Methods Workshops

Note: All workshops are co-hosted with the Canada Research Chair in Electoral Democracy

PAST WORKSHOPS
Friday June 19, 2020: Shawna N. Smith (University of Michigan). “Such a nuisance (parameter): Interactions & cross-model comparisons in binary response models” 13h-14h, On Zoom.

As social scientists, nearly all of us have a binary dependent variable (with discrete categories 0/1) that is of scientific interest. Most of us also know that appropriate models for these types of outcome variables require use of a nonlinear link function that maps an unbounded set of parameters (e.g., Y*=B0 + B1X1 +B2X2 + … +BkXk) onto a probability space where Y = [0, 1]. Most commonly used link functions result in parameters that are interpretable with respect to the logistic cumulative density function (for logit) or the Gaussian cumulative density function (for probit). Although these coefficients are less directly interpretable than coefficients for continuous outcomes, they are still often used for assessing direction of effects, significance, and even relative magnitude. However, even this limited utility breaks down when scientific interest moves beyond main effects and into exploring interactions in binary models (e.g., effect moderation) or comparing coefficients across binary models (e.g., effect mediation). Rather, with respect to interactions, mediation and cross-model comparisons, the identification assumptions and functional forms prerequisite of logit and probit models make the coefficients they produce little more the nuisance parameters that, if interpreted directly, can lead to incorrect conclusions. This lecture will provide an overview of these issues and suggest several alternative methods for appropriately exploring these questions with binary outcomes, largely focused around examining changes in average marginal effects.

Thursday May 7, 2020: Mirya Holman (Tulane University) “How to survive (and thrive) in academia: a practical guide” 13h-14h, On Zoom.

Academic life is often very stressful, filled with seemingly never ending work. Perfectionism, over-prepping for instruction, and too much service all limit our productivity. It does not have to be that way! You can make concrete changes to your approach that will reduce your stress and increase your productivity. I will discuss practical methods of managing your workload, writing (a lot), and maintaining work-life balance. Come to workshop with a to-do list and your calendar or planner.

Thursday April 30, 2020: Arthur Spirling (New York University). “Word Embeddings: What works, what doesn’t, and how to tell the difference for applied research” 13h-14h, On Zoom.

Word embeddings are becoming popular for political science research, yet we know little about their properties and performance. To help scholars seeking to use these techniques, we explore the effects of key parameter choices---including context window length, embedding vector dimensions and pre-trained vs locally fit variants---on the efficiency and quality of inferences possible with these models. Reassuringly we show that results are generally robust to such choices for political corpora of various sizes and in various languages. Beyond reporting extensive technical findings, we provide a novel crowdsourced ``Turing test"-style method for examining the relative performance of any two models that produce substantive, text-based outputs. Our results are encouraging: popular, easily available pre-trained embeddings perform at a level close to---or surpassing---both human coders and more complicated locally-fit models. For completeness, we provide best practice advice for cases where local fitting is required

Friday April 24, 2020: Vincent Arel-Bundock (Université de Montréal). “Régression Multi-Niveau pour les Nuls” 13h-14h, On Zoom.

Vous ne comprenez pas trop comment la Régression Multiniveau fonctionne? Ça tombe bien, Vincent non plus! Tentez de démêler les choses avec lui lors d’une séance spéciale de l’atelier méthodologique Science Po UdeM : Régression Multi-Niveau pour les Nuls.

Friday April 17, 2020: Alexander Coppock (Yale University). “Visualize as You Randomize: Design-Based Statistical Graphs for Randomized Experiments” 13h-14h, On Zoom.

A good statistical graph for a randomized experiment simultaneously conveys the study's design, analysis, and results. It reveals the experimental design by mapping design elements to aesthetic parameters. It illuminates the analysis by plotting the statistical model in “dataspace." When the design and analysis of an experiment are encoded in a plot, the interpretation of the experimental results is clarified. “Analyze as you randomize" is a dictum attributed to Fisher that guides interpretations of experimental data. This chapter extends that principle to visualizations of randomized experiments. While not every experiment requires a visualization, those that do should be visualized in ways that communicate the design and results together.

