Method and Inquiry

Claudius Gräbner-Radkowitsch

Author

1 Contributing to the Method and Inquiry

This guide is written for future contributors who want to add or update resource pages. Read it alongside CLAUDE.md if you are working with Claude Code; CLAUDE.md contains the same conventions plus instructions specific to AI-assisted editing.

1.1 What the hub is and isn’t

The hub is a general-purpose resource library — not a course companion. Resource pages should be useful to any student encountering the topic, regardless of course context. If you are writing a resource that only makes sense in the context of a specific assignment or syllabus, it does not belong here.

Course sites link into the hub. The hub does not link into course sites (except from teaching.qmd, which lists courses with external links).

1.2 Page granularity

The default unit is workflow-level: one page per meaningful task or concept a student sits down to do or understand (20–40 minutes). Titles are action- or understanding-oriented:

✓ “Reshaping Data from Wide to Long”
✓ “Understanding Statistical Power”
✗ “Data” (too broad)
✗ “Introduction to R, Chapter 3” (course-specific framing)

Short concept pages are allowed where no workflow framing makes sense — e.g., “What is Tidy Data?” or “What is Epistemology?”. Test: can a student finish this page and have accomplished or understood one specific thing?

1.3 Front matter schema

Every resource page must include this exact front matter:

---
title: ""
description: ""
author: "Claudius Gräbner-Radkowitsch"
date: YYYY-MM-DD
date-modified: YYYY-MM-DD
image: ../../assets/images/placeholder.png

learning-objectives:
  - "Understand what tidy data is and why it matters"
  - "Use pivot_longer() to reshape a wide dataset"

categories:
  - [at least one thematic tag]

level:
  - BA          # BA | MA | PhD — list all that apply

content-type:   # exactly one of: tutorial | video | slides | reading | exercise | concept

citation:
  id: graebner2026-[slug]
  type: entry-encyclopedia
  container-title: "Method and Inquiry"
  editor:
    - name: "Claudius Gräbner-Radkowitsch"
  url: https://euf-methodology-hub.netlify.app/resources/[slug]
---

citation rules: - type must be entry-encyclopedia. It is the only CSL type that carries container-title through to the generated BibTeX (as booktitle). With entry, webpage, or article, “Method and Inquiry” appears in the human-readable citation but is silently dropped from the BibTeX - id sets the BibTeX key and is mandatory. Without it every page on the hub produces the same key, so anyone citing two pages gets a collision. Use graebner2026-[slug] (or your own surname and the page’s year) - url uses the flat form /resources/[slug] — there are no category subfolders

learning-objectives rules: - 2–4 objectives; more than 4 signals the page scope is too broad - Each objective is a concrete skill or understanding, not a description of page content - Write from the student’s perspective: “Explain X”, “Apply X to Y”, “Distinguish X from Y” - ✓ “Use pivot_longer() to reshape a wide dataset” - ✗ “This page covers data reshaping in R”

author vs editor: - If you are a contributor writing a new page, put your name in author - The editor field lives inside the citation: block and is always Claudius Gräbner-Radkowitsch — it reflects editorial responsibility for the hub, not who wrote the page - Do not add a top-level editor: key — Quarto 1.9+ reserves that key for visual editor configuration and will reject a name string

1.4 Tag taxonomy

Use existing tags. Adding a new thematic tag requires also adding it to the taxonomy in CLAUDE.md.

Thematic: data-cleaning, R, visualization, experiments, statistics, regression, causal-inference, literature-review, research-design, qualitative-methods, reproducibility, quarto, survey-methods, epistemology, tidyverse

Content type (pick exactly one): tutorial, video, slides, reading, exercise, concept

Level (list all that apply): BA, MA, PhD

1.5 Level differentiation

Pages are written at BA level by default. Two patterns are allowed for adding MA/PhD content.

Pattern 1 — inline callout (default)

Use when the page is fundamentally the same workflow for all levels, with added depth for MA/PhD. Wrap MA/PhD-specific content in:

::: {.callout-note .level-advanced}
## 🎓 MA / PhD
Content here.
:::

This renders with a muted teal left border and lighter background, visually distinct from a standard note.

Pattern 2 — separate pages per level

Use when the conceptual treatment genuinely differs across levels — not just in depth but in framing, assumed knowledge, or purpose. Examples: - An epistemology page introducing the concept from scratch (BA) vs. engaging with the literature (PhD) - A statistics page using intuition and visuals (BA) vs. formal notation (PhD)

When using Pattern 2: - Each page is a standalone .qmd with its own front matter - The level field contains only the single applicable level - Add a sibling-link callout at the top of each page:

::: {.callout-tip}
## 📖 Level: BA
This page is written for BA students. There is also a version for
[MA/PhD students](../epistemology-phd.qmd).
:::

File naming: [topic]-ba.qmd, [topic]-ma.qmd, [topic]-phd.qmd
All sibling pages share the same thematic tags so they appear together in the listing

Deciding: Default to Pattern 1. Switch to Pattern 2 only when the callout block would be so large and structurally different that it dominates the page.

