Welcome to Part II

PSYC 81.09: Storytelling with Data

Welcome to Part II!

The structured learning phase is done. From here on, you drive the content.

Part I gave you tools for telling and evaluating stories, plus hands-on experience with Python's data science stack. Now it's time to put those skills to work on questions you care about.

The Part II cycle

We'll cycle through these stages continuously:

  1. Pitch & brainstorm — present story ideas to the class, form project groups, and begin homing in on questions and datasets
  2. Refine — workshop your ideas and code, work through implementation challenges, and learn new tools as needed
  3. Critique — share your finished stories, give and receive constructive feedback, and discuss what worked and what didn't

Different groups will move through these stages at different rates — and that's by design.

Weekly rhythm

Day Focus
Monday Review and critique recent data stories
Wednesday New tools, demos, and workshops
Thursday Office hours or hackathon sessions
Friday Hackathon + brainstorming new ideas

This schedule is flexible — some weeks we'll spend more time on one stage than another, depending on where the class is in the cycle.

What makes a great data story project?

  • A compelling question — something that makes people curious
  • Real data — publicly available datasets that speak to your question
  • Thoughtful analysis — appropriate methods, honest about limitations
  • Clear narrative — a beginning, middle, and end that a non-expert could follow
  • Effective visualization — figures that communicate, not just decorate

The best stories don't need the fanciest methods — they need the clearest thinking.

Resources available to you

  • Dartmouth AI tools — Claude, GitHub Copilot, and other AI assistants available through Dartmouth
  • The Python data science stack — NumPy, Pandas, Matplotlib, Seaborn, Hypertools, and more
  • Each other — your classmates are your most valuable resource for feedback and collaboration
  • Office hours — bring questions, bugs, or half-formed ideas
  • Slack — post questions anytime; chances are someone else has the same one

Expectations

Produce at least 3 data stories over the remaining weeks. For each story:

  • A Jupyter notebook with your analysis (runnable in Google Colab)
  • A 5-minute video presenting your story
  • A README describing your project

You can work solo or in groups of any size. Submit by making pull requests to the course repository, using the same demo template from Assignment 4.

Let's get started!

Let's brainstorm. Think about:

  • What questions interest you? What are you curious about?
  • What datasets would you explore? Check out Kaggle, FiveThirtyEight, or Awesome Public Datasets for inspiration
  • What kind of story do you want to tell? Persuasive? Exploratory? Surprising?

Some strategies that work well: start with a question and find data, start with a dataset and form questions, or find a visualization you want to create and work backward.

Questions? Want to chat more?

📧 Email me
💬 Join our Slack
💁 Come to office hours
  • Friday: Hackathon + brainstorming