter alia – social, political, and institutional features. At the theoretical level, the diversity of dimensions requires careful consideration of how to integrate contextual features into individual-level models of political behavior. Moreover, combining data from different levels of aggregation to examine the role of contexts in individual-level behavior raises several methodological issues. In this seminar, we will address the conceptual, theoretical, and methodological issues in the analysis of contextual effects on individual-level political behavior. Students will review the latest empirical studies in the field and prepare research papers in which they analyze specific questions using available data sets.
Course requirements: Oral presentation of a literature review and active participation during the sessions.
Academic assessment: Term Paper (ca. 8,000 words)
Seminar | |||||||
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | B 143 in A 5, 6 entrance B | Link | |||
This course will focus on the recent developments in European politics from both a European integration and comparative politics perspective. The first round of sessions concerns the recent history of European integration since the Maastricht Treaty came into force in 1993. This period is characterized by institutional change with the empowering of the European Parliament and the transfer of policy competencies from the national to the EU level, such as monetary, immigration and asylum policies. Another characteristic of this period relate to the economic and debt crisis, the Brexit vote, and the accession of Central and Eastern European countries. The second round of sessions is devoted to the implications of this history for party politics and national populism in the member states. In addition to the European elections, particular attention will be on cross-country analyses of national elections and public support. Further topics include the rise of Euroskeptic parties, which enter into governmental office.
Participants are expected to write a term paper, preferably with an empirical study on these topics. The main goals of the course are to sharpen analytical, presentation, and writing skills with a focus on the interplay between European integration and the recent developments in party politics and national populism in the member states.
Literature
Seminar | |||||||
06.09.18 – 06.12.18 | Thursday | 12:00 – 13:00 | 309 in B 6, 30–32 | Link | |||
This one-semester course introduces students to some of the major topics related to the study of re-distributive politics in the context of economic interdependence. It examines from an international and comparative political economy perspective why some governments in open economies are more re-distributive than others. It covers several key debates about the role of trade and investment on material and other interests, institutions and political parties. Following a brief overview of patterns of openness and redistribution, we start by asking whether and to what extent globalization increases the economic costs and benefits of re-distributive policies. We then turn to those factors that shape the political feasibility of redistribution. We begin by presenting standard models of international economics as well as redistribution. We examine how voters form preferences for redistribution in open economies, and extend this framework by exploring the role of partisanship, organised interests and electoral institutions. A final part looks at economic integration as a dependent variable, and considers the links between compensation strategies and the decision to open the economy.
Seminar | |||||||
10.09.18 – 03.12.18 | Monday | 15:30 – 17:00 | B 318 in A5, 6 entrance B | ||||
This seminar focuses on the outbreak and the dynamics of violent conflict as well as ways to end wars and sustain peace and stability. We will start by looking at the concepts of “peace” and “war” and examine empirical trends. Is the world more violent or more peaceful than in the past? How can we measure peace and conflict? In the following we will take a closer look at the relationship between religion and war, as well as economics and war. Does religion cause war? Or are economic factors more important for explaining why people fight? We will devote sessions to study terrorism as an alternative way of fighting (as opposed to conventional and civil wars), migration as a source of peace and conflict and the various forms of violence used by different militant groups. Why are some groups so much more violent than others? What is the rational behind using different forms of violence? Finally we will have a look at empirical evidence concerning ways to end conflicts and sustain peace. Are peace missions for example an effective tool to make and sustain peace? Does development assistance foster development?
Course requirement: Class presentation & leading the class discussion.
Academic assessment: Research paper
Seminar | |||||||
24.09.18 | Monday | 15:30 – 17:00 | A 204 in B6, 23–25 | ||||
16.11.18 – 18.11.18 | Friday to Sunday | 08:30 – 17:00 | A 204 in B6, 23–25 | ||||
The course “Current Research Perspectives” introduces first year doctoral students to the theoretically informed research approaches and substantive research fields that build the strongholds of social science research in Mannheim. A series of talks provides first year doctoral students with an overview of current debates and ongoing research in the fields of psychology, political science and sociology. CDSS faculty members will present an overview of their research fields, report on prime examples of their current research, and provide an outlook on potential topics for future research. Doctoral students will have the opportunity to discuss the short talks with the respective lecturer during the remaining discussion time.
