From stories to data stories

PSYC 81.09: Storytelling with Data

Jeremy R. Manning
Dartmouth College
Spring 2026

Today's agenda

  1. The bridge from Assignment 1 — you already know how to tell a great story
  2. How data can amplify a story — and how it can kill one
  3. Examples — Hans Rosling's Gapminder and the classic "Powers of Ten"
  4. Workshop — brainstorm your data story idea
  5. Assignment 2 release — data story remix

You already know how to tell a story

In Assignment 1, you told a 5-minute story about anything. You learned that great stories:

  • Hook the audience from the first moment
  • Create emotional connection — the audience has to care
  • Have a clear arc — setup, conflict, resolution
  • Leave the audience with a takeaway they remember

None of that changes. Assignment 2 adds one new ingredient: data.

How can data add to a story?

Data can make your story more powerful in several ways:

  • Surprise: a statistic that defies expectation is inherently compelling ("Did you know that...?")
  • Scale: numbers make abstract problems feel real ("That's 3 jumbo jets crashing every day")
  • Credibility: data transforms opinion into evidence
  • Visualization: the right chart can convey in 1 second what takes 100 words to explain
  • Narrative structure: tracking a trend over time is a story (beginning → middle → end)

The danger: death by data

The most common mistake: letting data replace the story instead of supporting it.

  • "Here's a chart. And here's another chart. And here's a table."
  • Starting with methodology instead of a question
  • Showing every analysis you ran instead of the one that matters
  • Reading numbers off a slide

Your audience came for a story, not a statistics lecture. Data should be the evidence, not the plot.

How to keep the audience with you

  • Lead with a question, not a dataset — "What if I told you that..." is always more engaging than "I downloaded this CSV..."
  • One chart, one point. If a visualization doesn't directly support your narrative, cut it.
  • Narrate the data. Don't just show a graph — walk the audience through what they should see. Keep their cognitive load low.
  • Use surprise. Find the moments in the data where something unexpected happens, and build your story around those.
  • Make it personal. Connect the data to something your audience has experienced.
  • Less is more. Three well-chosen data points beat twenty that blur together.

Hans Rosling — 200 countries, 200 years, 4 minutes

Example: Powers of Ten (1977)

Discussion: what data would strengthen your story?

Think back to your Assignment 1 story, or a new idea:

  • What question could you explore with data?
  • What would surprise your audience if they saw the numbers?
  • What visualization would make your point land instantly?
  • How could data add credibility or emotional weight to a narrative you already care about?

Sports stats, public health data, social media trends, economic indicators, environmental data, campus surveys, historical records, music streaming data, election results, scientific studies...

Workshop: pitch your data story

  1. Share your idea — what's the story, and where does data come in?
  2. Help each other find the "so what?" — why should the audience care?
  3. Identify the surprise — what's the moment where data changes everything?

Half-baked ideas are encouraged. You can pitch multiple ideas and get help choosing.

Assignment 2: data story remix

Tell a data-driven story as a 5-minute YouTube video.

  • You can remix an existing data story (new question, new angle, new visualization)
  • Or create an entirely new story built around data you find compelling
  • You can sketch, screenshot, or create your own visualizations — no coding required yet
  • The storytelling skills from Assignment 1 still apply — hook, arc, emotional connection
  • Due: Monday, April 13

Questions? Want to chat more?

📧 Email me
💬 Join our Slack
💁 Come to office hours
  • Monday: Review Assignment 2 stories + peer feedback

_class: no-content _paginate: false scale-90