1.6 File and folder naming

Resource pages: kebab-case.qmd matching the page title closely
Category folders: kebab-case/ (e.g., r-programming/, literature-review/)
Track pages: kebab-case.qmd in tracks/
Data files: kebab-case.csv / .rds etc.
Images: kebab-case.png in assets/images/

1.7 Data files

In the repo: synthetic/self-generated data, small modified external datasets, small freely redistributable datasets
On OSF (linked, not hosted): large datasets, verbatim third-party data, datasets where acquisition is part of the learning
Data files live co-located with the resource page in a data/ subfolder within the category folder — never in a global /data folder
R scripts go in .qmd code blocks by default; standalone .R files only when genuinely needed

Example structure:

resources/
  data-cleaning/
    reshaping-data-wide-to-long.qmd
    data/
      survey-responses.csv

1.8 Relative paths

Never hardcode the domain URL in .qmd files. Use relative paths for all internal links. This keeps the site portable when the domain changes.

1.9 Build and deploy workflow

While writing a single page:

quarto preview resources/[category]/[page].qmd

After finishing a page:

quarto render resources/[category]/[page].qmd
quarto render resources/index.qmd

Publish to the live site — double-click Update.command in the repo root. It works out which sources changed since the last deployment, renders, and uploads the result to Netlify. From a terminal the equivalent is:

quarto render && quarto publish netlify --no-render

Commit and push — a separate step. _site/ is git-ignored (it is uploaded to Netlify, not committed); _freeze/ is committed, so that a fresh clone does not have to re-execute every R chunk:

git add -A
git commit -m "add: [short description]"
git push

Netlify is not connected to this repository and never runs a build — it only receives the _site/ folder that quarto publish uploads. Pushing to GitHub does not deploy, and publishing does not commit. Do both.

Citation

BibTeX citation:

@online{graebner2026-method-and-inquiry,
  author = {Gräbner-Radkowitsch, Claudius},
  title = {Method and {Inquiry}},
  url = {https://euf-methodology-hub.netlify.app/CONTRIBUTING.html},
  langid = {en}
}

For attribution, please cite this work as:

Gräbner-Radkowitsch, Claudius. n.d. “Method and Inquiry.” https://euf-methodology-hub.netlify.app/CONTRIBUTING.html.

