Course Material | AQM 2025

Week 1: Introduction. OLS Recap.

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Wed, Feb 11	8:30–10:00	Lecture	A5, 6, B244
Thu, Feb 12		Lab (Domantas)	B6, 23-25, A 103
Fri, Feb 13		Lab (Lisa)	A5, 6, B143

Study Notes

Welcome back!

Readings


	Course syllabus	Required
	Cunningham, Scott. Causal Inference - The Mixtape. Chapter 2.	Required

Week 10: Conditional Logit Model I

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Apr 28	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Apr 29	8:30–10:00	Lecture	A5,6 B244
	12:00	Homework 5 will be assigned	Github
Thu, Apr 30	10:15–11:45	Lab (Dom)	B6, 23-25, A 103
Fri, May 01	10:15–11:45	Public holiday, no lab	Go to Dom’s Lab

Study Notes

We will discuss conditional logit models. Please closely read Alvarez and Nagler’s 1998 AJPS piece and re-read Long’s Chapter 6.

Readings

Week 11: Selection Bias and Multi-Equation Models

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, May 05	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, May 06	8:30–10:00	Lecture	A5,6 B244
	12:00	Homework 6 will be assigned	Github
Thu, May 07	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, May 08	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

We will discuss selection bias models, particularly tobit and heckman models. For tobit models please closely read Long’s chapter 7 and for a nice application take a look at the recent award-winning Dellmuth and Stoel paper. In order to better understand the Heckman model browse through Rich Timpone’s APSR piece.

In the tutorial, we will talk about conditional logit models

Readings


	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 6.	Required
	Alvarez, R. Michael, and Jonathan Nagler. 1998 “When Politics and Models Collide: Estimating Models of Multiparty Elections.” American Journal of Political Science 42(1): 55-96.	Required

Week 12: Multi-level Models

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, May 12	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, May 13	8:30–10:00	Lecture	A5,6 B244
	23:59	Homework 5 is due	Github
Thu, May 14	10:15–11:45	Public Holiday	Go to Lisa’s Lab
Fri, May 15	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

Make sure to closely read all three chapters of Gelman & Hill and skim the Clark & Linzer article. We will discuss in class as an application the model of Steenbergen & Jones. For those who deeply care about TSCS data, take also a close look at the Bell & Jones (2015) paper.

Readings


	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 7.	Required
	Dellmuth, Lisa Maria, and Michael F. Stoffel. 2012. “Distributive politics and intergovernmental transfers: The local allocation of European Union structural funds.” European Union Politics 13(3): 413-433.	Required
	Timpone, Richard J. 1998. “Structure, Behavior, and Voter Turnout in the United States”. American Political Science Review 92(1): 145-158.	Required

Week 13: Baby Bayes

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, May 19	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, May 20	8:30–10:00	Lecture	A5,6 B244
	23:59	Homework 6 due	Github

Study Notes

Finally, we will cover some ground in terms of Bayesian Analysis. Please read the intro article by Andrew Martin. People interested in Bayesian multi-level models should also consult the now “classic” Stegmueller piece.

Readings


	Gelman, Andrew, and Jennifer Hill. 2006. Data Analysis Using Regression and Multilevel/Hierarchical Models. New York: Cambridge University Press. Chapters 11–13.	Required
	Bell, Andrew, and Kelvyn Jones. 2015. “Explaining Fixed Effects: Random Effects modeling of Time-Series Cross-Sectional and Panel Data”. Political Science Research and Methods 3(1): 133-153.	Required
	Clark, Tom S., and Drew A. Linzer. 2015. “Should I Use Fixed or Random Effects?” Political Science Research and Methods 3(2): 399-408.	Required
	Steenbergen, Marco R., and Bradford S. Jones. 2002. “Modeling Multilevel Data Structures.” American Journal of Political Science 46(1): 218-37.	Required

Week 14: Student Presentations

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, May 28	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, May 29	8:30–10:00	Lecture	A5,6 B244

Study Notes

In order to provide you with feedback on your final papers we will have short (< 5 min.) presentations of your hypothesis and the key results of your paper. Please email your presentations to Thomas until 8am that day.

Week 2: OLS in Matrix Form.

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Mon, Feb 16	by 12:00	Inform tutors about your homework teams
Tue, Feb 17	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Feb 18	8:30–10:00	Lecture	A5, 6, B244
Thu, Feb 19	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Feb 20	10:15–11:45	Lab (Lisa)	A5, 6, B143
	12:00	Homework 1 is assigned	Github

Study Notes

Work through Wooldridge’s appendix D first, because it will be heavily used in appendix E. Try to test yourself by doing some concrete examples (2 × 2- or 2 × 3-matrices are totally fine) to make sure you understand what’s going on. Moreover, closely read section E.1 – E.3 and skim the rest of this appendix. Also, in order review the linear model in matrix form take a look at Scott Long’s chapter 1–2, in particularly sections 2.1–2.5.

