Employee Attrition Analysis – ABC Manufacturing Ltd

Data Science Internship Programme – AnalystLab Africa Week 1: Business Understanding & Data Exploration

Overview

This project simulates a consulting engagement for ABC Manufacturing Ltd, which wants to understand the drivers of employee attrition before investing in predictive machine learning models. Using the IBM HR Analytics dataset (1,470 employees, 35 variables) as a stand-in for the company's real HR data, this Week 1 deliverable focuses on business understanding, data inspection, and exploratory data analysis.

Dataset

IBM HR Analytics – Employee Attrition & Performance Kaggle link 1,470 employees · 35 variables · no missing values · no duplicate rows

Business Questions Addressed

1. What does the company's workforce look like?
2. Which departments have the highest employee attrition?
3. Does age influence attrition?
4. Does monthly income affect retention?
5. Does overtime influence attrition?
6. Which job roles experience the highest turnover?
7. Which variables appear important for future predictive modelling?

Key Findings

• Employees who work overtime leave at a substantially higher rate than those who don't.
• Monthly income is noticeably lower among employees who left.
• Younger employees show higher attrition than older, more tenured staff.
• Sales Representatives and Laboratory Technicians show the highest turnover by job role.
• Overall attrition rate: 16.1% (237 of 1,470 employees).

Project Structure

├── data/
│   ├── raw/              # Source dataset (untouched)
│   └── processed/        # Cleaned data (if applicable)
├── notebooks/
│   └── 01_week1_exploration.ipynb
├── reports/
│   ├── business_understanding_report.docx
│   ├── dataset_inspection_report.docx
│   └── reflection_report.docx
├── visuals/              # Exported charts (PNG)
├── docs/                 # Official assignment brief
├── requirements.txt
└── README.md

Tech Stack

• Python 3.13
• pandas, numpy
• matplotlib, seaborn
• Jupyter Notebook

Installation

python -m venv venv
venv\Scripts\Activate.ps1      # Windows
pip install -r requirements.txt

Then open notebooks/01_week1_exploration.ipynb in VS Code or Jupyter and select the venv kernel.

Deliverables

• Business Understanding Report
• Dataset Inspection Report
• Jupyter Notebook (EDA + visualisations)
• Reflection Report
• GitHub Repository (this repo)
• LinkedIn Post
• X Post
• Google Drive Link Submitted

Author

Caleb Djarabé — Data Science Intern, AnalystLab Africa