Homework assignment (3 pages)
Lecture | |||||||
04.09.18 – 11.09.18 | Tuesday | 13:45 – 17:00 | 211 in B6, 30–32 | Link | |||
12.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
26.09.18 | Wednesday | 10:15 – 13:30 | 212 in B6, 30–32 | ||||
It is increasingly important for modern social scientists to have a level of mathematical literacy, as mathematical research methods such as statistics and formal modelling have entered the main stream. This course is intended to provide an introduction to mathematical logic and rigour, and to some fundamental mathematical concepts that form the foundation of the modern subject. The course covers introductory set and function theory, including analysis of functions, and includes sections on both probability and linear algebra, which together are the basis of data analysis.
The exam is scheduled for 11 December at 9.30am in room 108 in B6, 30–32.
Basic readings:
Additional readings:
Lecture | |||||||
05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | 211 in B6, 30–32 | ||||
All researchers face similar challenges with core issues of research design. A research design is a plan that specifies how you are going to carry out a research project and, particularly, how to use evidence to answer your research question. The goal of this course is to jump-start students with their dissertation proposal. This course should help students to see the trade-offs involved in choosing a particular research design in their research projects. Consequently students are expected to develop own ideas about potential research questions and actively participate in those seminar-style meetings that are organized within this lecture course.
Workshop | |||||||
04.09.18 – 04.12.18 | Tuesday | 12:00 – 13:30 | A 203 in B6, 23–25 | Link | |||
Questions of cause and effect are at the heart of social science. And yet, establishing credible causal effects in empirical analyses is a difficult enterprise. This course will introduce some of the key conceptual and methodological approaches to tackle the causal inference problem: the potential outcomes model of causal inference, experimental designs, matching and regression, instrumental variables, regression discontinuity designs as well as difference-in-differences and fixed effects.
Workshop | |||||||
06.09.18 – 11.10.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
not on 12 Oct | 07.09.18 – 19.10.18 | Friday | 10:15 – 13:30 | 310 in B6, 30–32 | |||
Please check with individual chairs in the Psychology Department for dates and times of research colloquia as well as registration.
CSSR, TBCI, Dissertation Proposal Workshop
Research in Cognitive Psychology: Research projects in cognitive psychology and neuropsychology are planned, conducted, analyzed, and discussed.
Application via 'Studierendenportal' is necessary to have access to the course material provided in ILIAS.
Open office hours:
Prof. Dr. Erdfelder: Thursday, 10.15h – 11.45h.
Literature: References will be given during the course.
Improvement in research skills and communication of research results.
Workshop | |||||||
03.09.18 – 03.12.18 | Monday | 15:30 – 17:00 | EO 259 | Link | |||
You should be familiar with the basics of regression models and maximum likelihood estimation. No previous knowledge of software for Bayesian inference is required. However, we will use R as a front-end to JAGS and for graphical displays. Resources to learn R basics are UCLA’s Stat Consulting Site as well as the official manuals at http: //www.r-project.org/.
This course is intended as an introduction to applied Bayesian statistics for social scientists. Bayesian analysis involves two key aspects – inference based on probability theory and estimation using stochastic simulation. The course will spend some time on the basic principles of both aspects (Where do Bayesian priors come from? How does MCMC work?) and then apply them to some of the workhorse models of the social sciences: linear and discrete outcome models as well as their hierarchical counterparts. We will also discuss practical issues of applied Bayesian analysis, such as MCMC convergence diagnostics as well as Bayesian model checking and parameter summary.
Course readings
Workshop | |||||||
18.10.18 – 06.12.18 | Thursday | 12:00 – 15:15 | 211 in B6, 30–32 | ||||
02.11.18 – 07.12.18 | Friday | 10:15 – 13:30 | 211 | ||||
The course offers an overview and several hands-on experiences on some of the most relevant methods and tools developed in the field of natural language processing, which have been often adopted as basis for quantitative content analyses in social science research. Attention is dedicated to tasks such as collection building, topic modeling, sentiment analysis, text classification, clustering and scaling as well as to the application of methods such as latent Dirichlet allocations, word embeddings and entity linking. A brief introduction to practices such as web scraping and text pre-processing (e.g. tokenisation, part-of-speech tagging, lemmatisation and stemming) is also offered.
