PDF version

Learning Goals

The goal of this course is three-fold:

  1. to prepare students to conduct research using appropriate statistical models and to communicate their results to a non-technical audience;
  2. provide a foundation in the theory of maximum likelihood so students can investigate and implement a wide range of advanced statistical models; and
  3. provide students with the tools necessary to fine-tune existing or to develop new statistical models of political phenomena.

How to Succeed in This Course

  • Work through the assigned readings ahead of time. We expect everyone to come to class fully prepared.
  • Expect that this will take considerably longer than in a substantive seminar. Do not skip equations!
  • Instead, take notes, prepare questions and team-up with others to answer them, or as last resort, ask them in class. After every class we expect you to go over the lecture notes and your notes again.
  • We additionally offer the possibility to send us questions by Wednesday night. We will try to address them in the lab session on Thursday.
  • There is no point in getting lost — particularly not in an elective class. Nevertheless, understand that the bulk of learning in this course will take place outside the classroom, by reading, practicing using statistical software, and solving problem sets.

Graduate students in political science in the M.A. Political Science and CDSS PhD students as well as Mannheim Master in Data Science (MMDS) students. Interested PhD students from other GESS centers can participate subject to the availability of seats.


Master students (M.A. Political Science, MMDS, MMBR) should have successfully passed the previous course in the political science methods sequence Multivariate Analyses and the accompanying Tutorial Multivariate Analyses, preferably with a final grade of 2.0 or better. PhD students should have passed equivalent courses. If you know what $$(X’X)^{-1} X’y$$ is, you have the necessary background to take this class.

Course Registration

Students who wish to take the course should register for the lecture and tutorial at the student portal. Please note, that the course registration is only complete when you are admitted to the ILIAS group of the course. Furthermore, to be able to receive the grades, students are required to register for the examination in both the lecture and the tutorial during the semester, additionally to the course registration.

Note that this course is highly demanding and entails a substantial work load for students! Students who wish to audit this class should notify the instructor in advance (participation is subject to free room capacity). Please note that only registered students will receive feedback on their written work.

Teaching Organization: Online and Offline Teaching

Due to the ongoing COVID-19 pandemic, at least some of the course components will take place online.

  • The lecture will generally take place in person from the beginning of the semester.
  • All office hours will be carried out online via Zoom, until further notice from the University.
  • At least one lab will be carried out in person on campus. Another lab may be carried out online until the spring break. Both labs will be carried out in person after the spring break.
  • Please not that the mode of participation in the class is subject to change, depending on the current pandemic situation and regulations.


We will not use a single textbook for this course. All the reading will be available on the course website (with gated links to ILIAS). The following books will be used in the course:

  • Eliason, Scott R. 1993. Maximum Likelihood Estimation: Logic and Practice. Newbury Park: Sage.

  • King, Gary. 1989. Unifying Political Methodology. Ann Arbor: University of Michigan Press.

  • Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Newbury Park.: Sage.


Students need to bring their own computers to lab sessions. R will be the software package of choice. There will be homework problems that require you to edit and write some R-code. The open-source statistical programming language R is particularly suited for carrying out state-of-the-art computer-based simulations and programming advanced statistical models. It also generates really nice publication-quality graphics. The software runs under a wide array of operating systems. R can be downloaded for free at

A very good graphical user interface for R (which we will also use during the lab sessions) is RStudio. In recent years a growing number of features have been added to this graphical user interface, which makes it the preferred choice for learning R – also for beginners. It is cross-platform and open-source. RStudio can be downloaded for free at A style guide to make your code easier to read, share, and verify can be found at Please make sure to install the latest versions of R and RStudio before the first lab session.

To facilitate an efficient workflow, we integrate Github into the course. git is a version control system that makes it easy to track changes and work on code collaboratively. GitHub is a hosting service for git. You can think of it like a public Dropbox for code. We will use it to distribute code and assignments to you. And you will use it to keep track of your code and collaborate in teams. You can find the course on GitHub here.

Course Requirements

Grading will be based on the following components:

Homework Assignments (25%)

There will be a series of six homework assignments that will take the form of problem sets, replications, simulations, or extensions of the analysis in class and the lab. The assignments will be made available after the class on Wednesday and you are expected to upload a solution by Tuesday 23:59 (unless noted otherwise) a week later. You need to work through every task of the homework assignment, and the material in your Github repositories should be sufficient for replicating by the deadline. Late submissions and non-reproducible files will not be accepted.

You will work in small groups on the assignments. Usually 2–3 people per group works best. Please indicate with whom you worked on the assignment on your homework. Moreover, you are strongly encouraged to seek advice from the instructors during office hours or via Slack, also before the submission deadline. Note that instructive discussions about the material are best done during office hours rather than via Slack.

Final Paper (75 %)

There will be a final draft paper but no final exam. Each student will produce a co-authored manuscript (or a solo-authored manuscript, with permission of the instructor) that applies or develops an appropriate statistical model to an important substantive problem. Students will choose their own topics. What works particularly well is to start replicating an already published article to develop it into a different paper using your own argument. Our advice is to pick an article that interests you, was published within the last few years in a top journal, and uses methods we have or will talk about in class (or uses different methods at about the same level of sophistication).

The draft paper must include all analyses, tables, figures, and descriptions of the results. A good write-up of the draft paper should read like the third quarter of a journal article. The rest of the draft may be in detailed outline form, although it would be better to have it fully written.


We are aware that the current situation complicates communication and group work. To make communication for the course members and among group members as easy as possible, we set up a Slack workspace. Check our ILIAS group for an invitation link on ILIAS course page. Please sign-in with your address.

Mental Health and Wellness

We are living in unprecedented times and the changes, challenges and stressors brought on by the COVID-19 pandemic have impacted everyone, often in ways that are tax our academic performance and general well-being. If you experience significant stress or worry, changes in mood, or problems eating or sleeping this semester, whether because of this or other courses or factors, please do not hesitate to reach out to any of the course’s instructors to discuss. Everyone can benefit from support during challenging times. Not only are we happy to listen and make accommodations with deadlines as needed, we can also refer you to additional support structures on campus.

Accessibility and Accommodations

It is our goal that this class is an accessible and welcoming experience for all students, including those with disabilities that may impact learning in this class. If you anticipate or experience academic barriers based on your disability (including mental health, chronic or temporary medical conditions), please let us know immediately so that we can privately discuss options. After registration, make arrangements with us as soon as possible to discuss your accommodations so that they may be implemented in a timely fashion.