GESIS - Data Archive for the Social Sciences
Seminar: Mathematical Tools for Social Scientists
September 20 – 24, 2010
Research articles in social science journals and methods textbooks make increasingly use of matrix algebra and calculus. Thus, social scientists who seek to keep up with recent developments in data analysis methodology or want to get an in-depth understanding of common methods often find themselves confronted with mathematical notations they cannot read without special training. On the other hand, new data-analytical algorithms often are – long before they find their way into the more common packages such as SPSS or SAS – first implemented in matrix-oriented software packages such as R, Matlab, GAUSS.
The course ”Mathematical Tools for Social Sciences“ is targeted at social scientists who want to gain an understanding of contemporary methods of quantitative data analysis, but for this purpose need to acquire an understanding of advanced mathematical notations and their implementation in statistical software – or, who intend to refresh their knowledge in these topics. That is, the course aims to familiarize with the mathematical tools of matrix algebra and statistical computation and their application using the software package R. The course will help in preparing for the more specialist courses like GESIS Spring Seminar and other programs for post-graduates, but will also benefit all researchers who just want to keep up with recent developments in research methodology.
Design of the course:
The course will be organized as a mix of theory presentations and a hands-on-workshop. In the hands-on part, we will focus on programming with R. We chose R because it has gradually developed into a “lingua franca” of methodological research and advanced applied statistics. A growing number of courses on advanced statistical data reduction techniques make use of R in their practical lab sessions. There are four reasons for this development: first, R allows translating statistical concepts into software with relative ease, second, there are more than 2300 R extension packages that implement both basic and advanced data-analytical techniques, third, it greatly facilitates the creation of publication-ready, high-quality graphics, and, last but not least, it is free software. While the course heavily relies on practical exercises using R, the emphasis is less on the use of a software package or on programming skills. Rather the use of R is only a means for gaining a practical understanding of data-analytical procedures.
Topics will be:
- Notation and algebra of vectors and matrices
- Probability distributions and random numbers
- Geometry of linear regression and principal components
- Derivatives of matrix and vector functions
- Numerical solution of (non-linear) equations
Lecturer:
- Dr. Martin Elff, Lecturer in European and German Politics, Department of Government, University of Essex, UK
Venue:
- GESIS Leibniz Institute for the Social Sciences, 50931 Cologne, Germany, Conference room, Liliencronstr. 6
Registration:
For your registration, please use the registration form. Participants will be accepted by order of application date, the number of participants is limited to 25 persons.
Accommodation:
Cologne hosts numerous international trade fairs. If you need an accommodation, we strongly recommend to book directly after your registration (we will send some information after registration).
Seminar fee and payment:
The participation fee is 240,00 €. Reductions or refunds are not possible except for students and Ph.D. candidates without work contract (or civil servant status), who will receive a reduction of one third of the fee, if they can officially document their status.
Travel costs, accommodation and meals are not included. Please make your own arrangements.
Funding:
For participants from Germany, it might be of interest that the GESIS Spring Seminar is acknowledged as „Bildungsurlaub“ www.bildungsurlaub.info.
The GESIS Seminar (‘Köln Spring School’) is sponsored by the European Consortium for Political Research (ECPR), University of Essex, UK www.essex.ac.uk/ecpr/events/summerschools/index.aspx.
Please also have a look at the funding web-page of the University of Essex
www.essex.ac.uk/ecpr/funding/mobilityfund/index.aspx.
Contacts:
Address:
GESIS – Leibniz-Institute for the Social Sciences
Dept. Data Archive for the Social Sciences
Bachenmer Str. 40
50931 Koeln
Germany
Scientific Coordinator:
Maria Rohlinger
Phone/Fax: +49 221 47694.45/77
Administrative Coordinator:
Angelika Ruf
Phone/Fax: +49 221 47694.11/77
Email-contact:
springseminar(at)gesis(dot)org
Timetable:
| Tuesday, 21-09-2010 | Wednesday, 22-09-2010 | Thursday, 23-09-2010 | Friday, 24-09-2010 | |
09.00-12.30 | Vectors and matrices – their geometry and elementary algebra; Systems of linear equations and inverse matrices; Data matrices | Probabilities and random numbers; Elementary probability distributions; The multivariate normal distribution | Rotation matrices; Matrix factorization; The geometry of linear regression and principal components | Derivatives of vector and matrix functions; Linear regression as an optimization problem | Numerical approaches to solution of equations and to the optimization of functions |
12.30-15.00 | Break | Break | Break | Break | Break |
15.00-18.00 | Introduction to the R software; Basic computation in R; Matrix algebra in R. | Practicals: generating random numbers and summarizing empirical distributions using graphics | Introduction to regression models in R. Practicals: regression using R standard functions and regression in matrix form | Elementary programming techniques: defining branches, loops and, functions; Debugging | Practicals: solving equations numerically using R; maximizing likelihood functions |
18.00 | Come together party | Joint dinner at a Cologne brewery |
READING LIST:
Basic mathematical tools:
Gill, Jeff. 2006. Essential Mathematics for Political and Social Research. Cambridge: Cambridge University Press.
Advanced mathematical tools:
Harville, David A. 1997. Matrix Algebra From A Statistician's Perspective. New York: Springer.
Introductory texts on R:
Dalgaard, Peter. 2008. Introductory Statistics with R. 2nd edition.New York: Springer.
Maindonald, John and John Braun. 2007. Data Analysis and Graphics Using R. 2nd edition.Cambridge: Cambridge University Press.
Advanced programming and data analysis in R:
Chambers, John M. 2008. Software for Data Analysis: Programming with R. New York: Springer.
Venables, William N. and Brian D. Ripley. 2002. Modern Applied Statistics with S. Fourth Edition. New York: Springer.