Friday April 10, 2020: John Poe (University of Michigan). “Multilevel Modeling and Omitted Variable Bias” 13h-14h, On Zoom.

Bias in statistical models is probably the most common concern in the social sciences. Multilevel models can be used to build bias mitigation strategies in ways that are not common understood. This approach is different from standard causal inference techniques based on approximating randomized controlled trials that are extremely useful but often only narrowly applied in observational settings. This talk will walk through strategies for bias mitigation and the weaknesses inherent in the approach.

Friday April 3, 2020: Korhan Koçak (Princeton University). Workshop. “What Do We Learn About Voter Preferences From Conjoint Experiments?” 13h-14h, On Zoom.

Political scientists frequently interpret the results of conjoint experiments as reflective of voter preferences. In this paper we show that the target estimand of conjoint experiments, the AMCE, is not well-defined in these terms. Even with individually rational experimental subjects, unbiased estimates of the AMCE can indicate the opposite of the true preference of the majority. To show this, we characterize the social choice correspondence implied by AMCE and demonstrate its several undesirable properties. With this result we provide a method for placing sharp bounds on the proportion of experimental subjects with a strict preference for a given candidate-feature. We provide testable assumptions that help reduce the size of these bounds. Finally, we offer a structural interpretation of the AMCE and highlight that the problem we describe persists even when a model of voting is imposed. In this talk, I will present our paper and provide a few applications that I like.

Thursday March 26, 2020: Semra Sevi (Université de Montréal). Workshop. “Writing in LaTeX” 13h-14h, On Zoom.

LaTeX is a programming language. This workshop will go over what LaTeX is and why students should learn it. It will provide students with the tools needed to make their CV and documents in LaTeX. No familiarity with LaTeX is assumed. The workshop will go over the basic formatting and document layout (with templates that students can take away with them). Please either create an account on Overleaf <https://www.overleaf.com> or download TexShop here <https://www.tug.org/texlive/> .

Thursday February 27, 2020: Semra Sevi (Université de Montréal). Workshop. “Designing a Survey on Qualtrics”. Pavillon Lionel-Groulx, Université de Montréal. 17h-18h, C-4019.

In this Qualtrics methodological workshop, I will show you how to design a survey. The workshop will cover forced response, randomization, quotas, and will answer any questions you have.

Thursday February 20, 2020: John Holbein (University of Virginia). Workshop. “Regression Discontinuity Designs”. Pavillion Lionel-Groulx, Université de Montréal. 16h-17h, C-6070-9.

Regression Discontinuity Designs (RDDs) have exploded in recent years as a quasi-experimental research design in recent years as a means of approaching causal inference in observational settings. What do you need to know to be a good consumer and producer of RDDs? In this workshop, I will discuss the theoretical underpinnings of RDDs, discuss their strengths and weaknesses, provide tools for executing a RDD, and outline recent research that had used RDDs.

Friday January 17, 2020: Annika Fredén (Karlstad University) and Sverker Sikström (Lund University). Workshop. "Quantitative Text Analysis Workshop". Pavillon Lionel-Groulx, Université de Montréal. 15h-17h, C-4145.

The workshop will focus on two techniques: the more “supervised” Linguistic Inquiry and Word Count, and the more “unsupervised” Latent semantic analysis (plots and correlation analyses) in different forms. Participants need to bring their own laptop.
Here is the online version of the text analysis program that will be used: https://semanticexcel.com/semantic/login

Friday November 2, 2018: Jamie Druckman (Northwestern University). Workshop. "Why Replications Do Not Fix the Reproducibility Crisis: A Model and Evidence from a Large-Scale Vignette Experiment". Pavillon Lionel-Groulx, Université de Montréal. 15h-17h, C-6070-9.

Workshop led by Jamie Druckman on the Reproducibility Crisis.