Integration of Formal Theory and Quantitative Methods

Presenter: Tara Slough

Plan

Quantitative (statistical) methods and formal theory represent two “toolkits” central to graduate training in political science. However, we rarely discuss the merits and limitations of combining the two approaches, whether within a single work, a substantive research agenda, or something in between. This workshop will be a discussion framed with several works that discuss the integration of quantitative and formal work, with a focus on empirical methods of causal inference. We will conclude with a brief presentation of two working papers that utilize innovative approaches to integrating formal and empirical work.

Link to Workshop Materials

• Slides for the presentation

A Practical Guide to Large Scale Data Collection in Challenging Environments

ROOM/TIME CHANGE: IAB 711, 4:00-5:30pm

Presenter: Jasper Cooper

Plan

This workshop highlights some of the key practical considerations that need to be addressed when implementing a large scale data collection effort in a challenging environment. For example

• How should one go about recruiting enumerators and what kind of contract should one draw up with them?
• How does one train enumerators? How do transport and accommodation options affect the optimum team size?
• What kinds of roles should the data collection team feature and how much should the different members of the team be paid?
• What are good budget-monitoring practices and how should one access research funds in low-security environments where banking services are sparse or unreliable?
• What is the best way to collect data and what open source (free) options are available?
• When using electronic tablets, how can one plan for contingencies such as power cuts or working in areas without access to electricity?
• How long is the project likely to take, given team size, IRB requirements, sampling strategy, survey length, and other factors that influence the duration of the project?
• How does project duration impact the budget, and what are strategies to get the most out of a small budget?

I will draw examples principally from my experiences implementing large scale data collection in Papua New Guinea and Uganda. The workshop is intended as an exercise in highlighting the sorts of issues that arise when doing big surveys in challenging environments. Rather than giving all of the answers to these questions, this workshop is intended as an opportunity to share information and discuss solutions that have worked for others in the past.

Link to Workshop Materials

• Slides for the presentation

Bayesian Approach to Exclusion Restriction

Presenter: Rick McAlexander

Plan

Many works in political science use an instrumental variables (IV) regression to fix endogeneity problems and to identify causal effects. Standard approaches assume that an instrumental variable is correlated with the endogenous dependent variable, but that the instrument has no direct effect on the dependent variable of interest. Often, we have reason to belief that this assumption-the exclusion restriction-is violated, or at the least, we are uncertain if the assumption holds exactly. In this workshop, I show how to easily relax the exclusion restriction using a Bayesian model coded in Stan. I discuss what the implications are for our inferences from studies using IV models, and how the parameters from these models behave as the type of uncertainty changes. Please have Stan and rstan updated and installed!

Link to Workshop Materials

• Slides for the presentation
• Example R Script

Q&A on Bayesian Statistics and Causal Inference

Presenter: Andrew Gelman

Plan

Professor Gelman will be holding a Question and Answer session on

• Bayesian statistics
• Non-Bayesian methods
• Causal Inference
• The role of survey methods in this past election.

This workshop should be a great opportunity to have an in-depth discussion on these issues. Here are two links which have topics Professor Gelman has presented or commented on to help us all get started with questions: Prof. Gelman’s Personal Website and Prof. Gelman’s Blog.

If you plan on attending, please come with two prepared questions that you would like to discuss!

Link to Workshop Materials

• Some answers to selected questions will be posted here after the workshop.

Text Analysis

Presenter: Joe Sutherland

Plan

We will cover the following topics

• Models of structured and unstructured data;
• Document Vectorization;
• Text features – what are they?
• Pre-Processing: Stemming and Stopping;
• Bag-of-words model;
• Sparse matrices.

Setup Instructions

We will use python and homebrew to install additional packages, so please make sure you have them up and running. You can find tips on installation here

Link to Workshop Materials

• Workshop presentation and useful materials for learning about text analysis can be found in this Dropbox Folder.