• Carolina Duque Chopitea

Understanding World Happiness & Identifying its Drivers | EDA - Python, Tableau


Python code here


The goal of this exploratory data analysis (EDA) was to understand the state of world happiness and identify factors policy makers should focus on if they want to create policies that increase happiness in their countries and regions.


The analysis was conducted using the World Happiness Reports from 2015, 2016, 2017. The Report is an annual publication by the Sustainable Development Network Solutions that ranks national happiness based on individuals ratings of their own life (from 1 to 10), which the Report then correlates with other factors such as economy and GDP, health, family, government trust, freedom, and generosity. The factors are then benchmark against Dystopia, a hypothetical country that has the least happy people.


From the analysis it was found that between 2015 to 2017 world average happiness score has stayed the same at 5.4 points. This means that there has been no improvement or deterioration of world happiness.




Individual country scores have also stay constants with only a handful of countries exhibit scores changes above 20%. For example Venezuela's score decrease and Togo's' score increase by 23% respectively. Other countries like Central African Republic, Lesotho, Liberia and Haiti have shown a happiness decreased of over 20%. These countries have in recent years or are experiencing internal conflicts, which may explain the large fluctuation is happiness scores over time.


Moreover, we can observe that the happiest countries (dark blue) are concentrated in developed countries with Switzerland taking the lead. In fact, the top 10 happiest countries are all located in the developed world, with only a few nations like Costa Rica, Brazil, Israel, and Puerto Rico making it to the top 20 of the list. Also note that the least happiest countries (brown) are concentrated in the developing world, particularly in the African region and parts of Asia.


To better understand the distribution and state of work happiness, scores where looked at in terms of their regional distribution.


In the box-plot above we can clearly see that there is a happiness gap between and within regions. Again, regions with the most developed nations score the highest (AUS/NZ, North America, Western Europe). This may suggest that there is a link between development and happiness, however this is beyond the scope of this analysis.


From the box-plot it can be observed that there are 3 regions (Southeaster Asia, Southern Asia, Sub-Saharan Africa) that are below work happiness average. In fact there is a significant difference between the happiest and least happy region of 3 points.


Furthermore, the top 5 happiest countries are all in Northern Europe, which belong to the Western European block. Yet the region's score in only in third place as it is brought down by countries like Greece, Portugal, Italy and Cyprus whose sores are significantly lower than their Northern European counterparts. For example the difference between the region's happiest and least happiest, Switzerland and Greece respectively, country is quite significant: 2.5 points.


Nevertheless, scores in the Middle East and North African region are even more spread out with really high performing countries like Israel (11th place), Oman (25th place), and worst performing countries like Syria and Yemen that have the lowest happiness scores.


Based on the data analysis there are two main findings: happiness is concentrated in developed countries, and not only there is a staggering inequality between countries but also withing some regions.


So, what factors can world leaders, particularly those in developing areas, focus on if they want to move away from a period of happiness stagnation, increase happiness in their countries, and bridge happiness inequality?


From the 7 factors said to drive happiness leaders should place particular attention to GDP, health, and family as these have the strongest correlation to happiness (0.81, 0.78 and 0.75 correlation respectively).


Scatter Plots of happiness and happiness drivers


As you can see from the charts above in which happiness is plotted against the factors that are believe to drive happiness, a clear linear relationship is exhibit between happiness and GDP, family, health and to a lesser degree Dystopia residuals (the unexplained components of happiness scores). As such, policymakers should focus on policies that strengthen these factors, for example by fostering policies that:


  • Foster economic growth and innovation to increase production of good and services

  • Strengthen the foundations of family through improvement of social support like parental leave.

  • Encourage preventive healthcare and facilitate access to medicines to increase life expectancy.

  • They should strive to be far away from Dystopia - far from poor health, social support, no freedoms, no generosity, and far from corruption and poverty.

Carolina Duque Chopitea