Logo Lanfrica

kmiringu/njenga-data

Domaine:

socioeconomic

Type de record:

dataset
Créateur:
kmi
Hôte:
Kenya Construction Cost Transparency Dashboard — Moringa Hackathon 2026 # NjengaData Helping ordinary Kenyans understand the real cost of building their own home. NjengaData pulls together four public data sources into a single pipeline that answers the questions an incremental builder like James — earning KSh 40,000 a month — actually asks: what will it cost, where is it cheaper, are materials getting more expensive, and can I afford it? --- ## Data sources | Source | Description | Years | |---|---|---| | KNBS CIPI | Construction Input Price Index — quarterly material prices | 2019–2023 | | CAHF | Centre for Affordable Housing Finance Africa — county housing costs | 2022 | | KCHS | Kenya Continuous Household Survey — county median household income | 2022 | | Integrum | Regional construction cost per m2 | 2021–2024 | All sources are publicly available. Raw files are in `data/raw/`. --- ## Running the pipeline Run the three notebooks in order: | Notebook | Owner | What it does | |---|---|---| | `01_data_cleaning.ipynb` Loads, cleans and writes all four sources to `njenga.db` | | `02_analysis.ipynb` Runs SQL queries against `njenga.db`, produces four findings | | `03_dashboard.ipynb` Reads findings and generates four charts to `data/processed/charts/` | Open each notebook and run **Kernel > Restart and Run All**. Start with 01 — it builds the database that 02 and 03 depend on. --- ## The four findings - **F1** — A 2BR low-rise in Nairobi costs KSh 2.2M–2.5M. Labour is the largest cost at 38–42%. - **F2** — Steel prices rose 19.6% between 2019 and 2023. Cement rose 11.2%. - **F3** — Coastal counties build 18–22% cheaper than Nairobi. - **F4** — A household on median Nairobi income needs 25.4 years of savings to afford a 2BR build. --- njenga-data/ ├── data/ │ ├── raw/ # original source files, untouched │ ├── processed/ # cleaned outputs and charts │ └── njenga.db # SQLite database built by notebook 01 ├── notebooks/ │ ├── 01_data_cleaning.ipynb │ ├── 02_analysis.ipynb │ └── 03_dashboard.ipynb ├── a …