Lecture Notes for Advanced Quantitative Political Methodology
These are materials from class (videos and slides for lectures, annotated with R code), assignments (videos and slides from previous years, in a special collaborative annotation system), and section (videos, slides, other materials). When I lecture, I go through as much material as possible at each lecture, subject to the constraint that everyone follows what I'm doing. The speed at which I go is therefore dependent on the composition of each year's class and the questions that arise. As such, the slides below are not broken up into distinct weeks (I seem to go through roughly 15-20 pages in a weekly session lasting almost 2 hours, but issues and topics not represented here are covered most weeks).
In class, in addition to lecturing, I use LearningCatalytics (which I helped develop), along with some other demos and collaborative learning opportunities.
Separate PDF versions of the slides appear for teaching (i.e,. with pauses, etc.) and for (color) printing. R code is provided for each section. I update this material almost continuously while I teach. I'd appreciate if you would contact me with any comments, corrections, or suggestions.
Videos for 2013 lecture and section will be available here. Access requires being signed up through the extension school or having a Harvard ID.
The Basics
Course introduction, What is the field of political methodology?, What is statistics?, notation, probability, probability densities, statistical simulation.
2013 Lecture
Lecture 1 Video (2 hrs) Lecture 2 Video (2 hrs)
Slides used in class Printable version of slides
R code examples
2013 Section
Section 2 Video (1 hr) Section 2 Slides Section 1 Video (1.5 hrs) Section 1 Slides
Intro to R section materials
2012 Lecture
Lecture 1 Video (2 hrs) Lecture 2 Video (2 hrs) 
Slides used in class 
Printable version of slides R code examples
2012 Section
Section 0: Introduction to R (optional) Section 1: Probability Review
2010 Lecture (selections):
Introduction: Advanced Quantitative Research Methodology Course Introduction (17:49) What is Statistics? (24:24) Statistical Models - Part 1 (15:31) Statistical Models - Part 2 (32:51) Simulation (11:34) Probability Densities (16:57) Bernoulli and Binomial Distributions (23:28)
Theories of Inference
Alternative theories of inference (Bayes, likelihood, Neyman-Pearson hypothesis testing, etc.)
2013 Lecture
Lecture 3 Video (2 hrs) Lecture 4 Video (2 hrs) Lecture 5 Video (2 hrs) Slides used in class Slides for printing
2013 Section
Section 3 Video (1 hr) Section 3 Slides Section 3 Slides (printable) Section 4 Slides Section 4 Video (1 hr)
Computation Section Code Section 5 Video (1 hr) Section 5 Slides Section 5 Slides (Printable)
2012 Lecture
Lecture 3 Video (2 hrs) Lecture 4 Video (2 hrs) Slides used in class Printable version of slides
2012 Section
Section 2: Introduction to Likelihood Section 2: Slides Section 3: Optimization Section 3: Slides Section 4: Likelihood (cont.) Section 4: Slides
2010 Lecture (selections):
Slides used in these videos Problem of Inference and Likelihood Theory of Inference (36:34) Bayesian Theory of Inference (22:14) Neyman-Pearson Hypothesis Testing (13:03) What is the Best Theory of Inference? (09:29) Likelihood of Stylized Normal (28:44) Hyperdimensional Space (11:04) Finding Maximum of Likelihood (05:15) Properties of Likelihood Maximum + Uncertainty (23:02) Standard Errors of Maximum Likelihood Estimates (29:20) Simulation for any Maximum Likelihood Model (08:37) Forecasting Presidential Elections - Part 1 (18:42) Forecasting Presidential Elections - Part 2 (21:01) Variance Function Models (08:05)
Single Equation Models
Binary variable models, interpretation and presentation via simulation, ordinal dependent variables, how do you know which model is better?, grouped binary variables, event counts, simple duration models and censoring.
2013 Lecture
Lecture 5 Video (2 hrs) Lecture 6 Video (2 hrs) Lecture 7 Video (2 hrs) Lecture 8 Video (2 hrs) Slides used in class Slides for printing
2013 Section
Section 6 Video (1 hour) (Part 1 is Dichotomous Outcomes; Part 2 is Robust Standard Errors) Section 7 Video (1 hour) Section 8 Video (1 hour) Section 9 Video (1 hour) Section 5 Slides Section 5 Slides (printable) Section 6 Slides Section 6 Slides (printable) Section 7 Slides Section 8 Slides (printable) Section 9 Slides (printable)
2012 Lecture
Lecture 5 Video (2 hrs) Lecture 6 Video (2 hrs) Slides used in class Printable version of slides Single Equation Models
2012 Section
Section 6: Dichotomous Outcomes and Presenting Quantities of Interest Section 6: Slides Section 7 Section 7: Slides Section 8: Count Models, Counterfactuals, Duration Models Section 8: Slides
2010 Lecture, selections assigned this year:
smodelsp2010.pdf Logit Model - Part 1 (10:49)
Logit Model - Part 2 (17:42)
Alternative Interpretation of Logit Model (25:13)
Presenting Statistical Results - Part 1 (11:19)
Presenting Statistical Results - Part 2 (09:11)
Simulating Quantities of Interest and Associated Uncertainty (38:44)
Application of Clarify Package (06:18)
Ordinal Dependent Variable Models (26:01)
Model Validation - Part 1 (13:21)
Model Validation - Part 2 (25:51)
Grouped Uncorrelated Random Variables (20:20)
Event Count Models (27:37)
Duration Models: Exponential (17:38)
Causal Inference
Detecting Model Dependence
Sensitivity to parametric assumptions, revealing inferences too far from the data to have empirical answers, the curse of dimensionality, extrapolation, measures of distance from the data.
