After completing the course, you will:
- have gained insight into the use and theory of regression analysis in the context of theory-testing political science research
- know how to interpret (simple and multiple) linear and logistic regression models
- know how to interpret these models when they include continuous and categorical predictor variables
- know how to interpret these models when they include non-linear effects (through the inclusion of polynomial, exponential and logarithmic functions)
- know how to interpret these models when they are used for moderation and mediation analysis
- have gained insight into the benefits and limitations of regression analysis for making inferential statements, including statements about causality
- understand how these limitations are related to uncertainty in (frequentist) inferential statistics in general, and to the assumptions associated with regression models in particular
- have practical knowledge of the application of different types of regression analysis and related data challenges such as dealing with outliers and collinearity
- know how to practically prepare your data and execute various types of regression analysis in scientifically reproducable manner with syntax commands
- be able to interpret the computer output of the relevant statistics
- know how to present your findings in clear and well-structured tables
Regression analysis is an incredibly widely used method of inquiry in many scientific fields, including political science. Flip through the issues of many major scientific journals, and you will commonly encounter some type of regression analysis as the method of choice in the articles. Even scientific studies that do not use regression methods often still require some understanding of regression methods. This is not only true for other statistical methods, but also for many so-called qualitative research methods. For instance, strategies for case-selection in case study research are sometimes justified by regression-based reasoning. Other qualitative research methods are sometimes justified explicitly as alternatives to regression analysis, and thus require a good understanding of the strengths and weaknesses of this method. In short, it is nowadays virtually impossible to become a scientist in our field without a good understanding of regression analysis. The method is also widely used in applied research and is frequently is employed as a basis for policy advise, and hence of great importance to professionals concerned with policy choices and policy evaluation.
Political Science Research Methods II (PSRM II) addresses the use of linear and logistic regression analysis in theory-driven research. It teaches you how regression analysis works, how it can be used to tackle substantive research questions, but also what its limitations are. In this way, the course builds on the knowledge and skills that you have acquired in your previous introductory methods and statistics courses. You will apply the knowledge and skills learned in PSRM II directly in the course Project 3: Democracy and Representation.
BA-2/Pre-masters. Intermediate level.
For bachelor students: statistics and methods courses taught in the first year, in particular OIM-A, OIM-B and Project 1.|
For pre-master students: statistics and methods courses taught in your previous studies and in the first semester of the pre-master, in particular Statistics.
Students who lack appropriate foreknowledge should be aware that they will be unlikely to pass PSRM II.
The final grade is determined by a written exam held during exam week; the passing of weekly assignments is a prerequisite for passing the course.|
The written exam consists of multiple choice questions and is graded with chance-correction. It is not allowed to use the book, notes or other additional resources during the written exam. It is allowed to use a simple scientific calculator and a dictionary.
There are 7 mandatory, individual assignments. Each assignment will be graded as either 'pass' or 'fail'. In order to pass the course, it is necessary to have passed at least 6 of these 7 assignments by the end of the course. There will be 1 re-take opportunity for each of the 7 assignments. Additionally, there will be 1 'wildcard' second re-take opportunity, which can be used for 1 of the first 5 assignments. All relevant deadlines for the assignments and re-take opportunities will be listed in the course manual. The results of the assignments will be communicated through Brightspace.
For students that have passed at least 6 of the 7 assignments by the end of the course, the final grade will be determined by the written exam, with a grade >=5.5 representing a passing grade. Students that have not passed at least 6 of the 7 assignments by the end of the course do not pass the course.
The final grade will be registered in Osiris.
Partial results from previous years are not valid.