Week 01 · August 24–26

Frame the study.

Begin with a consequential environmental-policy problem, test whether accessible data can support it, and establish a reproducible place to record the project’s decisions.

Mon & Wed · 12:30–1:45 PM South 227B

Note

Office hours: Mondays and Wednesdays, 1:45–3:15 PM, in South 227B—immediately after class.

This week

Research question

  • Identify two or three environmental-policy questions.
  • Name the affected people, places, institutions, and possible mechanism.
  • Separate descriptive, predictive, and causal ambitions.

Data feasibility

  • Find two candidate open-data sources.
  • Record unit, time, geography, provenance, and access method.
  • Identify the first important coverage or quality limitation.

Reproducible setup

  • Open the course R project and a research log.
  • Keep source data separate from analysis-ready data.
  • Record every consequential choice as the project changes.

Before class

  • Read the course syllabus.
  • Browse the Build one coherent data stack and Potential environmental data sources sections of the Data Lab.
  • Bring a laptop with R and RStudio ready to use.
  • Arrive with one environmental issue, place, resource, or policy that you would genuinely like to understand.

Lecture slides

TipLecture slide shortcuts

Click inside the slide preview, then use:

  • ←/→ or ↑/↓ to page through the slides;
  • M to open the menu;
  • F to enter fullscreen;
  • E to enter or exit PDF export mode;
  • Ctrl + Shift + F to search within the slides;
  • Esc to exit the menu or fullscreen; and
  • T to switch between dark and light mode.

Use the arrow keys to move through the deck. Press F for full screen and T to switch the slide theme.

Course launch and research website setup Open full screen ↗

Data transformation with dplyr Open full screen ↗

Class focus

Across the week, we will:

  • review the course arc, logistics, and evaluation;
  • move from broad environmental concerns to researchable policy questions;
  • identify stakeholders, mechanisms, and possible data sources;
  • establish a reproducible project structure and research log;
  • inspect a manageable file or API response for its unit of observation, variables, coverage, missingness, and provenance; and
  • create one quick diagnostic figure and compare candidate data paths.

Use familiar tools—readr, dplyr, and ggplot2—for the initial data inspection. Spatial and machine-learning tools come later, after the question and data structure are clear.

By the end of Week 1

Prepare a brief Question + Data Scan containing:

  1. two or three candidate research questions;
  2. one preferred direction and why its policy stakes matter;
  3. two candidate datasets with source links;
  4. the likely unit of observation, time span, geographic coverage, and provenance of each; and
  5. one initial plot or data-quality screenshot, plus the most important limitation it reveals.

This is a formative working artifact, not a new graded course requirement. It becomes the starting point for Week 2’s literature and evidence work.

This week’s materials

Back to top