The programming language adopted is Python (and in particular the use of Jupyter Notebooks). No previous programming experience is needed.
Further information is available on Federico's web page.
A maximum of 10 ECTS can be obtained for the successful completion of this course.
6 ECTS written exam
4 ECTS coding exercise
Seminar | |||||||
not on 3, 24 and 31 October | 05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | C 108 Methods lab in A5, 6 entrance C | Link | ||
01.10.18 | Monday | 10:15 – 11:45 | 209 in B6, 30–32 | ||||
17.10.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
07.11.18 | Wednesday | 15:30 – 17:00 | EO 162 CIP-Pool | ||||
Sound understanding of linear regression models (OLS), knowledge in linear algebra and calculus, and being familiar with the statistical package Stata.
The main focus lies on the introduction to statistical models and estimators beyond linear regression useful to a social scientists. A good understanding of the classical linear regression model is a prerequisite and required for the further topics of the course. We will first discuss violations of the asymptotic properties of the linear regression model and ways to address these violations (heteroscedasticity, endogeneity, proxy variables, IV-estimator). The second part of the class is dedicated to rst the maximum likelihood estimator and second to generalized linear models (GLS) for binary choice decisions (Logit, Probit), ordinal dependent variables, and count data (Poisson, Negative Binomial). Classes will be accompanied by lab sessions to repeat and practice the topics from the classes. We will use the statistical package Stata.
Assessment type: written exam (90 min)
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 13:45 – 15:15 | 008.2 in B 6, 30–32 | ||||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 15:30 – 17:00 | C-108 in A5,6 entrance C | ||||
05.09.18 – 05.12.18 | Wednesday | 13:45 – 15:15 | C-108 in A5,6 entrance C |
Depending on teaching needs and room availabilities, lectures and tutorials may be swapped, such that two lectures or two tutorial take place on one day within the time slots allocated to the course, i.e. on Tuesdays, 8.30 – 11.45. Such swaps will be announced before the beginning of the semester.
Summary
This course gives an overview of data used in political science and their measurement properties. At the beginning of the course we will focus on survey data and traditional statistics and move to data science approaches for big data towards the end of the course. Topics covered include the Total Survey Error (TSE) framework, operationalizing research questions, guidelines for writing survey questions, testing questions with cognitive interviews and eye-tracking, sampling, coverage, and nonresponse of survey and big data, and data analytics approaches in data science.
Zu erbringende Prüfungsleistung
Credits for the lectures are obtained through a graded term paper. The term paper covers selected topics from the complete course and has to be submitted by ...tbc. Term papers can be written individually or in groups of max. 3 students/
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 08:30 – 10:00 | A 102 in B6, 23–25 entrance A | Link | |||
Tutorial | |||||||
04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | C-108 in A 5, 6 entrance C | Link |
Basic knowledge of probability and statistics.
Machine Learning is about designing algorithms that are able to make predictions about data or extract knowledge from data. The aim of this module is to study algorithms, underlying concepts, and theoretic principles that allow for algorithms to automatically learn how to make predictions. The course focuses on selected “hot topics” and their applications, which include:
• Basics of machine learning and probability theory
• Probabilistic graphical models
• Inference and parameter estimation
• Neural networks
Methods: The course consists of a lecture accompanied by theoretical and practical exercises as well as case studies with real data. In the exercises, students will deepen the material discussed in the lecture, apply the methods in practice, and present the result.
