Readings

Every work the slides cite, week by week, with a note on why it’s worth your time. All of it is optional. Each week’s list goes up after that week’s class. For the three textbooks (IW, BF, SG), the schedule lists the sections that go with each week.

Week 1: Introduction

The résumé experiment and its critics

  • Bertrand and Mullainathan (2004) — The study we worked through in class.
  • Imai and Williams (2022), chapter 2 — Where our copy of the data comes from, analyzed in R.
  • Fryer and Levitt (2004) — The source of the skeptic’s point, a name signals class as well as race.
  • Gaddis (2017) — Asks people what race they think these names signal.
  • Sen and Wasow (2016) — How can race have an effect if no one can assign it?

Descriptive or causal?

  • Hernán (2018) — A four-page argument against hiding a causal question behind “is associated with.”
  • Grosz, Rohrer, and Thoemmes (2020) — The same argument, made to psychologists.
  • Gerring (2012) — The case for description as a contribution in its own right.
  • Kaufman and Rogowski (2024) — The abstract we argued about in class.

What a cause is

  • Bueno de Mesquita and Fowler (2021), The definition of a causal effect we used, why we never see one directly.
  • Holland (1986) — The paper that named the “fundamental problem of causal inference.”

Selecting on the dependent variable

  • Geddes (1990) — The Asian tigers example from class, and the same critique aimed at Skocpol’s theory of revolutions.
  • Sekhon (2004) — A reply to Geddes.
  • Skocpol (1979) — A positive theory tested with three revolutions instead of a dataset.
  • Gladwell (2008) — The 10,000-hour rule.

Data-generating processes

  • Gailmard (2014), chapters 1 and 3 — The most mathematical of our three books.

Week 2: Bias, simulation, and measurement

Estimates, bias, and noise

  • Bueno de Mesquita and Fowler (2021), chapters 6 and 9 — Chapter 6 defines estimands, estimates, bias, and noise (the curling analogy is on p. 96). Chapter 9 splits a difference in means into the effect plus a baseline difference, the split we worked through for democracy and civil war. Its “Signing the Bias” section (pp. 176–79) is the rule we practiced, with its limits.

Confounding

  • McElreath (2020), chapter 5 — Where the Waffle House example comes from, in a graduate Bayesian course that is also one of the best places to learn to think in terms of data-generating processes.
  • Fearon and Laitin (2003) — The civil war data from class. Their regressions point where our income groups did: income matters a great deal, and democracy much less.

Measurement

  • Adcock and Collier (2001) — The standard political science treatment of measurement validity, written for quantitative and qualitative researchers alike.
  • Stevens (1946) — Where nominal, ordinal, interval, and ratio come from. Four pages.
  • Paxton (2000) — What happens to democracy scores when you count who is allowed to vote.
  • Hegre et al. (2001) — The classic finding that regimes in the middle of the Polity scale have the most civil wars.
  • Vreeland (2008) — Part of the Polity scale is coded from political violence, so the link between anocracy and civil war is partly built in.
  • Jones and Lupu (2018) — A newer look at the same question with machine learning, which finds the pattern is still there.
  • Weidmann (2016) — How news-based conflict data can make an effect appear where reporters can see violence.
  • Fryer and Levitt (2004) — Back from week 1: a name signals class as well as race, a construct validity problem.

Week 3: Averages, spread, and relationships

Summarizing one variable

  • Imai and Williams (2022) — Where the face-competence data come from, with a careful walk through correlation, the regression line, and what a slope does and doesn’t tell you.
  • Carter, Shaver, and Wright (2019) — Spread as the thing we study: rugged terrain, a lot of variation in elevation, let separate ethnic groups form and survive, and part of terrain’s link to civil war runs through their exclusion from power.
  • Bonica (2014) — The source of the ideology data behind the polarization slide: a way to place candidates, donors, and groups on one left-right scale using who gives money to whom.

Relationships

  • Bueno de Mesquita and Fowler (2021) — Chapter 2 is where our definition of correlation comes from. It builds the idea from a two-by-two table of oil and autocracy before any formula, then gets to slopes and lines.
  • Todorov et al. (2005) — The study behind the face data: one-second judgments of competence predicted Senate races about 70% of the time. Worth reading for how they handled the obvious objections, incumbency among them.

