Computational Analysis of Educational Data: A Field Guide Using R

This repository contains the source for a Quarto book on computational methods for educational research.

Book at a glance

This book is a practical field guide for educational researchers who want to run computational analyses in R with reproducible workflows. It combines method explanation, runnable code, and communication guidance for publishing transparent results.

The current structure includes four parts:

• Getting Started: environment setup, R workflow basics, and tidyverse foundations
• Data Science Methods: text, network, and numeric data analysis workflows
• LLM Methods: cloud/local LLM setup and applications for text and image workflows
• Communication and Conclusion: reporting, open dissemination, and reproducibility practices

Main chapter topics

• Chapter 1-3: Positron/RStudio setup, R workflow fundamentals, and tidyverse essentials
• Chapter 4: Text data methods (tokenization, frequency, sentiment, and topic modeling)
• Chapter 5: Network data methods using social network analysis workflows
• Chapter 6: Numeric data methods and modeling workflows with educational datasets
• Chapter 7-10: LLM environments and applied LLM workflows for educational research tasks
• Chapter 11 + back matter: communication strategies, conclusion, colophon, and full references

What this project is

• A multi-chapter Quarto book (.qmd) built with R
• Rendered website and PDF outputs committed under docs/
• Reproducible chapter workflows for text, network, numeric, and LLM-supported methods

Repository layout

• _quarto.yml - book configuration, chapter order, output settings
• chapter-*.qmd - chapter source files
• section-*-Intro.qmd - part introduction pages
• index.qmd - preface/front page
• conclusion.qmd, colophon.qmd, references.qmd - back matter
• references.bib, apa.csl - bibliography and citation style
• data/ - source datasets and static images used in chapters
• styles/ - custom theme/style overrides
• docs/ - rendered site and PDF for GitHub Pages
• Artwork/ - figure/table inventory and naming-tracking CSV files

Build and preview

Build full book:

quarto render

Preview with live reload:

quarto preview

Render without executing code chunks (fast layout check):

quarto render --no-execute

Render one chapter:

quarto render chapter-1.qmd

Render PDF output:

quarto render --to pdf

Authoring conventions

• Use relative paths (e.g., data/...) and do not use setwd()
• Prefer tidyverse-style R code and native pipe |>
• Use Quarto chunk options with #| syntax
• Keep narrative context above code chunks
• Keep rendered outputs in docs/ for publication

Publishing

The book is published from this repository via GitHub Pages using files in docs/.

Typical publish flow:

1. Edit .qmd source files
2. Run quarto render
3. Commit source + updated docs/
4. Push to main

Notes for collaborators

• No dedicated unit test suite is configured; render checks are the main validation.
• Internal/private review notes should stay local and not be committed.
• AI assistant config files are intentionally not required in this repository.