Which countries are the happiest? What factors predict happiness?
1. Intro
I’m graduating from college in a year. My life will be uprooted — I will no longer have classes, homework, clubs, and a schedule. As conventional wisdom goes, your twenties are your time to explore and live life to the fullest. One question remains on the top of my mind: where should I live?
I want to be happy. So today, I’m going to take a look at the “happiest” countries in the world and investigate what factors influence happiness.
Why this matters
The COVID-19 pandemic and the economic recession have negatively affected many people’s mental health. For instance, 4 in 10 adults in the U.S. have reported symptoms of anxiety or depressive order, up from 1 in 10 in January 2019. Even before the pandemic, negative feelings — worry, sadness, and anger — have risen 27% from 2010 to 2018 in the World Happiness Report.
We are in an era of rising tensions and negative emotions — Economist Jeffrey D. Sachs
Civil unrest is unfolding across the world, especially in the U.S and many European nations. Issues like police brutality, unfair elections, racial unrest plague the television screens. It has gotten so bad that John Krasinki even made a series called Some Good News to combat the negativity.
It has become increasingly important to stay optimistic during these difficult times. The pandemic and the global unrest have tainted many’s views of the world. Combined with the fact that over 50% of people will be diagnosed with a mental illness or disorder at some point in their lifetime, there is a lot of unhappiness in the world.
The goal of this data project is to shift the focus away from bleakness and back to happiness. I want those who are struggling to see the good in the world to realize there is a world of opportunity out there for them to be happy. On the other hand, I also want to help social organizations and government agencies understand the underpinnings of this happiness. Anyone — from a relocating worker to an NGO executive, to a casual reader — can benefit from these insights about happiness on earth.
2. Data Sources
World Happiness Report
The World Happiness Report is a landmark publication by the United Nations Sustainable Development Solutions Network. This report is based on respondent ratings of their own lives, which is correlated with various quality of life factors. The World Happiness Index is continuing to gain recognition as governments, organizations, and civil society increasingly use these happiness indicators to inform their policy-making decisions.
World Health Organization Suicide Statistics
Close to around 800,000 people commit suicide every year, which is one person every 40 seconds. Suicide is a global phenomenon and reflects the epitome of unhappiness. This data analysis will utilize World Health Organization (WHO)suicide data from its Mortality Database. The aggregate numbers cover 1970–2016, by country, year, age, groups, and sex. This data will shed light on some of the unhappiest countries in the world.
United Nations Data — Country Indicators
UNData is a dataset service created by the United Nations Statistics Division offered to the global user community. The dataset contains key statistical indicators of countries, including general information, economic indicators, social indicators, and environment & infrastructure indicators. This data can add an additional layer of nuance to the data above and help provide basic information regarding each country.
Country Continent Data
To add another level of granularity in this project, I also integrated a continent dataset from Kaggle to analyze Happiness and country-specific factors on a continent level.
3. Data Wrangling
All of the datasets were manually cleaned in Microsoft Excel prior to import into R Studio. The process involved manually renaming the country names and eliminating outlier data — especially in the WHO Suicide Database.
The WHO suicide data needed to be cleaned extensively. I filtered through the years, removed the bracketed estimated range in each cell using stringr, then dcasted the data frame to fit horizontally. The other datasets were also cleaned and wrangled.
I merged all four data sets to create the final data frame. In addition, I also merged all the World Happiness Reports from 2015 to 2019 for additional analysis.
4. Analysis
Part 1. Happiest Countries
In the 2019 report, four Scandinavian countries finished as the top 4, with Finland being ranked the happiest country.
But how consistent are these rankings? Which countries are regular contenders? I combined the World Happiness Report from 2015 to 2019 to show a boxplot of the distribution of happiness scores in the past 5 years.
Finland has the most variability in happiness in the past 5 years, while Iceland has the least variability. There seem to be different “leagues” of happiness:
- Top Contenders: Switzerland + Scandinavians (except Sweden)
- Strong Finishers: Sweden, New Zealand, Netherland, and Canada.
- Cut-Off: Austria
Don’t be surprised if you see another one of the Scandinavian countries or Switzerland finish top next year.
Now, social media and press generate a lot of buzz for the Top 10 countries every year. But what about the rest?
Here is a wordcloud of countries with frequent appearances in the 10–60 spots on the rankings in the past 5 years. The size of each country factors in the number of appearances and recency — recent rankings matter more.
Past the top 10, we begin to see a mix of developed economies around the world. We also see a heavy Latin America presence, which is surprising because Latin America is comparably less developed than the rest of the pack — and development often correlates with happiness.
