Project Proposals
Week 4 · Tuesday, Oct 20 · Presentations
Today’s plan
| Time | What |
|---|---|
| 1:30 | DS 3 discussion check, on paper |
| 1:50 | How the proposals work, and the running order |
| 1:55 | Proposals: 9 groups, back to back |
| 2:55 | What happens next |
Today’s goals: close out DS 3, and every group leaves with feedback on its question, its data and its plan, early enough to change course if needed.
DS 3 discussion check
Warm-up, in pairs
⏱ 5 min
Swap printouts with your neighbor. Check all their answers: are they similar to yours? If not, how so? Pick the concept that was harder to understand in all the practice.
Proposals
How each proposal runs
| Minutes | Who | |
|---|---|---|
| Presentation | 3 | the team |
| Questions | about 3 | I ask; each partner answers separately |
| Switch | under 0.5 | next team walks up while I write notes |
- Your slides are already loaded, from the PDF you uploaded. You click, I keep time
- At 2:30 I raise a card. At 3:00 I stop you, mid-sentence if needed. That’s fair to the teams after you
- No AI, no notes on your phone, no screens other than your slides
What you cover
- Question: the data-driven question, in one or two sentences
- Data: what, where it comes from, how big, which variables. The first rows
- Plan: which plots and which model(s), and why
- Risks: what could go wrong
- Team: alone or a pair
Last slide: your AI-use note. “None” is a complete answer.
What I look for
A question the data can actually answer.
- A dataset you have already opened, loaded by a script
- A plan that is realistic for the rest of the quarter
- Risks you’ve actually found, not generic ones
- Both partners can talk about every part of it
This is 10% of the project and the first, roughest version of it. I’m not expecting results. I’m expecting a question, data in R, and a plan.
The kind of question you’ll get
About the question
- Which variable in your data is the answer to your question? Point to it in the first rows
- What result would surprise you?
About the data
- How many rows do you have, and what is one row?
- Which column do you expect to need cleaning first? How will you check it?
About the plan
- Why that plot, and not another one? One sentence
- What would you do if the model you planned doesn’t fit?
Each partner gets a different question. “I don’t know yet, but here’s how I’d find out” is a good answer.
Your job when you’re not presenting
For every team, on the feedback sheet:
- One thing that was clear
- One question you’d ask them
- Write it while they present; hand the post it to each team at the end
- Kind and specific: “I didn’t get what one row is” helps; “good job” doesn’t
- Not graded. It’s what you’ll practice formally in the Week 9 peer review
Laptops closed while people present, please. Your attention is part of the feedback.
The running order
Drawn now, in R, with today’s date as the seed:
teams <- c("Group 1: Trisha & David", "Group 2: Nickan", "Group 3: Victor",
"Group 4: Claire & Madeleine", "Group 5: Gwen & Shihang", "Group 6: Meher Naz",
"Group 7: Gursajan", "Group 8: Kimberly", "Group 9: Mia")
set.seed(20261020)
order <- sample(teams)
data.frame(slot = seq_along(order), team = order)Note
Re-run it and you get the same order: that’s what set.seed() is for. Same idea as your project: anyone re-running your code should get your numbers.
Running order
1 of 9 · Group 5: Gwen & Shihang
3 minutes · questions for each partner · Next up: Group 7: Gursajan, get ready
2 of 9 · Group 7: Gursajan
3 minutes · questions for each partner · Next up: Group 3: Victor, get ready
3 of 9 · Group 3: Victor
3 minutes · questions for each partner · Next up: Group 8: Kimberly, get ready
4 of 9 · Group 8: Kimberly
3 minutes · questions for each partner · Next up: Group 2: Nickan, get ready
5 of 9 · Group 2: Nickan
3 minutes · questions for each partner · Next up: Group 6: Meher Naz, get ready
6 of 9 · Group 6: Meher Naz
3 minutes · questions for each partner · Next up: Group 9: Mia, get ready
7 of 9 · Group 9: Mia
3 minutes · questions for each partner · Next up: Group 4: Claire & Madeleine, get ready
8 of 9 · Group 4: Claire & Madeleine
3 minutes · questions for each partner · Next up: Group 1: Trisha & David, get ready
9 of 9 · Group 1: Trisha & David
3 minutes · questions for each partner · Last group of the day
What happens next
After today
Feedback: written feedback on Canvas within a week, including one or two feedback questions for your team.
Revision: like every deliverable, you get one chance to improve. For the proposal, it’s a short written follow-up answering the feedback questions.
Next milestone: mid-quarter report, Tue, Nov 3. Data cleaned by a script, first figures, first model.
Last Thursday’s ggplot2 and this Thursday’s functions are the tools for that report. Start using them on your own data this week.
Before Thursday
STAT 266A · Fall 2026 · UC Santa Cruz