01 — Getting Started

Course outlook, good research practice, and modern AI tools.

Lift off.

This module introduces the course, covers good research practice and reproducibility, and surveys modern AI tools for economists.

Lecture slides

View fullscreen

Setup

Please complete the following before the course.

1. GitHub

• Sign up for a free GitHub account.
• Fork the course repository.

2. Install core tools

macOS (recommended path)

Install Homebrew first (the macOS package manager):

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

Then install R, Quarto, and Git:

brew install --cask r
brew install --cask quarto
brew install git

Windows

• R 4.4+ — download the Windows installer
• Quarto 1.6+ — download the Windows installer
• Git for Windows — includes Git Bash

All platforms

• Install Visual Studio Code (or Positron)
• In VS Code, install the R extension and the Quarto extension

3. R packages

Open R and run:

install.packages("pacman")
pacman::p_load(
  tidyverse,    # data wrangling & visualization
  data.table,   # fast data manipulation
  ggplot2,      # plotting
  rvest,        # web scraping (module 04)
  sf,           # spatial data (module 07)
  terra,        # raster data (module 07)
  quanteda,     # text analysis (module 06)
  torch,        # deep learning backend (module 10)
  ellmer         # LLM API access (module 10)
)

For module 10 (LLMs), you additionally need to install the torch backend:

torch::install_torch()

Further resources

Reproducibility

• Gentzkow and Shapiro, Code and Data for the Social Sciences — the framework behind the 8 building blocks covered in this module
• Dreber (2025), A Framework for Evaluating Reproducibility, Economic Inquiry — referenced in the slides; ~30% of economics studies meet basic reproducibility standards
• Karl Broman’s initial steps toward reproducible research — short, practical checklist
• The Turing Way, Guide for Reproducible Research — comprehensive open guide to reproducible workflows

TipGoing further: renv for dependency management

The renv package pins exact R package versions so your project is reproducible months or years later. Run renv::init() in a project to create a lockfile, and renv::snapshot() to update it. See the renv documentation for details.

Git and version control

Start here:

• Jenny Bryan’s Happy Git and GitHub for the useR — the best resource for R users learning Git; covers setup on every OS, credential management, and common errors
• GitHub’s Git Handbook — concise overview of core concepts
• Learn Git Branching — interactive, visual exercises in the browser

Go deeper:

• Grant McDermott’s introduction to Git — for the economics context
• MIT’s The Missing Semester: Version Control — deeper dive into Git’s data model (snapshots, DAGs, objects)
• Pro Git by Scott Chacon — the definitive free reference

Docker and Dev Containers

• Rocker Project — Docker images for R, including rocker/tidyverse used in the slides
• VS Code Dev Containers documentation — how to set up the workflow shown in the lecture

Testing

• Wickham and Bryan, R Packages: Testing basics — the natural next step after the testthat introduction in the slides
• testthat documentation — reference for the testing framework