Write, review, reveal

Research Skills — Week 4 Application

Today’s plan

Today’s plan

• Phase 1: Write
• Phase 2: Review
• Phase 3: The reveal
• Wrap-up

Three phases

Phase	Time	What
Write	45 min	Write your briefing, commit it
Review	30 min	Read two others’, file an Issue on each
Reveal	25 min	Who were the traitors?

• [16:05–16:10] Orientation. Remind students of the briefing structure. Traitors know who they are. Everyone else: read critically.

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.

Briefing structure reminder

1. Summary — the answer, up front
2. Evidence — figures from Weeks 2–3, with captions
3. Caveats — what depends on assumptions
4. Conclusions — your recommendation

• Everyone writes in the briefing.md they started in Week 3, on github.com (pencil icon, Preview to check). New figures: right-click → Save image as… into figures/, then commit and push.

Phase 1: Write

• Today’s plan

Phase 1: Write

✏️ 45 minutes

• Phase 2: Review
• Phase 3: The reveal
• Wrap-up

Writing block

You have until the clock says 50 minutes.

• Reuse and refine figures from Weeks 2–3
• The analytical work is done — this is about synthesis
• Traitors: be credible, not cartoonish

Hard deadline: whatever is committed at minute 50 gets reviewed.

• [16:10–16:55] Circulate for questions about content, structure, and R code. This is the main working block.
• Traitors may need reassurance: “You’re not lying. You’re writing a technically sourced briefing that happens to cherry-pick.”

Phase 2: Review

• Today’s plan
• Phase 1: Write

Phase 2: Review

✏️ 30 minutes

• Phase 3: The reveal
• Wrap-up

Peer review assignments

Each of you has two repos to review.

Navigate to the first one on GitHub now.

• [16:55–17:00] Hand out assignments. Two each, drawn at random — so some of you will get two traitors and some none. Don’t tell them that; if asked, “the draw was random” is the honest answer.

One Issue per briefing

Not one per section. The Issue is a conversation — the author replies in it. Tick a box by clicking it, or by typing an x: - [x].

• Clarity: Is the main message clear? Could a non-specialist follow?
• Evidence: Are figures honest and well-labelled? Numbers in context?
• Uncertainty: Are limitations acknowledged?

Two answers a script reads

Verdict   FAITHFUL / SUSPECT
Argues    KEEP-OR-EXPAND / KEEP-ONLY-WITH-CCS / PHASE-OUT

Pick from the drop-downs. A script reads them — this is the vote.

• Say it plainly: there is no separate poll later.
• Repos accepted before 2026-10-05 have no Peer review form under New issue. Their reviewers copy the block from the Peer review template page and type one option under each heading; if that is mangled, the review still helps the author but vanishes from the class results.

Then two required paragraphs

Why I reached that verdict — point at a figure, a claim, an omission. Not “it feels wrong”.

What made me want to believe this — answer it even if you voted SUSPECT.

• [17:00–17:30] Quiet, focused work. Circulate, and push hardest on the second paragraph — students skip it, and it is the raw material for the debrief.
• Warn them their answers may be read out, unattributed.
• At minute 80, run the harvester for the first time and keep the output. It is your fallback if the API stalls during the reveal.

Attendance

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

Phase 3: The reveal

• Today’s plan
• Phase 1: Write
• Phase 2: Review

Phase 3: The reveal

💬🎓 20 minutes

• Wrap-up

Nobody stands up

There are 93 of us. You don’t know whose briefing you read, and they don’t know you read it.

So instead: the class reads its own data.

Every verdict you filed in the last half hour is already counted.

• Say this out loud early — it defuses the anxiety that has been sitting under the writing block all session.
• No traitor is named, tonight or ever. Anyone who wants the credit can claim it themselves afterwards.

What you believed

	Actually faithful	Actually traitor
Voted FAITHFUL	correct	fooled
Voted SUSPECT	wrongly accused	caught

• [17:30–17:33] Read the two off-diagonal cells first — they are the interesting ones.
• “Wrongly accused” is usually bigger than students expect, and it is the honest briefings that looked deceptive. Flag it; you come back to it in two slides.

The split that matters

Suspicion rate, broken down by what the briefing argued.

Were you tougher on briefings that disagreed with you?

• [17:33–17:36] This is the payoff for making Argues a separate field.
• Traitors were explicitly allowed to argue the popular conclusion by dishonest means. If those slipped through at a higher rate than traitors arguing the unpopular one, that is confirmation bias with the class’s own name on it.
• Don’t soften it, and don’t blame anyone: every reviewer in the room did the same thing.

The three that got away

The least-suspected traitor briefings. Anonymized.

Which techniques survived scrutiny?

• [17:36–17:41] Pick one and dissect it against the teaching notes — the techniques you held back from the content session.
• Credit the craft. These students did a hard job well.

And the one you all accused

An honest briefing. Most-flagged in the class.

What did it do that looked like deception?

• [17:41–17:44] Usually: a missing comparator, an unlabelled axis, a confident sentence with no caveat. Nothing deliberate.
• This is the richest slide in the deck. Honest mistakes and deliberate deception are indistinguishable from the outside.

The question

Not “how did you spot them?”

“What made me want to believe it — even though I thought it might be a traitor?”

• [17:44–17:50] Start by reading out four or five harvested answers. They are already written, so nobody has to compose an opinion in front of 92 people, and hearing someone else admit it first is what unlocks the room.
• Steer away from “it was obvious once I noticed X”. You want the honest version: the briefing was good, and that is the problem.

Add yours

PollEv.com/geol

text geol to 07480 781235

• Open response, not multiple choice — the vote already happened in the Issues.
• Leave the wall up while you talk. Late arrivals keep appearing and the room keeps reading them.

How you find out

Your reviewers’ verdicts, and whether the briefings you reviewed were traitors’, will be posted as a comment on each Issue.

Not by anyone announcing it in here.

• The staff script posts the closing comment after the session.
• This is also the formative feedback loop closing: your written comment on each Issue goes in at the same time.

The real lesson

Honest mistakes and deliberate deception can look identical.

The skills that help you spot a traitor also help you avoid being one by accident.

• This is the emotional peak of Phase 1. Let it resonate.

Wrap-up

• Today’s plan
• Phase 1: Write
• Phase 2: Review
• Phase 3: The reveal

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.

What you’ve achieved

In four weeks:

• Articulated a question
• Explored data
• Tested assumptions
• Produced figures
• Written a briefing
• Reviewed someone else’s work

Phase 2 starts next week. Same skills. Your question.

• Check the Q&A — surface any top-voted questions before finishing.
• [17:50–17:55] Celebrate the accomplishment. Phase 1 is complete.
• Preview: “Starting next week, you choose the topic. Groups of 3–4. You’ll design the investigation, analyse the data, and write an individual report.”