Quantifying Happiness: Exploring Factors Affecting Happiness

DataRes at UCLA
14 min readDec 21, 2023

Authors: Anya Smolentseva (Project Lead), Jasmine Jungreis, Yolanda Wu, Emma Vidal, Amay Jain, Isabel Tao, Megha Velakacharla


In a world that is constantly changing, the pursuit of happiness has been a constant for much of society. A plethora of research has been devoted to discovering the factors that drive happiness in many parts of the world. Most recently, the World Happiness Report has declared Finland, Denmark, and Iceland as the happiest countries in the world. Unsurprisingly, researchers around the world have dedicated their livelihoods to figuring out why.

This article quantifies happiness based on metrics from the World Happiness Report. This resource calculates the happiness score of countries around the world every year based on the Cantril ladder question where respondents are asked to imagine a ladder and rate their current lives on a 0 to 10 scale. The best possible life is assigned a value of 10 while the worst possible life is assigned a value of 0. Using these metrics and happiness scores, this article will dive into the intricate connection among many important aspects of life — specifically, age and GDP, the consumption of substances, the role of alcohol, the dynamics of social welfare, engagement of hobbies, and the influence of career with happiness around the world. By analyzing these factors, we hope to come closer to discovering the secret to happiness.


Figure 1

Here, after creating a scatterplot of the median age and the happiness index for 129 countries around the world, we can see a clear linear age-happiness relationship. The plot shows happier countries tend to have higher median ages around 40–45, and it is worth noticing that a decent number of these happier countries are concentrated in Western Europe such as Finland and Denmark — two of the top 10 happiest countries in the world. In contrast, Sub-Saharan Africa and South Asia exhibited lower happiness levels, corresponding to younger median ages, often around 20. To understand the factors contributing to higher median age and the associated happiness in some regions than others, we looked into the differences in migration rates, fertility rates, freedom to make life choices, and social support scores between Western European and Sub-Saharan African countries.

Figure 2

The four subgraphs reveal that compared with Sub-Saharan African countries, Western European countries associated with higher happiness levels appear to have higher migration rates, lower fertility rates, and higher freedom and social support scores. The higher migration rates are often attributed to the allure of stable social and economic environments, greater per capita land area, robust social welfare systems, high-quality healthcare, and enhanced educational opportunities. Lower fertility rates in these nations are likely influenced by various factors including better access to contraception and family planning services, as well as the emphasis on work-life balance and progressive social policies. Higher freedom to make life choices and elevated social support scores in Western Europe are indicatives of established societal structures, comprehensive social welfare systems, and policies that foster individual well-being such as parental leave, gender equality measures, and anti-discrimination laws. Collectively, these factors contribute to a more favorable environment for middle-aged individuals and families, enabling them to make choices aligned with their values and aspirations. The higher median age distribution in Western European countries can thus be understood as a consequence of a combination of factors that enhance overall life satisfaction and well-being, reflecting a societal landscape that promotes stability, opportunity, and personal fulfillment.


Figure 3

Next, we sought to analyze the relative importance of different recreational pastimes for happiness by creating a random forest regression model that determines which features are most important in predicting the happiness score of a country. It accomplishes this through the use of many decision trees; the model considers different groups of variables across many samples of the data, predicts each sample’s happiness score based on several conditions, and then calculates how important each variable is in predicting happiness score according to how accurate the model’s predictions are compared with the actual happiness scores. This model was created using a merged dataset that included the World Happiness Report in conjunction with a Kaggle dataset that contains pertinent statistics about how people spend their leisure time around the world. This dataset unfortunately only included data for 32 countries, but provided specific information that is hard to find anywhere else. This data includes how much time is spent per day sleeping, caring for one’s family, working, shopping, seeing friends, and much more. According to the random forest model, the variable that displays the highest importance when it comes to predicting the happiness score of a country is shopping. No other variable considered has an importance even half as high. This is an interesting finding as it seems to directly contradict the idea that money cannot buy one happiness. In order to further investigate, we created a histogram that displays the average time daily spent shopping by citizens of different countries and filtered it according to a country’s happiness score. Happiness scores range from as low as 1.8 to as high as 7.8, but we classified low happiness countries as having scores under 5, average happiness countries as having scores between 5 and 6, and high happiness countries as having scores over 6.

