Making the biomass case

Research Skills — Week 2 Application

Today’s plan

Today’s plan

• Chart audit
• Guided walkthrough
• Your figures
• Discussion
• Wrap-up

Goals

Produce briefing-quality figures of the biomass data.

By the end of this session:

• 2–3 committed figures in your repo
• The code that produced them
• A sense of the story the data tells

• [16:05–16:15] Setup: open WebR, open GitHub Desktop, pull latest.
• Today is about producing the figures for the Phase 1 briefing. Less new content, more hands-on making.

Questions?

Submit questions:

PollEv.com/geol

text geol to 07480 781235

• Open the Poll Everywhere Q&A activity before the session and leave it active throughout.
• Questions appear in your Poll Everywhere presenter view; upvoted questions rise to the top.
• Address during transitions or at the end of concept blocks, not mid-explanation.
• Reset (archive) the activity between sessions so questions don’t carry over from the previous week.

The figures you’re building

Figure	What it shows
Trend plot	Biomass generation over time
Comparison plot	CO₂ per MWh: biomass vs coal vs gas
Import plot	Where do the wood pellets come from?
Context plot	Biomass as % of total UK electricity

• These correspond to the four main exercises. Not every student will finish all four — that’s fine. One good figure is better than four rushed ones.

Chart audit

• Today’s plan

Chart audit

💬 Homework follow-up

• Guided walkthrough
• Your figures
• Discussion
• Wrap-up

Hall of shame: the satisfaction pyramid

Stack Overflow Developer Survey 2026

Volume encodes nothing: the 45% tier looks many times the size of the 22% one. And “not happy” sits between happy and complacent.

• [16:15–16:20 for the whole section] Before the session, open the Chart audit category and count posts per chart (titles start [Chart N]). Lead with the most-posted; skip any that drew nothing.
• Ask the author of a good post to say what they spotted, rather than reading the critique off the slide.
• Fix: three flat bars, in scale order: happy, complacent, not happy.

Hall of shame: two cubes

Stack Overflow Developer Survey 2026

Two cubes the same size, for 32% and 30%. Where are the other 38%? And only 14,428 of the 30,903 respondents answered.

• The missing 38% are neutral (15%), somewhat unfavourable (15%), very unfavourable (7%) and not sure (1%). They’re on the Dig deeper page.
• Compared to what? “62% like AI” means 62% of those who answered this question, under half the survey.

Hall of shame: the treemap

Stack Overflow Developer Survey 2026

The parts add up to 296%. Respondents could pick several languages, so these aren’t shares of a whole.

• A treemap, like a pie, claims to divide one whole. Multi-select answers can’t. Same lesson as the double-counted pie in the content session.
• Fix: sorted bars, each out of 100%.

Hall of shame: the language pyramid

Stack Overflow Developer Survey 2026

52% sits on top of 63%, and the tiers shrink far faster than the numbers. A pyramid implies a hierarchy that isn’t there.

• The five values span 52–63%. The drawing turns an 11-point spread into a sliver against a slab.
• Multi-select again: a pyramid of parts is meaningless when the parts overlap.

The same data, done properly

Stack Overflow Developer Survey 2026, Dig deeper: AI usage

Every answer shown, in scale order, every bar labelled, n stated. Their caption even says: “This is not proof … but they are certainly related.”

• The two cubes, redrawn by Stack Overflow themselves one click away.
• The quoted caption is about experience and favourability: correlation isn’t causation, and they say so. Credit where it’s due.
• Segue: “Now make yours look like this, not like the landing page.”

Guided walkthrough

• Today’s plan
• Chart audit

Guided walkthrough

🎓💻 Building a biomass figure

• Your figures
• Discussion
• Wrap-up

Live demo

🖥️ Building a multi-fuel line chart in WebR

• [16:20–16:35] Build one complete ggplot on screen. Suggested plot: UK electricity by fuel type over time (line chart, colour by fuel).
• Walk through: load data → reshape if needed (pivot_longer) → ggplot → add labels → add caption. Narrate each step.
• Emphasize: “This is the kind of figure that goes in a policy briefing. Notice the caption tells the reader what to see.”

The anatomy of a good figure

Every figure should have a “so what” — what should the reader take from it?

• Build this live, one layer at a time. Don’t paste it. The students have the same data open on the exercise page — tell them to see whether they can beat you to a legible four-fuel chart, and start typing a minute in so the slower ones can follow.

• Anticipated code (the same solution as Exercise 1 on the exercise page, so the two agree):

  elec <- read.csv("data/uk_electricity.csv")
  
  elec_long <- pivot_longer(
    elec,
    cols = c(coal_twh, gas_twh, wind_twh, bioenergy_twh),
    names_to = "fuel",
    values_to = "twh"
  )
  
  ggplot(elec_long, aes(x = year, y = twh, colour = fuel)) +
    geom_line() +
    labs(
      x = "Year",
      y = "Generation (TWh)",
      colour = "Fuel type",
      caption = "Source: DUKES 2026 Table 5.6.B"
    )

• Suggested build order: plot one fuel with geom_line() first, hit the wall of needing a second geom_ per fuel, then reshape. The reshape earns its place that way.

• Show the completed plot. Ask: “What’s the story here?” Students should see biomass rising as coal falls.

Your figures

• Today’s plan
• Chart audit
• Guided walkthrough

Your figures

✏️💻 Independent exercises

• Discussion
• Wrap-up

Work through the exercises

The WebR page has four main exercises with hints:

1. Trend plot — biomass over time, with context
2. Comparison plot — emissions per MWh by fuel
3. Import plot — pellet origins over time
4. Context plot — biomass as % of total

Extension: Can you make biomass look more important — or less important — than it really is?

• [16:35–17:25] The bulk of the session. Circulate actively.
• Common issues: wrong column names, wide vs long format confusion, forgetting to load ggplot2, aesthetic mapping errors.
• The extension is a preview of Week 4. Don’t belabour it, but students who try it will be ahead of the game.

Attendance

Show the QR from https://attendance.durham.ac.uk/mytimetable, then ← back to the task.

Discussion

• Today’s plan
• Chart audit
• Guided walkthrough
• Your figures

Discussion

💬 What story do your figures tell?

• Wrap-up

Show and tell

“What story do your figures tell?”

2–3 volunteers: share a figure. Class discusses:

• What does this figure show clearly?
• What’s missing?
• How do your choices (axis ranges, colours, what’s included) affect the impression?

• [17:25–17:40] This is important: first time students see each other’s choices and recognize that the same data can produce different impressions. Primes Week 4.
• Highlight specific choices: “You started the y-axis at 0 — good. What would happen if you started at 30?” “You showed all fuels — what if you only showed biomass and coal?”

Wrap-up

• Today’s plan
• Chart audit
• Guided walkthrough
• Your figures
• Discussion

Wrap-up

Any questions we missed?

Submit questions:

PollEv.com/geol

text geol to 07480 781235

• Continuation of the same Poll Everywhere Q&A activity opened at the start of the session — this is not a new poll, so don’t reset/archive it yet.
• Surface any questions that have come in since the early Questions slide, especially ones raised during today’s content.
• Once you’ve finished with this slide, archive (reset) the activity at the end of the session so questions don’t carry over.

Commit your work

Save your code to week2.R. Commit and push.

Your repo now has two weeks of work:

• week1.R — exploratory code
• week2.R — figures

Next week: “How confident should we be?”

We’ll ask whether the biomass numbers really add up.

• Check the Q&A — surface any top-voted questions before finishing.
• Ensure everyone commits. Students who haven’t finished can continue at home — localStorage preserves their WebR code.