Deepening your analysis

Research Skills — Week 7 Application

Today’s plan

Today’s plan

• Brief guided example
• Group work
• Cross-group check-in
• Wrap-up

A flexible session

Today is about whatever your project needs most.

Track A: ANOVA

For groups with 3+ categories.

ANOVA → TukeyHSD → effect sizes.

Track B: Refine t-test

For groups with 2-group comparisons.

Assumptions → transform → Cohen’s d → CI.

• [16:05–16:15] Setup. Pull latest changes. Today should feel like a supervised lab, not a structured lecture.

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.

Goals for everyone

By the end of today:

1. Run and interpret your main statistical test
2. Check assumptions
3. Calculate and report an effect size
4. Write a results paragraph for your report

• Students who finish: secondary analysis, subgroup exploration, or start producing report-quality figures.

Brief guided example

• Today’s plan

Brief guided example

🎓💻 Quick demo

• Group work
• Cross-group check-in
• Wrap-up

Complete analysis in 10 minutes

Track A: ANOVA → TukeyHSD → η²

Track B: t-test → QQ plot → log-transform → Cohen’s d + CI

• [16:15–16:30] Keep it brief. One complete worked example, typed live, not pasted. Pick whichever track most of the room needs. Students have the same arrows on screen — invite them to run ahead.

• Anticipated code, Track A (the wind data from Week 7 content):

  wind <- read.csv("data/wind_output.csv")
  
  model <- aov(capacity_pct ~ region, data = wind)
  summary(model)
  
  TukeyHSD(model)
  
  tab <- summary(model)[[1]]
  tab[1, "Sum Sq"] / sum(tab[, "Sum Sq"])   # eta squared

• Track A numbers: F(3, 96) = 14.2, p = 9.8e-08, η² = 0.31. Tukey gives two clusters — Scotland vs Northern England is the one non- significant pair (p = 0.99); every comparison across the north/south divide is significant. Read the two-cluster story off the adjusted p-values, not off the F.

• Anticipated code, Track B (the borehole data from Week 6):

  borehole <- read.csv("data/borehole_temp.csv")
  
  t.test(temperature_c ~ formation, data = borehole)
  qqnorm(borehole$temperature_c); qqline(borehole$temperature_c)
  t.test(log(temperature_c) ~ formation, data = borehole)

• Track B numbers: p = 0.030 raw, p = 0.22 on the log scale.

• Students need maximum work time today — if the demo runs long, cut the effect size and let them compute it in their own analysis.

Group work

• Today’s plan
• Brief guided example

Group work

✏️💻 Your analysis

• Cross-group check-in
• Wrap-up

Work time

55 minutes. Work on your project.

We’ll be circulating. If you need help, raise a hand.

Challenge questions we’ll ask:

• What’s your effect size?
• What’s the confidence interval?
• Would you make a policy recommendation based on this?

• [16:30–17:25] Circulate actively. Priority: help groups who are stuck; challenge groups who are coasting.

Attendance

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

Cross-group check-in

• Today’s plan
• Brief guided example
• Group work

Cross-group check-in

💬 30-second updates

• Wrap-up

What did you test? What did you find?

Each group: 30 seconds.

• What you tested
• What you found
• What you’re worried about

• [17:25–17:40] Quick round-robin. Instructor highlights common themes across groups (e.g., “Several of you have skewed data — remember to check and transform”).

Wrap-up

• Today’s plan
• Brief guided example
• Group work
• Cross-group check-in

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

Commit analysis code and results to your group repo.

Accept the report assignment: your individual report goes in its own repo.

Next week: “Models and their limits”

What is a model? What can go wrong when you build one?

And a new question to add to your collection:

“What is the model not capturing?”

• Check the Q&A — surface any top-voted questions before finishing.
• By the end of this session, every group should have run at least one statistical test with effect size and assumption checks.