Countries with the Biggest Happiness Changes
African countries are very heavily represented on both lists, which reflects their underlying geopolitical instability. Most on the lists are also smaller, lesser-known countries. It may be easier for smaller countries to change than it is for larger countries.
It’s not surprising to see Venezuela at the bottom of the list given its recent crisis. This list also heavily features African countries. If you’re looking to be happy, African countries are a mixed bag. But how do continents compare on a macro level?
Happiest Continent
Oceania and Europe have the highest median happiness, followed by the Americas and Asia. Africa trails the rest of the pack. The variability seems to correspond with the number of countries on the continent.
One interesting observation is that there are many outliers for the Americas. Observing the world happiness map below, the high outliers are Canada, U.S, and various Latin American countries like Brazil. The low outliers are likely the chain of small islands around Cuba/Bahamas.
Part 2. Understanding Happiness
Now that we know the happiest countries, why are they happiest? What factors correlate the most with happiness?
The World Happiness Report uses 6 factors in its evaluation of various countries: GDP per capita, healthy life expectancy, generosity, social support, freedom to make life choices, and perception of correction. These sub-factors help explain the happiness score of a country.
The top 10 countries have very similar breakdowns. One observation is Iceland’s low score on perception of corruption, as it ranks 11th highest on the Corruption Index.
But how much do these individual factors correlate with or predict happiness? We examine the relationship for the 2019 Happiness Report. Note: factor values are calculated by the World Happiness Report — more on that here.
We see a strong relationship between happiness and social support. This makes sense because social support alleviates psychological problems that make people unhappy. Countries in Europe, Oceania rank highest for social support while Africa has low social support.
We see a weaker relationship between happiness and freedom. Again, many European countries are happier relative to their lack of freedom while African countries are unhappy given their freedom. It seems that freedom would be a good supporting factor to include in a logistic regression.
Generosity seems to be a poor predictor of happiness given the weak correlation. Perhaps this factor could be replaced by a Gender Equality factor — which is discussed later.
Countries with the lowest perception of corruption seem to be the happiest. However, corruption doesn’t seem to predict happiness among the countries with high perceptions of corruption.
Continent seems to be a strong predictor of life expectancy, and life expectancy seems to be a strong predictor of happiness. This isn’t surprising as happier people live longer.
For the last factor, I wanted to see the relationship between the real GDP per Capita figures instead of the Report’s GDP factor scores. I used the UN data to plot the following graph.
I roughly grouped the countries above into the following quadrants.
- Third World — Mainly African countries. GDP per capita seems to have a linear relationship with happiness. No country with a happiness score of <5 has a GDP per capita of over $11,500.
- Developed Asia — Have you seen the movie Crazy Rich Asians? These are the Singapores, Hong Kongs, Japans of the world. High GDP but relatively low Happiness-to-GDP-per-capita ratio.
- 6.5+ Club — Almost entirely European with a few Oceanic and North American countries (Canada, U.S).
- Latin America — I saved the best for last. In line with our observations earlier, Latin American countries are extremely happy relative to their economic status. Almost all Latam countries have a Happiness Score over 5 despite none having a GDP per capita over $16,000.
I made the data interactive for you to explore. Note: hover over the points to see the country names.
Performance Analysis
Using GDP per capita as a proxy for the economic resources available to a country’s government, we can view happiness as a return on investment (ROI). In a sense, we can view governments as investors and treat countries as funds, with happiness being the desired return.
Using the smoothed line from the chart above, we can see whether each country outperforms or underperforms. Underperforming countries lie above the smoothed line, meaning they are unhappier than average despite their resources. Outperforming countries can be found below the smoothed line because they are happier than average despite their lack of resources.
Other Happiness Factors
I examined over 50 social, environmental, economic, and infrastructure variables in each country’s U.N data and plotted over 20 relationships in R. Here are the two most interesting results:
People living in rural communities surrounded by nature must be happier, right? Turns out that urban population as a % of total population has a positive correlation with happiness score.
This may be a manifestation of the GDP per capita vs happiness relationship, but it warrants further investigation.
I also investigated the impact of gender equality, using the proxy variable of % of seats held by women in national Parliament. There seems to be a weak relationship between the two variables. Still — it appears to be a stronger predictor than the generosity variable in the World Happiness Report, making it a potential replacement.
Part 3. Understanding Suicides
One key area of investigation is whether suicides are related to happiness. After all, suicides represent the epitome of unhappiness and are completely unaccounted for in the World Happiness Report.
So, how do countries differ in suicide rates? Do suicide rates correlate with happiness? How can we explain high suicide rates in developed economies? Here is the world suicide map to start us off.