Literature:
• K.P. Murphy. Machine Learning: A Probabilistic Perspective, The MIT Press, 2012
• D. Koller, N. Friedman. Probabilistic graphical models. The MIT Press, 2009
• Additional material and articles provided in lecture notes
Form of assessment: Oral or written examination, homework assignments
Workload Hours per semester: 56h (4 SWS)
Self-study per semester: 98 h
• 70 h: pre- and post lecture studying and revision
• 28 h: exam preparation
Lecture | |||||||
Lecture & Exercises | 04.09.18 – 04.12.18 | Tuesday | 10:15 – 11:45 | A 101 in B 6, 23–25 entrance A | Link | ||
Lecture & Exercises | 05.09.18 – 05.12.18 | Wednesday | 08:30 – 10:00 | A 101 in B 6, 23–25 entrance A | |||
The course introduces students to quantitative methods in political science. During the first half of the course, we will focus on linear regression models. The topics covered include discussions of the mathematical bases for such models, their estimation and interpretation, model assumptions and techniques for addressing violations of those assumptions, and topics related to model specification and functional forms. During the second half of the course, students will be introduced to likelihood as a theory of inference, including models for binary and count data.
The main goals of this course are to develop sound critical judgment about quantitative studies of political problems, to understand the logic of statistical inference, to recognize and understand the basics of the linear regression model, to develop the skills necessary to work with datasets to perform basic quantitative analyses, and to provide a basis of knowledge for more advanced statistical methods.
In the accompanying course “Tutorial Multivariate Analyses” students will develop the necessary expertise in using statistical software to conduct quantitative research in political science.
Graded assignments include homeworks, a mid-term exam and data analysis projects.
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 08:30 – 10:00 | B244 in A5,6 entrance B | Link | |||
Tutorial | |||||||
Neunhoeffer, Marcel | 06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | Room C -108 (PC Lab) in A5,6, entrance C | |||
Sternberg, Sebastian | 10.09.18 – 03.12.18 | Monday | 12:00 – 13:30 | Room C -108 (PC Lab) in A5,6, entrance C |
In der Vorlesung sollen grundlegende Konzepte zur Behandlung von linearen und nichtlinearen Optimierungsaufgaben erarbeitet werden. Ziel ist es, Lösungsverfahren algorithmisch umzusetzen und nebenbei die Algorithmen in einen sicheren mathematischen Rahmen einzubetten. Das Hauptaugenmerk liegt hierbei auf folgenden Gebieten:
Lecture | |||||||
04.09.18 – 04.12.18 | Tuesday | 426154:15 – 11:45 | B 243 in A 5, 6 entrance B | ||||
05.09.18 – 05.12.18 | Wednesday | 10:15 – 11:45 | B 144 in A 5, 6 entrance B | ||||
Tutorial | |||||||
06.09.18 – 06.12.18 | Thursday | 10:15 – 11:45 | C 013 in A5, 6 entrance C | ||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | C 012 in A5, 6 entrance C | ||||
07.09.18 – 07.12.18 | Friday | 12:00 – 13:30 | C 015 in A5, 6 entrance C |
tbd
Lecture | |||||||
Lecture & Exercises | 06.09.18 – 06.12.18 | Thursday | 10:15 – 13:30 | C -109 PC Pool in A 5, 6 entrance C | Link | ||
The software R is a computer programming language designed for statistical analysis and graphics. The first part of the course deals with a basic introduction to R, i.e. data handling, basic statistical analyses, the creation of graphics, and linear modeling including test for specially designed hypotheses. In the second part we use R as a programming language for cognitive modeling. We will simulate data based on mathematical models of cognitive functions and analyze these data with maximum likelihood parameter estimation techniques. At the end, I will introduce some advanced techniques, for example the creation of statistical reports with R.
The software package R is free and available on all major platforms (www.r-project.org). I also recommend the free and platform independent Software RStudio as a comfortable IDE for R (www.rstudio.com). A basic introduction to R can be found under:
http://cran.r-project.org/doc/manuals/r-release/R-intro.pdf.
Literature will be given during the course
Course achievement – regular participation in the course; non-graded test
Academic assessment – graded homework
Seminar | |||||||
biweekly | 07.09.18 – 30.11.18 | Friday | 13:45 – 17:00 | EO 162, CIP-Pool | Link | ||
Knowledge of multiple linear regression and some familiarity with generalised linear models, to the level of Freedman (2005, see below) or equivalent. Familiarity with notions of research design in the social sciences and the statistical package Stata or R.
If you need to review material on regression models, please consult this excellent textbook: Freedman, David. 2005. Statistical Models: Theory and Practice. Cambridge University Press.