2013 Lecture
Lecture 9 Video (2 hrs) Lecture 8 Video (2 hrs) Slides used in class Printable version of slides 5modeldependencematching.r
2012 Lecture
Lecture 7 Video (2 hrs) Slides used in class Printable version of slides 5modeldependencematching.r
2012 Section
Section
Section: Slides
2010 Lecture, selections assigned this year:
Slides used in these videos Model Dependence (27:42) Example of Model Dependence (Doyle and Sambanis, APSR 2000) (13:13)
Matching Methods to Ameliorate Model Dependence
Matching methods as nonparametric preprocessing to reduce model dependence in parametric causal inference.
2013 Lecture
Lecture 9 Video (2 hrs) Slides used in class Printable version of slides R Code
2013 Section Section 10 Video (2 hrs) Section 10 Slides (printable)
2012 Lecture
Lecture 8 Video (2 hrs) Slides used in class Printable version of slides R Code
2012 Section
Section 9 section9.pdf Section 9 R Code
2010 Lecture, selections assigned this year:
psparap2011.pdf Intro to Matching and Review of Model Dependence (06:02) Matching as Solution to Model Dependence (27:23) Matching and Causal Quantities of Interest (07:44) 3 Methods of Matching (25:22) Space Graphs: Visualizing Bias-Variance Tradeoff in Matching (17:30) Reorientation on Matching Methods and Space Graphs (16:03) Propensity Score Matching as Approximating Random Matching (15:57) Problems with Popular Matching Methods and CEM as a Valuable Alternative (11:46)
Research Designs
An overview of how key features of various observational and experimental research designs, and the designs themselves, reduce components of error in estimating causal effects.
2012 Lecture
Lecture 10 Video (2 hrs)
Slides used in class Printable version of slides R Code
2013 Section
2012 Section
2010 Lecture, selections assigned this year:
matchsetlkp2010.pdf Causal Quantities of Interest and Error Decomposition (44:35) Benefits of Major Research Designs (38:28)
Multiple Equation Models
Code for multiple equation models and missing data.
Analysis Models
Identification, how multiple and single equation models differ, seemingly unrelated regression models, reciprocal causation, multiple equation reparametrization, multinomial choice models (multinomial probit, multinomial logit), independence of irrelevant alternatives, conditional logit.
2013 Lecture
Slides used in class Printable version of slides
2013 Section
Section12 Section 12 Slides
2012 Lecture
Lecture 10 Video (2 hrs) Slides used in class Printable version of slides
2012 Section
Section11 Section 11 Slides Section 11 R Code
2010 Lecture, selections assigned this year:
mmodelsp2010.pdf Multiple Equation Models and SURM (27:17) Reciprocal Causation (27:15)
Missing Data Modeling
Problems with listwise deletion, assumptions, general purpose methods, application specific methods, multiple imputation, computational algorithms, a detailed example, Amelia software.
2013 Lecture
Slides used in class Printable version of slides
Final lecture 2013: Big Data is Not About the Data! Slides used in class Printable version of slides
2012 Lecture
Lecture 11 Video (2 hrs) Slides used in class Printable version of slides
2012 Section
Section12 Section 12 Slides
2010 Lecture, selections assigned this year:
Printable slides to accompany videos Missing Data Notation and Setup (17:52) Why is Listwise Deletion Bad (13:36) General Purpose Existing Methods for Missing Data (08:41) Application-Specific Methods for Missing Data (11:27) Multiple Imputation for Missing Data - Part 1 (06:23) Multiple Imputation for Missing Data - Part 2 (15:40) Computational Algorithms for Missing Data (15:21) Amelia-Style Missing Data Imputation (16:20)
Text Analysis
Text Analysis: Statistical analysis when the observation is a text document of some kind.
Supervised Learning
Supervised Learning: Estimating the average in a category vs. individual classification, dealing with measurement error, extensions to an apparently unrelated application in epidemiology and public health.
2012 Lecture Notes: Slides [to print]
Videos ( Printable slides to accompany videos )
Quantities of Interest in Text Analysis and Representing Text as Numbers (22:27) Existing Statistical Approaches to Estimating Proportions in Categories (10:22)
Using Misclassification Rates to Correct Proportions (12:55) Verbal Autopsy Methods (07:18)
Unsupervised Learning
Unsupervised Learning: fully automated vs computer assisted cluster analysis; why humans are incapable of learning from text as well as computers & why computers can't do it without humans.
2012 Lecture Notes: Slides (no 'to print' version)
Videos ( Printable slides to accompany videos )
New Strategy for Cluster Analysis (28:22) Applications of Cluster Analysis (19:33)