Readings


	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 1-2.	Required
	Wooldridge, Jeffrey, M. 2009. Introductory Econometrics. A Modern Approach. Appendix D.	Required
	Wooldridge, Jeffrey, M. 2009. Introductory Econometrics. A Modern Approach. Appendix E.	Required

Week 3: OLS in Matrix Form and Probability Theory

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Feb 24	13:30–14:30	Office hours (Thomas)	A5, 6, B 220
	15:30–17:00	Office hours (Domantas)	A5, 6, B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Feb 25	8:30–10:00	Lecture	A5, 6, B244
	12:00	Homework 2 is assigned	Github
	23:59	Homework 1 due	Github
Thu, Feb 26	10:15–11:45	Lab (Domantas)	B6, 23-25, A 102
Fri, Feb 27	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

We will wrap-up our discussion of the linear model in matrix form and do some applications with probability distributions to get more familiarity with them. The core reading will be chapter 1 and 3 from King (1989). Furthermore, a nice overview about probability theory and particular probability distributions (which was also covered last semester) is provided by Moore and Segal (2013).

Note for further reading I also provide John Fox’s appendix in its entirety. If you want to read more about linear algebra take a look at his treatment in Appendix B. It includes nice examples and graphs that provide some more intuition. Also check out sections D.1–D.5 if you prefer additional reading on probability distributions.

Readings


	King, Gary. 1989. Unifying Political Methodology. Ann Arbor: University of Michigan Press. Chapter 1+3.	Required
	Moore, Will H., and David A. Siegel. 2013. A Mathematics Course for Political and Social Research. Ann Arbor: Princeton University Press. Chapter 9.	Required
	Moore, Will H., and David A. Siegel. 2013. A Mathematics Course for Political and Social Research. Ann Arbor: Princeton University Press. Chapter 10.	Required
	Moore, Will H., and David A. Siegel. 2013. A Mathematics Course for Political and Social Research. Ann Arbor: Princeton University Press. Chapter 11.	Required
	Fox, John. 2007. Applied Regression, Generalized Linear Models, and Related Methods, 2 Edition, Thousand Oaks, CA: Sage. Appendix B+D	Required

Week 4: A first peek at Maximum Likelihood

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Mar 03	13:30–14:30	Office hours (Thomas)	A5, 6, B 220
	15:30–17:00	Office hours (Domantas)	A5, 6, B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Mar 04	8:30–10:00	Lecture	A5, 6, B244
	23:59	Homework 2 due	Github
Thu, Mar 05	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Mar 06	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

We will finally start with an introduction of the likelihood theory of inference. Please read chapter 4 (only until section 4.3) of King’s UPM book. For a quick peek at MLE, you can refer to Long’s chapter 2.6.

Readings


	King, Gary. 1989. Unifying Political Methodology. Ann Arbor: University of Michigan Press. Chapter 4.1-4.3.	Required
	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 2.6.	Required

Week 5: Maximum Likelihood Estimation and Heteroskedastic Regression

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Mar 10	13:30–14:30	Office hours (Thomas)	A5, 6, B 220
	15:30–17:00	Office hours (Domantas)	A5, 6, B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Mar 11	8:30–10:00	Lecture	A5, 6, B244
	12:00	Homework 3 is assigned	Github
Thu, Mar 12	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Mar 13	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

Make sure you closely (re)-read the entire King’s UPM, chapter 4. For those of you who appreciate a slightly different take on MLE take a look at Eliason (1993). Please also read a short section in Long (1997) chapter 3.6.1 and 3.6.2 in order to get a sense of how to actually estimate standard errors using maximum likelihood. For an nice application on how to set-up a heteroskedastic regression model take a look at the “classic” Franklin (1991) paper (this paper will also be relevant for your third homework!). Alternatively, another interesting application of a heteroskedastic regression model is found in Golder and Lloyd (2014).

Readings


	King, Gary. 1989. Unifying Political Methodology. Ann Arbor: University of Michigan Press. Chapter 4.	Required
	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 3.6.1 - 3.6.2.	Required
	Eliason, Scott R. 1993. Maximum Likelihood Estimation: Logic and Practice. Newbury Park: Sage. Chapter 1-4.	Required
	Franklin, Charles H. 1991. “Eschewing Obfuscation? Campaigns and the Perception of Senate Incumbents”. American Political Science Review 85(4): 1193–1214.
	Golder, Matt, and Gabriella Lloyd. 2014. “Re-Evaluating the Relationship between Electoral Rules and Ideological Congruence”. European Journal of Political Research 53(1): 200–212.