This course teaches the fundamental concepts behind the estimation of causal effects, including potential obstacles to causal inference, faulty measurement, spuriousness, specification errors, and other problems that can lead to inappropriate causal inferences. We will discuss the benefits and the difficulties of randomization in survey research in the first half of the class. The focus of the second half is on the design of observational studies. Real-world examples will be discussed, with an emphasis on examples from survey methodology. Students will come away with an understanding of how to estimate causal effects in both randomized and observational settings, with a particular focus on careful design of both types of studies.
There will be several assignments/
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 13:45 – 15:15 | B 243 in A5, 6 entrance B | Link | |||
Tutorial | |||||||
13.09.18 – 06.12.18 | Thursday | 15:30 – 17:00 | A 102 in B6, 23–25 entrance A |
Further SMiP courses open to CDSS doctoral students are:
Hierarchical Bayesian Models with Computer Applications (Instructor: Jeffrey Rouder, Date: 01.10. (10:00 – 18:00) and 02.10.2018 (09:00 – 17:00), Location: Mannheim)
Analysis of Response Time (Instructor: Trisha van Zandt, Date: 16.10. (10:00 – 18:00) and 17.10. (09:00 – 17:00), Location: Mannheim)
An Introduction to modern R, Statistical Modeling, and Mixed Models (Instructor: Henrik Singmann, Date: 15.01. and 16.01.2019, Location: Freiburg)
This course gives an advanced overview of standard multivariate methods and current developments in multivariate modeling. The topics include general and generalized linear models, structural equation models, multilevel modeling and specific models for the analysis of time-related effects. In the morning sessions, we will provide the formal foundations and theoretical background of the model classes for continuous and discrete observations. The afternoon sessions focus on empirical applications and hands-on exercises in data analysis.
The goals are to bring the PhD candidates to a common level of statistical knowledge and data analytic skills and to set the stage for the more specialized topics in the Foundations 2 course and workshops in the following semesters.
24.10.2018: 11:00 – 14:00 in C-107 in A5, 6 15:00 – 18:00 in Schloss, EO 162
25.10.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
08.11.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
09.11.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
06.12.2018: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
07.12.2018: 09:00 – 12:00 in Schloss, EO 162 13:00 – 16:00 in Schloss, EO 162
11.01.2019: 10:00 – 13:00 in Schloss, EO 162 14:00 – 17:00 in Schloss, EO 162
Further details can be found on the web page of the RTG 'Statistical Modeling in Psychology'
In the course, an introduction in structural equation modeling with software Mplus and R is provided. The students are introduced in the usage of the software and fundamentals of structural equation modeling. The topics such as path analysis, confirmatory factor analysis, measurement invariance and structural regression models are covered. The students work with existing survey data and conduct analyses for the selected research questions. They present both, statistical bases and results of their analyses.
Required examinations: oral presentation and written paper (5000 words)
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | EO 259 (Schloss, Ehrenhof Ost) | Link | |||
The objective of this course is to provide students with the basics of formal modeling in political science. The course has some breadth in coverage in the sense that it provides a graduate-level introduction and overview to di erent areas in game theory. It is also narrow in the sense that the emphasis is not on application and model testing but getting trained in reading and writing down formal models. At the conceptual level the course will cover the following topics: normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signaling games, preferences and individual choices, basics of decision theory and social choice. At the substantial level, we will use these concepts to study, as examples, candidate competition, political lobbying, and war and deterrence.
Literature
tba
Assessment: mid-term and final exam
Tutorial
This tutorial accompanies the graduate-level introductory lecture in game theory. Its main objective is to practice solution concepts for static and dynamic games of complete and incomplete information.The contents are centered around the material covered in the lecture. Thus, the following key areas will be discussed: preferences and individual choices, decision theory, normal form games, Nash equilibria, extensive form games, subgame perfect equilibria, repeated games, bargaining, games with incomplete and imperfect information, Bayesian perfect equilibria, signalling games. At the substantial level, we will use these concepts to study, for instance, candidate competition, political lobbying, and war and deterrence. Students are required to submit weekly problem sets. Moreover, active participation in class discussions is expected.