Week 4: Probability

Where randomness comes from

  • Gailmard (2014) — Chapter 3 is where the three sources of randomness (sampling, theoretical, fundamental) come from, with the “rerun the French Revolution” thought experiment (73–78). Chapter 4 covers the probability rules more formally.

Probability as a long run

  • Imai and Williams (2022) — Chapter 6 opens with the two readings of probability (long-run frequency and degree of belief), the rules, and the birthday problem, all with R simulations.

Finding the sampling distribution

  • Xu (2026b) — The source of the two paths slide: find an estimate’s distribution by simulation (rerun the study on a computer) or by analysis (derive it with math), and use each to check the other. A Stanford regression course, so the math runs ahead of ours; the framing is what matters.

Week 5: Expectation and variance

The rules

  • Imai and Williams (2022) — Section 6.3.5 defines expectation and variance for random variables and gives the rules for sums and scalings, with R checks; 6.3.6 (predicting elections with uncertainty) is the same rules at work on a poll. Section 6.4 covers the law of large numbers, which licenses the plug-in step at the end of the lecture.
  • Gailmard (2014) — Chapter 5 treats expectation, variance, and covariance of random variables more formally than lecture, including conditional expectation. The best place to see the three-line derivation of Var(X + Y) that the slides only state.

Where the formula came from

  • Bertrand and Mullainathan (2004) — Table 1 reports the callback rates and standard errors the lecture rebuilds by hand. Their footnotes explain that four résumés went to every ad and that standard errors are clustered by ad: the independence assumption the lecture flags, and its fix (week 10).

Checking math against simulation

  • Xu (2026b) — The source of the big-picture slide (goal, parameters, estimand, estimator, estimate), which is the first slide of his lecture. The lecture then does what ours did: it derives the expected value and variance of a sample mean, then checks each against a simulation. Xu posts the slides, a five-minute video, and the rest of the course, with R code for every figure.

Week 6: The normal, the law of large numbers, and the central limit theorem

The theorems

  • Imai and Williams (2022) — Section 6.3 introduces the uniform and normal distributions alongside the discrete ones, and Section 6.4 states the law of large numbers and the central limit theorem, with R simulations close to the ones from class.
  • Xu (2026b) — The second half of lecture 2-1 states the weak law of large numbers and the Lindeberg–Lévy central limit theorem formally, with the “convergence in probability” language week 7 uses for consistency.

What a standard error is

  • Wooldridge (2023) — Eight pages on what a standard error is. Section 1 is the source of the recap’s point that the standard error is itself an estimate: it separates the sampling SD (the true width, unknown) from the standard error (our estimate of it). The rest asks what a standard error means when the data hold the whole population, and when to cluster: background for weeks 8 and 10.

Week 7: Choosing an estimator

Properties of estimators

  • Imai and Williams (2022) — Section 7.1 treats estimators as random variables: their sampling distributions, bias, and standard errors, checked with simulations in R.
  • Xu (2026b) — The source of the four estimators of an average. The lecture defines finite-sample properties (unbiasedness, efficiency) and asymptotic ones (consistency, asymptotic normality), derives the bias and variance of each estimator, and proves that MSE equals bias squared plus variance. Its worksheet problems are good practice.

Counting the dead

  • Ball et al. (2003) — The estimate for Peru’s truth commission. Worth reading for how lists from different organizations were matched, and why the estimates were made separately for each perpetrator.
  • Rendón (2019) — A critique of that estimate, arguing that the method distorted the total, with a much lower count.
  • Manrique-Vallier, Ball, and Sulmont (2019) — A later reanalysis by the original team, adding a newly available list and newer methods. Read with Rendón (2019) to see how much the answer depends on assumptions about how the lists were generated.

Week 8: Confidence intervals

Building and reading an interval

  • Xu (2026a) — The source of the interval’s derivation, the coverage simulation (100 intervals, about 95 catch the truth), and the formal \(t\) distribution: Xu states the theorem, lists the \(t\)’s properties, and leaves the proof as a worksheet problem. His slides and R code are posted with the rest of the course.
  • Greenland et al. (2016) — Twenty-five common misreadings of p-values, confidence intervals, and power, each explained in a paragraph. Several of the activity’s wrong answers are on the list. It also describes an interval as the range of values reasonably compatible with the data, the practical reading from Part 2. Worth rereading after week 9.
  • Lakens (2022) — The confidence interval chapter of a free online textbook, with interactive simulations of coverage, overlapping intervals, and “capture percentages.” Its sample-size chapter covers planning a study for precision, as on the planning slide.

