akram-prog/Maji-Ndogo-Analysis

A data-driven investigation using MySQL, Jupyter to analyze water access, detect survey errors, and delivering actionable insights. # 💧 WEAVING THE DATA THREADS OF MAJI NDOGO'S NARRATIVE **Integrated Project: From Detecting Corruption to Designing Solutions** --- ## 📌 Project Overview The **Maji Ndogo Water Crisis Analysis** is a fictional but realistic case study exploring how **data analysis, SQL, and business intelligence** can improve access to water and strengthen accountability. The project unfolds in two key stages: ### 🔹 Part 1 – Detecting Discrepancies & Corruption - Compared surveyor reports with an audit dataset. - Identified **193 discrepancies** in quality scores. - Ruled out water source type as the cause of errors. - Exposed **four employees** (Zuriel Matembo, Malachi Mavuso, Bello Azibo, Lalitha Kaburi) as responsible for misconduct. ### 🔹 Part 2 – From Insights to Action - Combined multiple datasets to identify inequalities in water access. - Highlighted disparities between provinces, towns, and social groups. - Designed a `Project_progress` table for tracking repairs. - Proposed an actionable plan (new wells, filters, tap expansions, repair prioritization). - Developed **Power BI dashboards** to visualize the improvement plan, vendors, and regional disparities (`bi_dashboard/` folder). --- ## 🔑 Key Insights - **193 audit discrepancies** → Human error or misconduct suspected. - **4 employees** were responsible for most inaccurate survey entries. - **43%** of residents rely on shared taps, often overcrowded (~2,000 per tap). - **31%** of homes have taps, but **45% are broken**. - **18%** of residents use wells, only **28% safe**. - Wait times exceed **120 minutes** at peak times (Saturdays, mornings, evenings). - Rural areas like **Amanzi** and provinces like **Sokoto** are the most underserved. --- ## 🛠️ Plan of Action 1. Drill wells to replace unsafe river sources. 2. Install filters in wells (**RO for chemical**, **UV+RO for biological contaminants**). 3. Expand shared taps where queues exceed 30 minutes; deploy temporary tankers. 4. Repair failing infrastructure in * …