Lecture | |||||||
03.09.18 – 03.12.18 | Monday | 10:15 – 11:45 | B 317 in A5,6 entrance B | Link | |||
Tutorial | |||||||
05.09.18 – 05.12.18 | Wednesday | 17:15 – 18:45 | B 317 in A5,6 entrance B |
OpenSesame is a free, open-source, and cross-platform software for creating laboratory experiments. Many standard tasks can be implemented in OpenSesame via drag and drop using its graphical user interface. In addition, complex tasks can be realized through the underlying programming language Python. The goal of the workshop is to provide an introduction to both approaches. In doing so, the workshop involves both structured input from the instructor as well as a number of practical exercises so that participants can directly explore the features of OpenSesame. Besides, the workshop will introduce plug-ins that extend OpenSesame for specific purposes, e.g., the psynteract plug-ins that implement real-time interactions between participants (as required in many economic games), and the mousetrap plug-ins that implement mouse-tracking during decision tasks (a method that is becoming increasingly popular in the cognitive sciences to measure preference development). Additional topics will be covered depending on the preferences of the workshop participants. No prior knowledge of the software or Python is required.
As an assignment, participants will create their own experiment based on the requirements discussed in the workshop.
Software:
OpenSesame can be downloaded for free under http://osdoc.cogsci.nl/index.html, where you can also find an extensive documentation.
Literature:
Mathôt, S., Schreij, D., & Theeuwes, J. (2012). OpenSesame: An open-source, graphical experiment builder for the social sciences. Behavior Research Methods, 44(2), 314–324. https://dx.doi.org/10.3758/s13428-011-0168-7
Workshop | |||||||
12.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP Pool | Link | |||
13.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
26.10.18 | Friday | 10:15 – 15:15 | EO 162 CIP-Pool | ||||
27.10.18 | Saturday | 10:15 – 17:00 | EO 162 CIP-Pool | ||||
This seminar has a particular focus on research activities in social psychology. Unlike seminars that concentrate on one core thematic topic, this seminar will address a selected variety of different research topics in current social psychology. In each seminar session we will have a presentation either by participating students or by members of the social psychology group. Each presentation will address a current research topic in social psychology. The seminar provides the opportunity to actively discuss methodological, theoretical, and applied implications of the presented research. . A particular focus will rest on the discussion of general methodological aspects. Participants need to have passed at least one other social psychology seminar in the master program or, alternatively, they have to be enrolled in one other social psychology seminar during the same semester (for international visiting students these preconditions may differ, please contact the program manager psychology).
Literature: Will be announced in the seminar
Seminar | |||||||
03.09.18 – 03.12.18 | Monday | 13:45 – 15:15 | 309 in B6, 30–32 | Link | |||
This lecture series will present cutting edge research conducted in cognitive psychology at the University of Mannheim. After an introductory overview of cognitive psychology and its advanced methods (i.e., cognitive modeling) by B. Kuhlmann, various researchers will present their current work including research on judgment and decision making, memory, metacognition, and cognitive aging. The following researchers are planned as lecturers (changes possible): Dr. Nina Arnold, Dr. Martin Brandt, Prof. A. Bröder, Prof. E. Erdfelder, Dr. Michael Gräf, Dr. Meike Kroneisen, Dr. Lena Naderevic, Prof. Rüdiger Pohl und Dr. Monika Undorf. All lectures and materials will be in English.
Final test: written exam.
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 15:30 – 17:00 | EO 145 | Link | |||
Knowledge in work and organizational psychology (as acquired during bachelor studies). It is expected that students know the content of a text book such as Spector (2008) or Landy & Conte (2010).
This course provides an overview of core topic within work and organizational psychology. We will focus on recent theoretical approaches and empirical research findings (meta-analyses). In addition, we will discuss practical implications of core research findings. Topics include: Work motivation, stress and health, leadership, teams, personnel selection.
Methods comprise: Lecture, reading (as homework), teamwork assignments during class.
Literature
Journal papers; reading assignments will be given at the beginning of the semester.
Lecture | |||||||
06.09.18 – 06.12.18 | Thursday | 17:15 – 18:45 | EO 242 | Link | |||