Will the next study land in our interval?

  • Cumming and Maillardet (2006) — The source of the 83% figure: on average, a 95% interval holds the next study’s estimate only 83% of the time. Their point is about replication: landing outside the original interval is weak evidence that anything went wrong.

Overlapping intervals

  • Schenker and Gentleman (2001) — A short paper showing that checking whether two intervals overlap is too strict a test of whether two estimates differ. The source of the “overlap, yet a clear difference” zone.

When the data are the whole population

  • Abadie et al. (2020) — Opens with the all-50-states puzzle: what does a standard error mean when nothing was sampled? The answer depends on whether you want to describe these units or learn about the process that made them.
  • Wooldridge (2023) — The short, readable version of the same argument, plus when clustering helps and when it hurts (week 10).

References

Abadie, Alberto, Susan Athey, Guido W. Imbens, and Jeffrey M. Wooldridge. 2020. “Sampling-Based Versus Design-Based Uncertainty in Regression Analysis.” Econometrica 88(1): 265–96. doi:10.3982/ECTA12675.
Adcock, Robert, and David Collier. 2001. “Measurement Validity: A Shared Standard for Qualitative and Quantitative Research.” American Political Science Review 95(3): 529–46.
Ball, Patrick, Jana Asher, David Sulmont, and Daniel Manrique. 2003. How Many Peruvians Have Died? An Estimate of the Total Number of Victims Killed or Disappeared in the Armed Internal Conflict Between 1980 and 2000. American Association for the Advancement of Science. Report.
Bertrand, Marianne, and Sendhil Mullainathan. 2004. “Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” American Economic Review 94(4): 991–1013. doi:10.1257/0002828042002561.
Bonica, Adam. 2014. “Mapping the Ideological Marketplace.” American Journal of Political Science 58(2): 367–86. doi:10.1111/ajps.12062.
Bueno de Mesquita, Ethan, and Anthony Fowler. 2021. Thinking Clearly with Data: A Guide to Quantitative Reasoning and Analysis. Princeton, NJ: Princeton University Press.
Carter, David B., Andrew C. Shaver, and Austin L. Wright. 2019. “Places to Hide: Terrain, Ethnicity, and Civil Conflict.” Journal of Politics 81(4): 1446–65. doi:10.1086/704597.
Cumming, Geoff, and Robert Maillardet. 2006. “Confidence Intervals and Replication: Where Will the Next Mean Fall?” Psychological Methods 11(3): 217–27. doi:10.1037/1082-989X.11.3.217.
Fearon, James D., and David D. Laitin. 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97(1): 75–90.
Fryer, Roland G., and Steven D. Levitt. 2004. “The Causes and Consequences of Distinctively Black Names.” Quarterly Journal of Economics 119(3): 767–805. doi:10.1162/0033553041502180.
Gaddis, S. Michael. 2017. “How Black Are Lakisha and Jamal? Racial Perceptions from Names Used in Correspondence Audit Studies.” Sociological Science 4: 469–89. doi:10.15195/v4.a19.
Gailmard, Sean. 2014. Statistical Modeling and Inference for Social Science. New York: Cambridge University Press.
Geddes, Barbara. 1990. “How the Cases You Choose Affect the Answers You Get: Selection Bias in Comparative Politics.” Political Analysis 2: 131–50.
Gerring, John. 2012. “Mere Description.” British Journal of Political Science 42(4): 721–46. doi:10.1017/S0007123412000130.
Gladwell, Malcolm. 2008. Outliers: The Story of Success. New York: Little, Brown and Company.
Greenland, Sander, Stephen J. Senn, Kenneth J. Rothman, John B. Carlin, Charles Poole, Steven N. Goodman, and Douglas G. Altman. 2016. “Statistical Tests, P Values, Confidence Intervals, and Power: A Guide to Misinterpretations.” European Journal of Epidemiology 31(4): 337–50. doi:10.1007/s10654-016-0149-3.
