API 209 MATH CAMP 2026
STUDENT BOOK/ CODE, DATA, AND JUDGMENT
PROGRAMPA/ID
DATESAUG 13-26
CONTACT12 HOURS

Lesson 1 | Lab 1 | Lesson 2 | Lab 2

Code, data, and judgment.

A practical introduction to R, Quarto, and coding agents for international development policy.

Math Camp begins with a policy question. You will examine development data, make the question precise, ask an agent for help, and decide what evidence is strong enough to trust.

Everyone receives a working dataset and code scaffold. Students new to programming can follow the main path. Experienced programmers can update the data, test alternate specifications, and challenge the agent.

FIG_001[ OLS MINIMIZES SQUARED RESIDUALS ]Σe² = —
OUTCOME PREDICTOR

Each vertical segment is a residual. OLS selects the intercept and slope that make Σe² as small as possible.

R + RSTUDIOLATEX + BEAMERCODEX / CLAUDE CODE

WHY THIS CAMP

A policy question becomes a computational investigation.

Math Camp introduces the workflow that students will use before the fall statistics course. It gives everyone a common way to enter a data project: begin with a question, learn what the records can represent, make a small move in R, and check whether the computer did what you intended. Students with different levels of programming experience will use the same materials and can take steps that fit their current skills.

The current materials have four parts: Lesson 1, Lab 1, Lesson 2, and Lab 2. Lesson 1 introduces coding, the tools we will use, and the judgment needed when working with a coding agent. Lab 1 is an installation clinic. Lesson 2 shows how a policy question becomes a checked country-year table, and Lab 2 gives students time to build that table. There is no group submission and no grade.

R does the computation. RStudio is the place where we read, run, and revise code. Codex is the common coding agent, while Claude Code is an optional alternative. LaTeX Beamer is used for the lesson slides. These tools do different jobs. A coding agent can read files and propose edits. The student remains responsible for framing the policy question and deciding whether a numerical pattern supports a causal statement.

By the end of camp, each student may keep a small personal artifact: an R script, a short Quarto notebook, or another compact record of the investigation. The artifact is optional and ungraded. Its value is practical—it gives you code, questions, and verification habits that can travel into the semester.

AUGUST 2026

Four parts for our first two meetings.

Lesson 1 and Lab 1 take place on August 13. Lesson 2 and Lab 2 take place on August 19 and 20. The lessons introduce the ideas and demonstrate the work. The labs give you time to try the work on your own laptop, ask questions, and check that the result is correct.

  1. 01 Lesson 1 · AI, coding, and judgmentLearn what the tools do, why coding is still useful, and how to ask for help without giving away the important decisions.
  2. 02 Lab 1 · Installation clinicBring every laptop to green: project, R, packages, Quarto, data, and Codex.
  3. 03 Lesson 2 · Build the datasetSelect, filter, transform, join, and document the rows needed for one policy question.
  4. 04 Lab 2 · Build-table sprintConstruct one country-year analysis table and make its key, sample, transformations, and checks visible.

POLICY QUESTIONS

One file. Four ways into development.

Everyone first completes the same health example, giving the room one common object to inspect and discuss. Students may then choose a development path. The paths share documentation and a verification protocol. Each complete recipe defines the unit, sample, comparison, and limitation required by its own question.

These are descriptive and associational exercises. Their purpose is to practice turning a broad concern into variables, comparisons, checks, and honest claim language. A visible pattern can motivate a better question. A policy effect requires a separate causal research design.

  1. 01HEALTH · INCOME AND SURVIVALHow is national income associated with under-five mortality, and does the pattern differ by income group?

    Use GDP per capita, under-five mortality, and income classifications to examine scale, transformation, and between-group comparisons.

  2. 02GENDER · SCHOOLING AND FERTILITYWhere is girls’ secondary enrollment associated with lower adolescent fertility, and where does the relationship depart from the overall pattern?

    Use female secondary enrollment and adolescent fertility to think about missing coverage, regional context, and observations that do not follow the overall pattern.

  3. 03ACCESS · ELECTRICITY AND CONNECTIONDoes electricity access travel with internet access, and which countries closed the combined access gap fastest?

    Use electricity access, internet use, year, and region to distinguish levels from changes and to compare trajectories over time.

  4. 04CLIMATE · ENERGY AND GROWTHCan renewable electricity expand while carbon intensity falls and income continues to grow?

    Use renewable electricity, carbon intensity, and GDP per capita growth to practice multi-variable reasoning without collapsing a complex transition into one score.

THE WORKING PROTOCOL

A five-step protocol for supervised agent work.

The labs repeat one short sequence that makes each use of an AI agent visible and reviewable. The order matters. Students first form an expectation, then make a small attempt, then ask for bounded assistance, and finally test the result against the data and documentation.

  1. 01FRAMEState the question before opening the agent.

    Name the population, unit of observation, variables, comparison, time period, and the pattern you expect. This gives you something concrete to evaluate when code or advice arrives.

  2. 02HUMAN PASSRead or modify a small piece of R.

    Run the smallest working example you can understand. Predict what should change when an input changes, and keep the output that records what actually happened.

  3. 03AGENT PASSRequest one bounded, reviewable task.

    Give Codex or Claude Code a goal, relevant project context, constraints, and a completion check. Ask it to separate facts found in files from inferences it makes.

  4. 04VERIFYMake the result earn your trust.

    Inspect rows, keys, types, ranges, missingness, figures, model output, and claim language. Classify important agent statements as confirmed, contradicted, or unresolved.

REFERENCE

Keep the working guides nearby.

The lessons explain the work in sequence. The references answer questions that return during class: how to repair the setup, where a variable came from, what a technical term means, or where to find an editable file. These pages are working guides for use during class.