🌍 World Happiness Report - Data Insights Dashboard

(World Happiness – Data Analysis & Visualization Project)

Project Overview (Resumen del Proyecto)

This project analyzes global happiness indicators using the World Happiness Report and complementary Gallup survey data to uncover social, economic, and behavioral factors influencing well-being across countries.

The workflow covers the full data lifecycle:

• Data exploration and transformation using Python and Jupyter Notebooks
• Integration of multiple datasets beyond the original report
• Analytical modeling and visualization using Power BI dashboards

Special focus was placed on methodological rigor, preserving original data structures, and contextualizing subjective metrics such as life satisfaction, generosity, and social support.

This project simulates a real-world business intelligence workflow combining ETL, analytics, and visualization.

---

Objectives (Objetivos)

• Explore and understand global happiness metrics
• Clean and transform multi-source datasets using Python
• Apply EDA techniques to uncover trends and patterns
• Build interactive dashboards for data-driven storytelling
• Communicate insights clearly and visually

---

Technologies Used (Tecnologías Utilizadas)

• Python (Pandas, NumPy)
• Jupyter Notebooks
• Power BI
• CSV & external datasets
• Git & GitHub

---

Project Structure (Estructura del Proyecto)

├── docs/ # Technical documentation
├── files/ # Datasets (raw and processed)
├── scr/ # Python helper scripts
├── EDA.ipynb # Exploratory data analysis
├── transformación-*.ipynb # Data cleaning and ETL
├── whr-power-bi-dashboard.pbix
├── whr-power-bi-dashboard.pdf
├── README.md
└── .gitignore

---

Workflow (Flujo de Trabajo)

1️⃣ Data Exploration (EDA)

• Initial inspection of World Happiness data
• Distribution analysis and missing value assessment
• Preliminary insights

Notebook: EDA.ipynb

---

2️⃣ Data Transformation (ETL)

• Cleaning and normalization
• Handling missing values (with and without imputation)
• Merging complementary datasets

Notebooks:

• transformación-gallup-info-2005-2025.ipynb
• transformacion-v2-sin-imputar.ipynb

---

3️⃣ Dashboard Development

• Interactive visualizations in Power BI
• Country comparisons
• Trend analysis over time
• Key happiness indicators

Files:

• whr-power-bi-dashboard.pbix
• whr-power-bi-dashboard.pdf

---

Methodological Considerations (Consideraciones Metodológicas)

• Original missing values were intentionally preserved (no imputation) to maintain data integrity and reflect the true structure of the World Happiness Report.

• Raw numerical indicators were transformed into more interpretable formats (percentages and categorical variables) to improve visualization clarity.

• External documentation and Gallup survey sources were incorporated to better understand how happiness metrics were collected and defined.

• Special attention was given to key subjective variables such as:

  • Life Ladder (self-reported life satisfaction)
  • Generosity
  • Social support

• Survey methodology differences across countries were analyzed to contextualize potential bias in results.

• Project coordination followed an agile sprint-based workflow, with Claudia Cervantes serving as Scrum Master.

---

Key Insights (Hallazgos Clave)

• Happiness outcomes are influenced not only by economic factors but strongly by social support, perceived freedom, generosity, and life satisfaction (Life Ladder score).

• Quantitative happiness indicators attempt to measure highly subjective concepts, requiring careful interpretation beyond raw numerical values.

• Converting complex numerical indicators into percentages and binary categories (yes/no) improved clarity and interpretability in visual analysis.

• Data completeness varies significantly by country, reinforcing the importance of working with original (non-imputed) values to reflect real-world reporting limitations.

• Additional sources (Gallup surveys and methodological reports) revealed differences in polling methods and sample sizes across countries, introducing potential measurement bias.

• Some countries surveyed over 4,000 individuals while others included fewer than 500, and collection methods (in-person vs. phone surveys) varied by region.

• These methodological differences highlight challenges in cross-country happiness comparisons and the need for contextual analysis.

---

Team & Credits (Equipo y Créditos)

This project was originally developed as a collaborative course project by:

• Claudia Cervantes — https://github.com/cloud9international
• Mayka Durán — https://github.com/Maykaduran
• Patricia Merchán — https://github.com/patrimerchan
• Andrea R. Virués — https://github.com/andreavirejos
• Ona Zaragoza — https://github.com/onaimusest

Original team repository:
https://github.com/Maykaduran/BI-WHR-beyond-the-data.git

This repository is a curated portfolio version maintained by Claudia Cervantes.

---

Final Notes (Notas Finales)

• The project follows a real-world analytics pipeline (EDA → ETL → Visualization)
• Multiple datasets were integrated for richer insights
• The Power BI dashboard provides an interactive exploration of results
• This work simulates a business intelligence workflow for global indicators analysis