Grosz, Michael P., Julia M. Rohrer, and Felix Thoemmes. 2020. “The Taboo Against Explicit Causal Inference in Nonexperimental Psychology.” Perspectives on Psychological Science 15(5): 1243–55. doi:10.1177/1745691620921521.
Hegre, Håvard, Tanja Ellingsen, Scott Gates, and Nils Petter Gleditsch. 2001. “Toward a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816–1992.” American Political Science Review 95(1): 33–48.
Hernán, Miguel A. 2018. “The C-Word: Scientific Euphemisms Do Not Improve Causal Inference from Observational Data.” American Journal of Public Health 108(5): 616–19. doi:10.2105/AJPH.2018.304337.
Holland, Paul W. 1986. “Statistics and Causal Inference.” Journal of the American Statistical Association 81(396): 945–60. doi:10.1080/01621459.1986.10478354.
Imai, Kosuke, and Nora Webb Williams. 2022. Quantitative Social Science: An Introduction in Tidyverse. Princeton, NJ: Princeton University Press.
Jones, Zachary M, and Yonatan Lupu. 2018. “Is There More Violence in the Middle?” American Journal of Political Science 62(3): 652–67.
Kaufman, Aaron R., and Jon C. Rogowski. 2024. “Divided Government, Strategic Substitution, and Presidential Unilateralism.” American Journal of Political Science 68(2): 816–31. doi:10.1111/ajps.12821.
Lakens, Daniël. 2022. Improving Your Statistical Inferences. doi:10.5281/zenodo.6409077.
Manrique-Vallier, Daniel, Patrick Ball, and David Sulmont. 2019. “Estimating the Number of Fatal Victims of the Peruvian Internal Armed Conflict, 1980–2000: An Application of Modern Multi-List Capture-Recapture Techniques.”
McElreath, Richard. 2020. Statistical Rethinking: A Bayesian Course with Examples in R and Stan. 2nd ed. Boca Raton, FL: CRC Press.
Paxton, Pamela. 2000. “Women’s Suffrage in the Measurement of Democracy: Problems of Operationalization.” Studies in Comparative International Development 35(3): 92–111.
Rendón, Silvio. 2019. “Capturing Correctly: A Reanalysis of the Indirect Capture–Recapture Methods in the Peruvian Truth and Reconciliation Commission.” Research & Politics 6(1). doi:10.1177/2053168018820375.
Schenker, Nathaniel, and Jane F. Gentleman. 2001. “On Judging the Significance of Differences by Examining the Overlap Between Confidence Intervals.” The American Statistician 55(3): 182–86. doi:10.1198/000313001317097960.
Sekhon, Jasjeet S. 2004. “Quality Meets Quantity: Case Studies, Conditional Probability, and Counterfactuals.” Perspectives on Politics 2(2): 281–93.
Sen, Maya, and Omar Wasow. 2016. “Race as a Bundle of Sticks: Designs That Estimate Effects of Seemingly Immutable Characteristics.” Annual Review of Political Science 19: 499–522. doi:10.1146/annurev-polisci-032015-010015.
Skocpol, Theda. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia, and China. Cambridge: Cambridge University Press.
Stevens, S. S. 1946. “On the Theory of Scales of Measurement.” Science 103(2684): 677–80.
Todorov, Alexander, Anesu N. Mandisodza, Amir Goren, and Crystal C. Hall. 2005. “Inferences of Competence from Faces Predict Election Outcomes.” Science 308(5728): 1623–26. doi:10.1126/science.1110589.
Vreeland, James Raymond. 2008. “The Effect of Political Regime on Civil War: Unpacking Anocracy.” Journal of Conflict Resolution 52(3): 401–25.
Weidmann, Nils B. 2016. “A Closer Look at Reporting Bias in Conflict Event Data.” American Journal of Political Science 60(1): 206–18. doi:10.1111/ajps.12196.
Wooldridge, Jeffrey M. 2023. “What Is a Standard Error? (And How Should We Compute It?).” Journal of Econometrics 237: 105517. doi:10.1016/j.jeconom.2023.105517.
Xu, Yiqing. 2026a. “Interval Estimation.” https://github.com/xuyiqing/regression-lectures.
Xu, Yiqing. 2026b. “Point Estimation.” https://github.com/xuyiqing/regression-lectures.