🇿🇦 Youth Unemployment Analysis in South Africa (Ages 15–24)

This project explores and analyzes youth unemployment trends in South Africa using a dataset derived from ILOSTAT's SDG indicator 8.5.2 — Unemployment Rate (%) by Age, Sex, and Country.

Tools Used

• Python (Google Colab) for data cleaning, encoding, and exploratory data analysis (EDA)
• Pandas, Matplotlib, Seaborn for visual analytics
• Power BI for interactive storytelling and dashboards
• GitHub for project version control and documentation

Goal

To understand the patterns, gender disparities, and time-based changes in unemployment among South African youth (ages 15–24), and present the insights using visual and statistical techniques.

Summary of Steps

1. Data Cleaning

   • Filtered only South African data
   • Focused on youth age group (15–24)
   • Removed unnecessary columns with excessive null values
   • Encoded categorical features for further analysis

2. Data Exploration

   • Correlation analysis
   • Time-series analysis by gender
   • Detected notable spikes (for example post-2020 due to COVID-19)

3. Visualization

   • Line plots show trends over time for Male, Female, and Total
   • Highest unemployment consistently observed among young females

4. Power BI Storytelling (Coming Next)

   • Visual dashboards with filters (gender, year)
   • Key insights for policymaking and socio-economic review

Key Insights

• Youth unemployment has fluctuated between 42% to 70% over the past two decades.
• Females consistently face higher unemployment rates than males.
• Sharp increase post-2020 indicates COVID-19’s economic impact on youth employment.

Future Plans

• Add forecasting models (e.g., ARIMA, Facebook Prophet)
• Compare trends across other Southern African countries
• Integrate education/training level to explore deeper insights

---