Skip to content

Preface

Nepal produces more data every year than ever before: a census every ten years, daily rainfall from hundreds of stations, prices at Kalimati market, air-quality sensors in the Kathmandu valley, school records, health-post registers, mobile money, maps drawn by volunteers after the 2015 earthquake. Decisions about where to build a school, which ward needs a health worker, when to warn farmers about floods and how to spend a municipality's budget are increasingly made — or should be made — with data.

Yet most students leave school knowing how to calculate a mean and draw a pie chart, but not how to ask a good question, collect data fairly, clean a messy table, tell a real pattern from chance, or notice who is missing from a dataset. This book is an attempt to close that gap.

What this book is

Data Science from Nepal is a five-year course for Grades 8 to 12. It was built for Nepal's schools, is open to learners everywhere, and is written so that

  • a student can follow it alone, chapter by chapter, with worked examples, exercises and answers to selected problems;
  • a teacher can teach from it, with timed lessons, activities that work with or without computers, common misconceptions, full answer keys (in the teacher edition) and a handbook on teaching data-science foundations;
  • a curriculum planner can see how ideas build from Grade 8 to Grade 12 and how they connect to existing Maths, Science, Social Studies and Computer Science subjects.

It is built around Nepali contexts and data: districts and provinces, the Terai–Hill–Mountain belts, monsoon rainfall, Kathmandu's winter air, migration and remittances, maternal health, Bikram Sambat dates. Where real data can be redistributed freely, exercises use it; where it cannot, the book ships small synthetic practice tables that look like the real thing and are clearly labelled, and points you to the real source.

How it is organised

  • Getting started — how to use the book, how to set up (from paper to Python), and the data-science cycle that runs through every chapter.
  • Grade 8: Data detectives — questions, collecting, graphs, summaries, comparisons, data stories and ethics, without any programming.
  • Grade 9: Data wranglers — tidy spreadsheets, cleaning, SQL, first Python, pandas, charts and rates.
  • Grade 10: Data analysts — cleaning real series, distributions, relationships, time series, a first model, AI literacy and dashboards.
  • Grade 11: Data scientists I — performance, advanced SQL, uncertainty and the bootstrap, testing, regression, seasonality, maps and reproducibility.
  • Grade 12: Data scientists II — the machine-learning workflow, evaluation, overfitting, fairness audits, feature engineering, communication and a capstone with a real client.
  • Teaching data-science foundations — a handbook for teachers: the progression of ideas, planning a year, teaching statistics and code, low-resource classrooms, projects, assessment, AI and inclusion.
  • After Grade 12 — how to keep learning: study methods, a roadmap, books, courses, building a portfolio, communities, and pathways into higher education and work.
  • Appendices — free datasets, an English–Nepali glossary, cheat sheets for spreadsheets, SQL and Python, and key formulas.

What it asks of you

Curiosity, honesty and patience. Data science is not mainly about clever code. It is about asking a clear question, looking carefully, being honest about what the data can and cannot show, and remembering that most data are about people.

Licence and credit

The text is licensed CC BY 4.0: you may copy, print, translate and adapt it, including for textbooks, as long as you credit the source. Code is MIT-licensed. Datasets keep their own licences, listed with each one. See the credit line at the front of the book.

How this book was made

This book began as a curriculum for Nepal's schools and is open to learners everywhere. It was structured and drafted with the assistance of AI, then curated, verified and adapted by the author: every worked example is run automatically, and sources are credited in the project's third-party log. If you find a passage that should credit someone, please report it through the repository so it can be fixed.

Thanks

To the teachers and students who will test these lessons and tell us what to fix; to the organisations that publish open data about Nepal; and to the many people worldwide who share teaching materials, software and datasets openly. Corrections and suggestions are welcome through the project's repository.

Ram Chandra Giri and contributors