# Contributing to the Method and Inquiry This guide is written for future contributors who want to add or update resource pages. Read it alongside `CLAUDE.md` if you are working with Claude Code; `CLAUDE.md` contains the same conventions plus instructions specific to AI-assisted editing. --- ## What the hub is and isn't The hub is a **general-purpose resource library** — not a course companion. Resource pages should be useful to any student encountering the topic, regardless of course context. If you are writing a resource that only makes sense in the context of a specific assignment or syllabus, it does not belong here. Course sites link *into* the hub. The hub does not link *into* course sites (except from `teaching.qmd`, which lists courses with external links). --- ## Page granularity The default unit is **workflow-level**: one page per meaningful task or concept a student sits down to do or understand (20–40 minutes). Titles are action- or understanding-oriented: - ✓ "Reshaping Data from Wide to Long" - ✓ "Understanding Statistical Power" - ✗ "Data" (too broad) - ✗ "Introduction to R, Chapter 3" (course-specific framing) **Short concept pages** are allowed where no workflow framing makes sense — e.g., "What is Tidy Data?" or "What is Epistemology?". Test: *can a student finish this page and have accomplished or understood one specific thing?* --- ## Front matter schema Every resource page must include this exact front matter: ```yaml --- title: "" description: "" author: "Claudius Gräbner-Radkowitsch" date: YYYY-MM-DD date-modified: YYYY-MM-DD image: ../../assets/images/placeholder.png learning-objectives: - "Understand what tidy data is and why it matters" - "Use pivot_longer() to reshape a wide dataset" categories: - [at least one thematic tag] level: - BA # BA | MA | PhD — list all that apply content-type: # exactly one of: tutorial | video | slides | reading | exercise | concept citation: id: graebner2026-[slug] type: entry-encyclopedia container-title: "Method and Inquiry" editor: - name: "Claudius Gräbner-Radkowitsch" url: https://euf-methodology-hub.netlify.app/resources/[slug] --- ``` **`citation` rules:** - `type` must be `entry-encyclopedia`. It is the only CSL type that carries `container-title` through to the generated BibTeX (as `booktitle`). With `entry`, `webpage`, or `article`, "Method and Inquiry" appears in the human-readable citation but is silently dropped from the BibTeX - `id` sets the BibTeX key and is mandatory. Without it every page on the hub produces the same key, so anyone citing two pages gets a collision. Use `graebner2026-[slug]` (or your own surname and the page's year) - `url` uses the flat form `/resources/[slug]` — there are no category subfolders **`learning-objectives` rules:** - 2–4 objectives; more than 4 signals the page scope is too broad - Each objective is a concrete skill or understanding, not a description of page content - Write from the student's perspective: "Explain X", "Apply X to Y", "Distinguish X from Y" - ✓ "Use `pivot_longer()` to reshape a wide dataset" - ✗ "This page covers data reshaping in R" **`author` vs `editor`:** - If you are a contributor writing a new page, put your name in `author` - The `editor` field lives inside the `citation:` block and is always `Claudius Gräbner-Radkowitsch` — it reflects editorial responsibility for the hub, not who wrote the page - Do not add a top-level `editor:` key — Quarto 1.9+ reserves that key for visual editor configuration and will reject a name string --- ## Tag taxonomy Use existing tags. Adding a new thematic tag requires also adding it to the taxonomy in `CLAUDE.md`. **Thematic:** `data-cleaning`, `R`, `visualization`, `experiments`, `statistics`, `regression`, `causal-inference`, `literature-review`, `research-design`, `qualitative-methods`, `reproducibility`, `quarto`, `survey-methods`, `epistemology`, `tidyverse` **Content type** (pick exactly one): `tutorial`, `video`, `slides`, `reading`, `exercise`, `concept` **Level** (list all that apply): `BA`, `MA`, `PhD` --- ## Level differentiation Pages are written at **BA level by default**. Two patterns are allowed for adding MA/PhD content. ### Pattern 1 — inline callout (default) Use when the page is fundamentally the same workflow for all levels, with added depth for MA/PhD. Wrap MA/PhD-specific content in: ```markdown ::: {.callout-note .level-advanced} ## 🎓 MA / PhD Content here. ::: ``` This renders with a muted teal left border and lighter background, visually distinct from a standard note. ### Pattern 2 — separate pages per level Use when the conceptual treatment genuinely differs across levels — not just in depth but in framing, assumed knowledge, or purpose. Examples: - An epistemology page introducing the concept from scratch (BA) vs. engaging with the literature (PhD) - A statistics page using intuition and visuals (BA) vs. formal notation (PhD) When using Pattern 2: - Each page is a standalone `.qmd` with its own front matter - The `level` field contains only the single applicable level - Add a sibling-link callout at the top of each page: ```markdown ::: {.callout-tip} ## 📖 Level: BA This page is written for BA students. There is also a version for [MA/PhD students](../epistemology-phd.qmd). ::: ``` - File naming: `[topic]-ba.qmd`, `[topic]-ma.qmd`, `[topic]-phd.qmd` - All sibling pages share the same thematic tags so they appear together in the listing **Deciding:** Default to Pattern 1. Switch to Pattern 2 only when the callout block would be so large and structurally different that it dominates the page. --- ## File and folder naming - Resource pages: `kebab-case.qmd` matching the page title closely - Category folders: `kebab-case/` (e.g., `r-programming/`, `literature-review/`) - Track pages: `kebab-case.qmd` in `tracks/` - Data files: `kebab-case.csv` / `.rds` etc. - Images: `kebab-case.png` in `assets/images/` --- ## Data files - **In the repo:** synthetic/self-generated data, small modified external datasets, small freely redistributable datasets - **On OSF (linked, not hosted):** large datasets, verbatim third-party data, datasets where acquisition is part of the learning - Data files live **co-located with the resource page** in a `data/` subfolder within the category folder — never in a global `/data` folder - R scripts go in `.qmd` code blocks by default; standalone `.R` files only when genuinely needed Example structure: ``` resources/ data-cleaning/ reshaping-data-wide-to-long.qmd data/ survey-responses.csv ``` --- ## Relative paths Never hardcode the domain URL in `.qmd` files. Use relative paths for all internal links. This keeps the site portable when the domain changes. --- ## Build and deploy workflow **While writing a single page:** ```bash quarto preview resources/[category]/[page].qmd ``` **After finishing a page:** ```bash quarto render resources/[category]/[page].qmd quarto render resources/index.qmd ``` **Publish to the live site** — double-click `Update.command` in the repo root. It works out which sources changed since the last deployment, renders, and uploads the result to Netlify. From a terminal the equivalent is: ```bash quarto render && quarto publish netlify --no-render ``` **Commit and push** — a separate step. `_site/` is git-ignored (it is uploaded to Netlify, not committed); `_freeze/` *is* committed, so that a fresh clone does not have to re-execute every R chunk: ```bash git add -A git commit -m "add: [short description]" git push ``` Netlify is not connected to this repository and never runs a build — it only receives the `_site/` folder that `quarto publish` uploads. Pushing to GitHub does not deploy, and publishing does not commit. Do both.