Quantitative Analysis for IA Practitioners

IAFF 6501

Welcome!

Professor Teitelbaum ejt@gwu.edu
Office Hours: Tuesdays, 3:00-5:00 p.m.
Monroe Hall Rm 411 or online
Go to my calendly page to sign up for a slot

Teaching Assistants

Yuchen Zhao
email: yuchen.zhao1@gwu.edu
Office Hours: Wednesdays, 1:00 - 3:00 p.m.
Location TBD

Jeffrey Kuo
email: jeffkuo@gwu.edu
Office Hours: Wedensdays, 4:00-6:00 p.m.
Gelman Library

Why Take this Course?

• International Affairs is changing!
• Data is everywhere and it is changing the way government works
• You will be a better consumer of data and research
• You can be a “bridge builder” between traditional analysts and data scientists on your team

Data Informed Dipomacy

Data is a critical instrument of diplomacy. When our workforce has data at their fingertips they are better prepared to engage diplomatically, manage effectively, and lead globally.

Secretary of State Anthony Blinken, 2022¹

1. https://www.state.gov/data/

State Department Goals

• Cultivate a Data Culture
• Accelerate Decisions through Analytics
• Establish Mission-Driven Data Management
• Enhance Enterprise Data Governance

Open Goverment Data Act (2018)

• Requires government data assets to be published as machine-readable data in open formats
• Requires Chief Data Officers (CDOs) to be appointed at federal agencies
• Requires CDOs to develop and maintain comprehensive data inventories
• Has led to a proliferation of data science roles in the federal government

https://crsreports.congress.gov/product/pdf/IF/IF12299

Open Source Intelligence

• More than 90% of the analysis in the intelligence community is based on open source information
• Government agencies use a lot of the same datasets that we will be using in this class
• Yet the OSINT community has only begun to scratch the surface of what is possible with data science

MLP’s digital tools provide fine-grained data tracking and forecasting major political events across nearly 60 countries by continuously scraping and process tens of millions of articles published by more than 300 local, regional, and international news sources in more than 30 languages. This infrastructure provides up-to-date data on recent and historical trends in civic space and foreign influence and builds forecasting models that learn from historical patterns to predict how conditions are likely to change in the near future.

Monitoring, Evaluation and Learning

• Another important use of data in international affairs is monitoring, evaluation and learning (MEL)
• MEL is a process that helps organizations track and assess the performance of their programs
• MEL is a key component of USAID, the World Bank and other agencies
• A major component of MEL is the use of randomized control trials (RCTs) and other designs, which you will learn about in this class

Skills/Knowledge You Will Gain

1. R coding skills (and RStudio), with focus on “tidy” approach and reproducible research
2. Quarto (html documents, PDFs, presentations, websites, books, blogs, …)
3. How to access and “clean” data so that you can analyze it
4. When you hear terms like “machine learning”, you’ll have some sense of what people are talking about

Structure of the Course

1. Data Visualization
   • Summarizing and communicating effectively with data
2. Statistical Inference
   • Making rigorous conclusions from data
3. Modeling
   • For prediction and forecasting
   • For drawing causal conclusions

How do I get an “A”? (requirements)

• Weekly quizzes (20 percent)
• 3 Data Analysis Assignments (45 percent; 15 percent each)
• Final Project (20 percent)

Course Website

• https://quant4ia.rocks
• Let’s take a quick tour!

Install R and RStudio

• If you haven’t already…
• Go to the RStudio download page
• Download R and then RStudio

Set up RStudio

• Go to Tools>Global Options
• Under Code, enable native pipe operator (|>)
• Under Appearance, choose a theme
• Configure panes
  • Go to Pane Layout
  • Move Source, Console, etc. to preferred positions

Illustration

03:00

Install key packages

• Install the Tidyverse group of packages from the console
  • install.packages("tidyverse")
• Install devtools
  • install.packages("devtools"))
• Install tinytex (for PDF rendering)
  • Go to your terminal and type quarto install tinytex

Illustration

03:00

Let’s get going . . .

Your first data visualizations…

(and making sure we have R and RStudio installed and ready to roll)

Example: Make a map!

Example: Plotting Democracy Over Time

Example: UN Voting Trends

Your Task

• Make sure R and RStudio installed (we can help if needed)
• Create a folder for this class somewhere on your machine
  • Create a sub-folder called “classwork”
  • Download and save week1-classwork.qmd in that folder
• Open the week1-classwork.qmd file in RStudio, which has code for 3 data viz activities
  • Map making
  • Democracy Over Time
  • UN Voting patterns
• Follow the instructions to update the code
• Click the little green arrow to run the code chunk
• Click Render to update your HTML output
• Complete as much as you can (no problem if you do not finish)