steph45acke-hue/kenya-census-data-pipeline

# 🇰🇪 Kenya Census Data Pipeline & Analytics This project provides a complete, end-to-end data science workflow designed to turn raw, messy government data tables into pristine, actionable insights for all 47 Kenyan counties. We built this to show how rigorous data engineering and statistical analysis can be applied to real-world demographic challenges. --- ## 📋 What The Project Does & Problems Involved * **The Problems Involved:** When you download raw demographic data straight from official government portals (like the 2019 Kenya Population and Housing Census), it is full of friction. The files contain messy metadata, unwanted header rows, hidden NaN values, unaligned columns, and formatting artifacts that instantly break standard data processing scripts. Furthermore, raw numbers alone don't explain underlying demographic realities—such as whether a county's physical size dictates its population or if major urban hubs are statistically distinct from rural regions. * **What The Project Does:** This project automates the entire Extract, Transform, Load (ETL) pipeline using Python. It cleans and structures the dataset across all 47 Kenyan counties, computes key population metrics, runs advanced econometric modeling, and builds an interactive web interface to visualize the findings. --- ## 🎯 Why We Built It & Solution Offered * **Why We Built It:** We wanted to bridge the gap between raw public records and actionable quantitative insights. Too often, data projects stop at basic cleaning; we wanted a complete, end-to-end portfolio piece that proves how rigorous data science and statistical testing apply to real-world demographics. * **Solution Offered:** This repository offers a clean, reproducible, and fully documented workflow. It turns messy government spreadsheets into pristine structured data (`kenya_counties_official.csv`), automates statistical hypothesis testing, and provides an interactive Streamlit dashboard so anyone can explore Kenya's demographic d …