Part 3 — Grade 10: Data analysts¶
Python for analysis, a first honest model, AI literacy and dashboards.
Grade 10 moves from wrangling to analysis: cleaning a real-world daily series, describing distributions and outliers, fitting lines and resisting causal claims, separating trend from season, and building a first model — k-nearest neighbours — that you test honestly against a baseline. Then you step back to ask how AI systems learn and fail, and you publish your work as a small dashboard.
| Unit | Title | Lessons | Project |
|---|---|---|---|
| 1 | pandas deep-dive: cleaning a real-world series | 4 | Clean series report |
| 2 | Distributions and outliers | 3 | What does a typical day look like? |
| 3 | Relationships and the line of best fit | 3 | Two variables, one honest line |
| 4 | Time series basics | 3 | Climate or weather? |
| 5 | What is a model? | 3 | Model vs baseline |
| 6 | AI literacy and ethics | 3 | AI charter and audit |
| 7 | Dashboards with Streamlit | 3 | Mini dashboard |
| 8 | Project: Air in my city | 2 | flagship project |
Tracks: Unit 6 is no-code (Track A) and suits any class; the others use Python (Track B).
Pacing: 24 lessons + projects ≈ 40 periods.
Key idea of the year: always compare with something — a baseline, a season, another group, chance — before claiming a result.