Exploratory Data Analysis of a World Happiness Report using R and tidyverse. Dataset source: https://www.kaggle.com/datasets/ajaypalsinghlo/world-happiness-report-2024
Author: Joshua Kohlmeyer
Dataset: World Happiness Report 2024
Countries: 143
Features: 11
My aim with this project was to present a full exploratory data analysis (EDA) of a 2024 World Happiness Report. I investigated the main factors associated with national happiness and applied the full EDA process using R, as outlined (and well-taught) in my University of Pretoria WST221 module.
I did this to gain practical experience working with real-world data in R. The goal was to uncover which factors most strongly explain international differences in happiness.
The main variables used in the analysis were:
- ladder_score – Overall happiness score
- gdp – Log GDP per capita
- life_expectancy – Healthy life expectancy
- social_support – Perceived social support
- freedom – Freedom to make life choices
- generosity – Generosity score
- corruption – Perceived corruption
- dystopia_residual – Unexplained component of happiness
- Loaded and cleaned the dataset in R (using tidyverse)
- Renamed variables for improved readability
- Looked at distributions using histograms (awesome function; applied coursework theory here)
- Detected outliers using boxplots (underrated graph)
- Built a correlation heatmap (never disappoints when checking for correlation between variables)
- Interpreted the results in plain English (for clear communication of results)
All the analysis was done using tidyverse and documented using RMarkdown.
The full rendered analysis is available here:
https://joshuakohlmeyer.github.io/World-Happiness-Report-EDA/
- GDP per capita
- Healthy life expectancy
Both variables showed very strong positive correlations (approximately 0.75 to 0.85) with happiness.
- Generosity
Generosity showed near-zero correlation with happiness across countries, which was interesting and quite unexpected.
- Corruption
As expected, higher perceived corruption is associated with lower happiness (correlation approximately -0.4 to -0.5, clear negative relationship).
- Happiness and GDP had the widest spread and strongest outliers
- Corruption and generosity were much more tightly distributed
National happiness is driven primarily by economic strength and population health, while social generosity plays a minimal role in happiness across countries. Corruption consistently reduces well-being across nations.
- Economic success and people’s health dominate national well-being
- Not all “positive” traits (like generosity) translate directly into higher happiness
- Visualization and correlation analysis are essential in making sense of real data
- R
- tidyverse
- RMarkdown
- Git and GitHub
- World-Happiness-Report-eda.Rmd
- data.csv
- README.md
- World-Happiness-Report-EDA.Rproj
If you would like to connect or give feedback on this project (much appreciated):