One interesting observation is that suicide rates seem to be lower in less developed countries. It also does not seem to match the World Happiness Map. Let’s investigate the relationship between suicides and GDP per capita.
Overall, suicide rates seem to positively correlate with GDP per Capita, albeit the relationship is very weak.
Interestingly, we see that most countries with GDP per capita over $25,000 seem to have high suicide rates except Asian countries. Specifically, Arab and Mediterranean countries seem to have far fewer suicides relative to their GDP per capita than other countries. This may be due to religion or culture:
- Western individualism — you consider yourself to be independent and self-contained
- Asian collectivism — you are entwined and interconnected with the other people around you, valuing the group over the individual
These cultural values may have something to do with preventing suicides in Arab and Mediterranean cultures. But why do developed economies have more suicides?
Maslow’s Hierarchy of Needs might provide some answers. According to the hierarchy, developed countries are preoccupied with physiological and safety needs. Before they satisfy these needs, they cannot move on to the next few psychological stages.
Developed nations tend to provide most residents with safety and physiological needs. Thus, they move on to love and belonging, esteem, and self-actualization. These “psychological” stages are related to the most common reasons for suicides: depression, hopelessness, isolation, stress, and burden. This theory may intuitively explain suicides, but further scientific research into this area is warranted.
Against expectations, suicide rate seems to have a weak positive relationship with happiness at the country level. We see pretty vague continent-groupings patterns on the chart.
- Europe and Oceania: High suicide rate, high happiness
- Africa: Low suicide rate, low happiness
- Americas: Low suicide rates, relatively high happiness
- Asia: All over the place
So the happiest country isn’t the one where there are the least suicides. If a country is extremely happy, it will most likely have a high suicide rate.
5. Conclusion
Happiest Countries: Scandinavian countries consistently reign supreme, followed by Oceania and my home country Canada.
Continent Analysis Takeaways
- African countries are volatile in happiness rankings— both good and bad volatility. Africa has the lowest median happiness ranking.
- Asian countries vary the most in happiness rankings, likely due to the size of the continent.
- Latin American continues to be a dark horse in this analysis. They are extremely happy relative to their economic status and have a strong presence in the happiness ranking leaderboard between the spots of 10 to 60 — where other countries tend to have far greater GDP per capita.
If you want to be happy and live cheap, Latin America should be your next destination.
Happiness Factor Takeaways
- Life Expectancy and Social Support correlate the most with happiness. Both factors are direct indicators of quality of life.
- Generosity and Perception of Corruption correlate the least with happiness. Generosity could be replaced by Gender Equality, which we can measure using the proxy variable % of seats held by women in national parliaments.
- Freedom to Make Decisions and Urban Population both have small correlations with happiness. The relationship observed for urban population could be related to the relationship between happiness and GDP per capita.
- Visualizing GDP Per Capita vs Happiness allows us to group different continents and see if countries are happier relative to their wealth and resources. A country can be considered to “outperform” if they are below the smoothed line and “underperform” if they are above.
Suicide Rate Takeaways
- Suicide rates have a weak positive correlation with national happiness.
- Maslow’s Hierarchy of needs could explain why developing countries tend to have fewer suicides than highly-developed countries.
- Continent appears to be a better predictor of suicides than happiness — likely due to different values of different cultures, such as individualism vs collectivism.
Further Research
- Why are Latin American countries much happier compared to other countries with similar economic status? Does it have to do with culture?
- Use logistic regression and classifier trees to understand what economic, social, political, environmental, or infrastructure factors are the best predictors of the happiness score.
- Look into the largest changes in the happiness scores in the past 10 years. Investigate the changes within the country that led to the increase in happiness score.
- Add cost of living as a variable to determine countries with the highest happiness to cost of living ratios, and vice versa.
- Further investigation into the positive relationship between suicide rates and national happiness.
Data Used for this Analysis
- World Happiness Report Data — 2015–2019
- WHO Global Health Data Repository — Suicide Rate Estimates 2000–2019
- United Nations Statistics Division — Country Indicators
- Continent-Country Mapping Data
Software Used for this Analysis
- R Notebooks — data analysis and visualization
- Excel — manual data cleaning
- Plotly API — interactive charts
About Davin
Davin Liu is a junior at the Wharton School of the University of Pennsylvania studying Finance and Business Analytics. In his spare time, Davin enjoys embarking on cross-country road trips, shredding the slopes of Whistler, and immersing himself within psychological thrillers films
This data project was conducted for Prasanna Tambe’s course: Analytics & The Digital Economy.
