Project

• Deadline: 9 January 2026, 23:59 (Europe/Stockholm)
• Presentations: 16 January 2026, 12.00-17.00

Overview

Your individual project is a blog-style data story based on a unique dataset you collect yourself. Use the course tools to obtain, transform, explore, visualize, and explain your data.

Submission: push your work to your course GitHub project repository and open an issue. You may push early and often; we will grade the version that exists at the deadline.

Presentation: a max 5-minute talk in front of the class. Attendance is compulsory for the full session in which you present.

Inspiration

A few examples of scope and style:

• The Olympic Medal Table Visualized Gapminder Style
• On a First Name Basis with Statistics Sweden
• Baby Weight Shiny app
• Are #python users more likely to get into Slytherin?

Data

During the course you’ve seen many ways to source data. Additional public, web-based sources include:

• Stockholm Open Data Portal
• APIs listed on Sweden’s national data portal
• COVID-19 data (Folkhälsomyndigheten)

We strongly recommend using data you personally collected and genuinely care about. Do not use sensitive data. Your raw data must be legally shareable on GitHub (no licensing or privacy violations).

What to do

1. Find data “in the wild.” This could be an open SQL database, an API, web-scraped data, personal logs (running watch, app exports), or hobby data.
2. Decide on a story. Examples: test a simple hypothesis, create a compelling visualization, write an educational explainer, or do a small piece of data journalism. Define your audience up front (general public, BSc students, ornithologists, …).
3. Analyze. Read → wrangle → visualize → summarize. Keep the story from step 2 in focus and interpret results accordingly.
4. Write the post. Swedish or English. Target ~1000–1500 words, ≤ 7 figures (tables count as figures), and ≤ 7 visible code chunks (it’s fine to hide non-essential code). The post must be reproducible.
5. Present. Prepare a 5-minute talk whose goal is to convince your classmates to read your post.

Pro tip: Be realistic. Whatever you think you can do—plan for half of that and you’ll still be busy. Get a minimal, working version early and iterate.

Submission & Reproducibility

• Push to your project course repo and open an issue to signal submission.

• Include:

  • A clear notebook (Jupyter/R Markdown) that reproduces your analysis end-to-end.

  • The data you used (or a script that downloads it), and brief notes on source/licensing.

  • A short README with: title, audience, how to run, and any package requirements.

    If you used extra packages, list them at the top of the notebook or in a simple requirements.txt/install note.

• We will clone your repo and run the notebook. Make sure it runs cleanly from a fresh checkout.

Grading

Each dimension has equal weight:

1. Technical difficulty. Data access/import effort, wrangling complexity, appropriate use of methods.
2. Code quality & reproducibility. Readable code, sensible structure, “runs from scratch.”
3. Visualization quality. Appropriate choices, clarity, and correct interpretation.
4. Writing & storytelling. Clear aims, concise narrative, audience-appropriate tone, spelling/grammar.
5. Presentation. Slide quality (if any), engagement, timing (≤ 5 minutes).