Project Overview
This comprehensive data analysis project investigated the factors contributing to physical exhaustion in the workplace using the European Work Life Dataset. The study applied multiple statistical and machine learning models to identify key predictors and moderating effects of managerial support on employee burnout.
Data Visualizations
Feature Correlations
Heatmap showing correlations between workplace factors and exhaustion levels.
Regression Analysis
Coefficients and significance levels for exhaustion predictors.
Key Findings
- Regression Analysis: Identified significant predictors of workplace exhaustion
- LASSO Model: Feature selection for optimal predictor identification
- Random Forest: Non-linear relationship discovery with 87% accuracy
- Quantile Regression: Analysis across different exhaustion levels
- Moderating Effects: Impact of managerial support and supervisor gender
- Actionable Insights: Data-driven recommendations for workplace wellness
Methods & Tools
Python
Regression
LASSO
Random Forest
Quantile Regression
Data Visualization
Impact
- Delivered actionable insights for workplace wellness program development
- Identified key intervention points for reducing employee burnout
- Provided evidence-based recommendations for management training
- Contributed to academic understanding of workplace exhaustion factors