Final Project: What Affects Happiness Levels Around the World?

For my final project, I explore the relationship between 6 different factors and happiness scores across different countries using the World Happiness Report 2023 dataset. The primary goal is to visually analyze and compare how various factors as outlined by the happiness report impact the happiness scores of different countries. I aim to make it easier to compare how each of the factors impact happiness differently, and uncover which factors affect happiness more than others. 

1. Visualizing how Happiness Score is impacted by 6 different factors

The following interactive map shows Happiness Score across countries worldwide, as well as their GDP per capita, Social support, Healthy life expectancy, Freedom to make life choices, Generosity and Perceptions of corruption. Users can click through each of the maps to see the how the trend of the factors vary. Users may also use the search tool to search individual countries and their scores & values. 

2. Scatterplots of Each of the Factors

While the above map gives a nice visual understanding and intuition of how happiness levels and the 6 factors interact, it does not concretely show nor compare them. The below scatterplots compare the factors more concretely, each of the dots being a country, with the x-axis being each of the 6 factors and the y-axis being their happiness score. Users can also see a regression line overlayed on each of the plots, showing the correlation between happiness and the factors more easily. 

3. Stacked Regression Lines

To compare each factor more easily, I normalized regression lines (since each of the variables had different dimensions) and put them in one plot. The slopes are noted in the legend. Social support had the highest slope of 4.74, closely followed by GDP per capita at 4.54. This indicates that across all countries, social support and financial conditions are most closely correlated with happiness levels. Notably, perceptions of corruption were negatively correlated with happiness levels, meaning that the lower the perceived corruption, the higher the happiness levels. However, the absolute value of the slope tends to be on the lower side at 2.37, indicating that corruption and happiness levels may not be very strongly correlated. Interestingly, Generosity had a slope of 0.25, the lowest of all 6 factors. This raises the question of why generosity seemingly has little to do with people's happiness levels, and if the way Generosity was measured affected these results. 

4. Conclusion

From our analysis, we see that Social Support shows the highest correlation to happiness, giving us a good outlook into what are the things we as individuals as well as governments should focus on in order to improve happiness and sense of well-being across citizens. At the individual level, we can try to make sure that we are well connected with our social networks and support systems, and at national levels, governments should work to create equitable and well-developed economies, develop healthcare, and ensure individual freedoms. 

It is important to note that we only identified correlational factors and not causation. This means that we cannot conclusively say that a certain factor directly causes happiness scores to go up or down. Rather, we need to have a holistic view of what contributes to happiness across the world. 

Several studies around the world look into happiness, trying to find which factors contribute to happiness the most. As societies continue to evolve, we must look at different factors including perceived gender inequality, physical health, digital media use & culture, rural vs. urban environments, etc. I would love to further investigate these factors in order to achieve a more holistic understanding of happiness across the world.