7. Teaching in low-resource classrooms¶
Most schools in Nepal do not have a well-equipped computer lab with reliable internet. This curriculum was designed for that reality. This chapter collects practical strategies.
No computers at all¶
- All of Grade 8 and the Track A versions of later units work on paper: tallies, tidy tables on the board, hand-drawn graphs, calculators.
- Printed extracts: print 10–30 rows of a dataset per group (the synthetic tables are small). Students compute summaries by hand, then you reveal the full-table result computed in advance (this book gives them).
- Unplugged versions of computational ideas:
- SQL: one student is the "database" holding district cards; others call out queries.
- Sorting and median: the "human median" line-up.
- k-NN: cards on a grid, a ruler for distance.
- Bootstrap: draw slips with replacement from a bag.
- Permutation test: shuffle outcome cards between two piles.
- Overfitting: "memorise last year's paper" role play.
- Paper spreadsheets: grid paper with column letters and row numbers teaches cell references and formulas before anyone touches a computer.
One computer and a projector¶
The teacher (or a rotating student) runs the notebook; everyone else predicts on paper first. This is surprisingly
effective with PRIMM. Use large fonts (Ctrl +).
A small lab (5–15 computers)¶
Groups of three with roles (driver, navigator, data steward); station rotation with paper activities.
No or unreliable internet¶
- Build the offline school bundle once (
python tools/build_offline_bundle.py) and copy it to every computer by USB. - Pre-download real datasets for the year's exercises in a single trip to an internet connection (respect licences: most listed in Appendix A allow classroom copies with credit).
- JupyterLite caches Python after the first load; open it once while online.
Power cuts¶
- Keep a paper version of every computer activity (each teacher guide has one).
- Save work often; teach Git (Grade 11) partly because it makes losing work less likely.
- Laptops on battery and phones can keep a small group going.
Large classes (50+)¶
- Groups of five with roles; one shared table per group.
- Mini-whiteboards (or slates, or A4 sheets in plastic sleeves) so every student answers Check questions at once.
- Gallery walks instead of presentations: posters on the wall, groups rotate and leave sticky-note feedback.
- Peer marking with the rubric; teacher samples and moderates.
Multi-grade classes and one-teacher schools¶
- Same theme, different depth (see chapter 3).
- Older students as data stewards and mentors.
- Shared class dataset for the whole school (e.g. travel to school), analysed at different levels.
Low cost materials list¶
| Item | Use |
|---|---|
| graph paper, rulers, coloured pencils | graphs, maps |
| index cards / cut paper slips | sorting, sampling, shuffling, SQL cards |
| a cloth bag or box | random draws |
| dice | random numbers, simulations |
| a stopwatch (phone) | measurement activities |
| tape measure, kitchen scale, thermometer, rain gauge (a straight-sided bottle) | collecting real measurements |
| sticky notes | feedback, gallery walks |
Equity and access¶
- Do not make phone or internet access a requirement for homework; always offer the offline fallback.
- Rotate access to computers fairly; track who has used them.
- Watch that boys don't monopolise the keyboard: assign driver roles deliberately.