First tests

Research Skills — Week 6 Application

Today’s plan

Today’s plan

• t-test workflow
• Group work
• Discussion
• Wrap-up

Goals

You have a research plan and a hypothesis. Today: test it.

1. Load and clean your project data
2. Visualize before you test
3. Check assumptions
4. Run the t-test
5. Interpret — honestly

• [16:05–16:15] Setup: pull latest changes from group repo. Recap.

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.

t-test workflow

• Today’s plan

t-test workflow

🎓💻 Guided example

• Group work
• Discussion
• Wrap-up

The complete workflow

1. Load data
2. Visualize: histograms, violins
3. Check assumptions: shapiro.test(), QQ plot
4. Run: t.test(y ~ group, data = df)
5. Interpret: CI, effect size, plain English
6. If violated: log-transform or wilcox.test()

• [16:15–16:35] Work through the full pipeline live in the cell — type it, don’t paste it. The list above is on screen for the students too: tell them they have the same information you do, and invite them to beat you to step 4. Start typing a minute in.

• Example data is the borehole set from the Week 6 content session, so the numbers are ones they have already seen.

• Anticipated code:

  borehole <- read.csv("data/borehole_temp.csv")
  
  ggplot(borehole, aes(x = formation, y = temperature_c)) +
    geom_violin() +
    geom_point()
  
  shapiro.test(borehole$temperature_c[borehole$formation == "Stainmore"])
  shapiro.test(borehole$temperature_c[borehole$formation == "Whin Sill"])
  
  t.test(temperature_c ~ formation, data = borehole)
  
  t.test(log(temperature_c) ~ formation, data = borehole)

• The numbers, so you can narrate without reading them off: raw data gives t = 2.24, df = 47.2, p = 0.030, 95% CI [0.86, 15.97], means 35.3 vs 26.9 °C. Shapiro-Wilk on Stainmore is p = 0.032 — the assumption fails. On the log scale p = 0.22 and the CI crosses zero.

• Emphasize: always visualize before you test. Let the room reach step 4 and celebrate the “significant” result before you run step 3’s verdict against it.

The golden rule

Always visualize before you test.

If you haven’t seen the data, you can’t interpret the result.

Group work

• Today’s plan
• t-test workflow

Group work

✏️💻 Apply to your data

• Discussion
• Wrap-up

Your workflow

Work through these steps on your project data:

1. Load your project data. Make sure it’s clean.
2. Define your groups / comparison.
3. Visualize. What does it look like?
4. Check assumptions.
5. Run the t-test.
6. Interpret.
7. Write up a paragraph of results.

Finished early? Try a second comparison or explore subgroups.

• [16:35–17:25] Groups work on their data. Circulate with demonstrators.
• Many groups will spend most of this on data cleaning. That’s realistic — say so: “In real research, 80% of the time is data preparation.”
• If a group’s data don’t suit a t-test (more than 2 groups, categorical variable), help them preview Week 7’s ANOVA.

Attendance

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

Discussion

• Today’s plan
• t-test workflow
• Group work

Discussion

💬 What did you find?

• Wrap-up

Share your results

2–3 groups share:

• What was the hypothesis?
• What was the p-value?
• What was the effect size?
• Do you believe the result?

• [17:25–17:40] For each group, model the conditional-thinking habit: “How plausible was this before you tested? Does the p-value change your mind a little or a lot?”

The question to always ask

“How plausible was this before we tested?”

Does the p-value change your mind a little — or a lot?

Wrap-up

• Today’s plan
• t-test workflow
• Group work
• 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

Commit analysis code and written results to your group repo via PR.

Next week: “What if you have more than two groups?”

ANOVA, multiple comparisons, and effect sizes.

• Check the Q&A — surface any top-voted questions before finishing.
• Encourage intermediate commits, even rough code. The commit history is part of the learning.