Week 6: Models for Binary Dependent Variables & Model Fit

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Mar 17	13:30–14:30	Office hours (Thomas)	A5, 6, B 220
	15:30–17:00	Office hours (Domantas)	A5, 6, B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Mar 18	8:30–10:00	Lecture	A5, 6, B244
	23:59	Homework 3 due	Github
Thu, Mar 19	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Mar 20	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

We will take a closer look at models for dichotomous dependent variables. Please take a close look at Chapter 5.1-5.3 of King (1989) and Chapter 3 in Long (1997). Also skim the Esarey/Pierce (2012) as well as the “separation plot” paper of Greenhill et al. (2011) for new strategies of how you could evaluate your model in terms of model fit. For an applied example of logistic regression models, have a look at the recent paper by Giger et al. (2020). They also present nice quantities of interest which we will cover next week. Finally, former students of our AQM class now contribute to this literature themselves! Take a look at Marcel and Sebastian’s work on cross-validation to assess model fit.

Readings


	King, Gary. 1989. Unifying Political Methodology. Ann Arbor: University of Michigan Press. Chapter 5.1 - 5.3.	Required
	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 3.	Required
	Esarey, Justin, and Andrew Pierce. 2012. “Assessing Fit Quality and Testing for Misspecification in Binary-Dependent Variable Models.” Political Analysis 20(4): 480–500.	Required
	Greenhill, Brian, Michael D. Ward, and Audrey Sacks. 2011 “The Separation Plot: A New Visual Method for Evaluating the Fit of Binary Models.” American Journal of Political Science 55(4): 991–1002.	Required
	Giger, Nathalie, Simon Lanz and Catherine de Vries. 2020. “The motivational basis of constituency work: how intrinsic and extrinsic motivations interact”. Political Science Research and Methods 8(3): 493–508.
	Neunhoeffer, Marcel, and Sternberg, Sebastian. 2019. “How Cross-Validation Can Go Wrong and What to Do About It”. Political Analysis 27(1): 101–106.

Week 7: Interpretation and Simulation

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Mar 24	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Mar 25	8:30–10:00	Lecture	A5,6 B244
	12:00	Homework 4 is assigned	Github
Thu, Mar 26	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Mar 27	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

Please read closely King et al (2000). If you have read it before - read it again! This piece is definitely on the “Top 10” list of papers every MA student has to digest. Also take a look at the Hanmer & Kalkan paper to understand the difference between average-case and observed-value approaches. Which one do you prefer? Finally, to see an example how much substance you can convey through simulating quantities of interest, take a look at Abrajano et al.

Readings


	King, Gary, Michael Tomz and Jason Wittenberg. 2000. “Making the Most of Statistical Analyses: Improving Interpretation and Presentation”. American Journal of Political Science 44(2): 347–361.	Required
	Hanmer, Michael J., and Kerem Ozan Kalkan. 2013. “Behind the Curve: Clarifying the Best Approach to Calculating Predicted Probabilities and Marginal Effects from Limited Dependent Variable Models.” American Journal of Political Science 57(1): 263–277.	Required
	Abrajano, Marisa A., R. Michael Alvarez, and Jonathan Nagler. 2008. “The Hispanic Vote in the 2004 Presidential Election: Insecurity and Moral Concerns.” The Journal of Politics 70(2): 368-82.	Required

Week 8: Ordered Choice Models & How to write a publishable Paper

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Apr 14	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Apr 15	8:30–10:00	Lecture	A5,6 B244
Thu, Apr 16	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Apr 17	10:15–11:45	Lab (Lisa)	A5, 6, B143

Study Notes

In the first part we will discuss ordered choice models. Focus on Long’s chapter 5. For a discussion of nifty applications of the ordered choice model take a look at Jackman’s lecture notes. In the second part of today’s lecture I will provide you with some strategies on how you can write a publishable paper in this class.

Readings


	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 5.	Required
	Jackman, Simon. 2000. Models for Ordered Outcomes. Lecture Notes.	Required

Week 9: Multinomial Choice Models

Mon, 01 Jan 0001 00:00:00 +0000

Schedule of the Week

Date	Time	Item	Place
Tue, Apr 21	13:30–14:30	Office hours (Thomas)	A5,6 B 220
	15:30–17:00	Office hours (Domantas)	A5,6 B 226
	15:30–17:00	Office hours (Lisa)	Zoom
Wed, Apr 22	8:30–10:00	Lecture	A5,6 B244
	23:59	Homework 4 due	Github
Thu, Apr 23	10:15–11:45	Lab (Domantas)	B6, 23-25, A 103
Fri, Apr 24	10:15–11:45	Lab (Muhammad)	A5, 6, B143

Study Notes

We will cover “multinomial choice models”. Please take a look at Long’s Chapter 6. In case you wanna see an application of this model, take a look and Gschwend & Leuffen’s BJPolS.

Readings


	Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. Chapter 6.	Required
	Gschwend, Thomas, and Leuffen, Dirk. 2005. “Divided We Stand–Unified We Govern? Cohabitation and Regime Voting in the 2002 French Elections.” British Journal of Political Science 35(4): 691–712.	Required