W14 practical (17 Jul) — written submission due 2026-07-22. Submit link to your rendered group-reflection.qmd page (within your project repo) on Moodle.
In-class discussion (W14 practical)
In the final session, you will discuss with other groups your project experiences in class. This is not a formal presentation of results — it is a structured conversation about what you learned, what surprised you, and what you would do differently.
Come prepared to speak to the following:
About your data and analysis
- What did you learn about your dataset that you did not expect when you chose it?
- Which analytical decision was hardest to make, and how did you reason through it?
- What is one thing your analysis cannot answer, and why?
About your workflow and collaboration
- What worked well in how your group coordinated? What was harder than expected?
- How did you use LLM assistance during the project — what was useful, what was not, and when did you have to override or correct it?
About your own skill development
- What is one concept or tool from the course that made more sense once you applied it to your own data?
- What would you want to learn next, given what this project revealed?
- One error or mistake in LLM output that you caught
The reflection is also good preparation for the oral exam: the questions above are representative of the kinds of reasoning you will be asked to demonstrate.
Written submission
Add a group-reflection.qmd to your project repository and render it as part of your Quarto website. It does not need to be polished — honest and specific answers are more useful than complete ones. This document should summarise things you’ve already documented in your git commit messages, GitHub issues, Contributions and AI disclosure statements.
group-reflection.qmd
---
title: "Group Reflection"
---
## Skills inventory
For each skill area, describe what your group actually used it for.
If you did not use a skill, say so and note why.
### Functions & refactoring (W2)
*What we did:*
*What worked well:*
*What was hard:*
### Debugging (W3)
*What we did:*
*What worked well:*
*What was hard:*
### Version control & remotes (W4)
*What we did:*
*What worked well:*
*What was hard:*
### Quarto websites & collaborative coding (W5)
*What we did:*
*What worked well:*
*What was hard:*
### R packages (W6)
*What we did:*
*What worked well:*
*What was hard:*
### Initial data analysis & data cleaning (W7)
*What we did:*
*What worked well:*
*What was hard:*
### Reproducibility, open data & renv (W9)
*What we did:*
*What worked well:*
*What was hard:*
### Analysis & inference strategies (W10)
*What we did:*
*What worked well:*
*What was hard:*
### Communication & visualisation (W11)
*What we did:*
*What worked well:*
*What was hard:*
## Key analytical decisions
Describe 2–3 decisions that had a meaningful effect on your results or workflow.
### Decision 1: [short title]
*What were we deciding between?*
*How did we reason through it?*
*In hindsight, would we decide differently?*
### Decision 2: [short title]
*What were we deciding between?*
*How did we reason through it?*
*In hindsight, would we decide differently?*
## Workflow & collaboration
*How did we divide work across group members?*
*What branching or Git workflow did we actually use — and did it match the proposal?*
*What would we change about how we coordinated?*
## LLM usage
Describe up to 5 significant uses of an LLM tool during the project.
### Usage 1: [task]
- **Tool / model:**
- **What it produced:**
- **What we had to correct or verify:**
### Usage 2: [task]
- **Tool / model:**
- **What it produced:**
- **What we had to correct or verify:**