Figure 4

Per Figure 4, it seems evident that the countries with the highest happiness scores tend to also spend the most time per day shopping. There are nearly no average or low happiness countries that spend over 20 minutes a day shopping. This result is likely due to the fact that how often a person shops is directly tied to their country’s wealth, which greatly affects a country’s happiness score as we will later see in Figure 7. Generally speaking, when a country has a greater GDP per capita, it also has a higher happiness score. For instance, there is only one country with a happiness score under 5 that has a GDP per capita score of over 10. This observation alone essentially means that if a country has an above average happiness score, they must also nearly always have above average wealth. As a result, this trend in the data helps to explain why countries whose residents spend more time shopping tend to be happier: if a person is from a country with greater resources and accessibility to goods and services, it makes logical sense that they can afford to spend more time and money shopping, since their priorities are likely different compared to a person from a less affluent country. Residents of these less fortunate countries are more likely to be concerned with doing what they can to survive and saving money rather than spending it.


Figure 5

Drugs are known to have impacts on mood, including long term mental health effects. In this article, we will investigate the global impact of four different drugs. The first drug we will investigate is cannabis, the most widely consumed drug worldwide for both recreational and medicinal purposes. Global estimates suggest that 5% of the international population has consumed cannabis in some form within the past year. The second class of drugs, opioids, includes the illegal drug heroin as well as pain relievers such as oxycodone. Opiates can also produce a euphoric effect. The third class of drugs we will inspect are amphetamines, a stimulant that includes drugs prescribed to treat ADHD as well as illegal drugs including speed. Finally, the powerfully addictive stimulant cocaine, which increases levels of dopamine in the brain.

Cannabis is most widely consumed in North America, consumed by 17% of the adult population within the past year. In contrast, only 1% of the adult population in the East and Southeast Asia region had consumed cannabis in the previous year. The impact of cannabis on mood and mental health is unclear. While acute levels of THC, the primary psychoactive ingredient in cannabis, cause increased dopamine release, long term cannabis use has been linked to depression and decreased levels of dopamine in the brain. Cannabis was found to be positively correlated with happiness score, however the relationship was not found to be significant.

North America is also the region with the highest prevalence of both opiates and amphetamines. Opioids are used in prescription drugs, and opioid overdoses are the leading contributor to global drug deaths. Despite this, opioids appear to be positively correlated with happiness score, with a correlation coefficient of 0.38. Amphetamines have the highest positive correlation with happiness score, with a correlation coefficient of 0.48.

Despite a correlation between the prevalence of certain drugs and the happiness score for these regions, we are unable to determine whether either the short term impact on mood or the long term mental health effects of these drugs have any significant impact on happiness score. These differences could be due to a variety of other factors including the prevalence of drugs within developed countries when compared to developing countries. Wealthier nations have a higher rate of drug use and drug addiction. A controlled experiment would be necessary to determine the impact of substance use on happiness score.


Figure 6

In this next section of the article, we will explore the relationship between happiness and alcohol consumption by measuring the consumption of different types of alcohol between the happiest and saddest countries.

Per Figure 6 we can see that the drinking ranges from anywhere from around 1–12 litres of total alcohol per capita. Although the degree of alcohol consumption varies among the happiest countries, they generally exhibit a trend of consuming similar quantities of beer and wine, with a comparatively lower consumption of spirits.

In contrast, Figure 6 reveals that the less content nations, on the whole, display a preference for spirits over wine, while still aligning with the happier countries in their inclination towards higher beer consumption. The general trend of higher beer consumption in both happier and sadder countries could be explained by the fact that beer is a relatively popular choice for many social contexts.

Considering the positive correlation between GDP and happiness, we can see that in general, affluence contributes to contentment. Given the relatively lower cost of beer and spirits compared to wine and other alcoholic beverages, it is reasonable to infer that countries with lower happiness levels — often correlated with lower economic status — tend to favor beer and spirits over wine in their alcohol consumption patterns.


Figure 6

Figure 7

We can see that the distributions of the happiness and logged GDP per capita scores from the 2023 World Happiness Report are very similar. Their Fischer’s coefficients are also close (-0.45 for the former, and -0.48 for the latter). The negative sign indicates that they are skewed left. In the context of the happiness score, a negative skew implies that while most countries have higher happiness scores, there are a few countries with very low happiness scores. For GDP per capita, the left skewness could suggest that while economic wealth is widespread, there are still countries with very low economic wealth, which points toward global economic disparities.

Figure 8

Both of the distributions are strongly correlated, as revealed by this scatterplot (Figure 8) and their high Pearson’s coefficient of 0.78. This relationship could be due to the fact that stronger economies often indicate better living standards, such as access to quality healthcare and education, which contribute to overall well-being. However, correlation does not imply causation and happiness could be influenced by other factors not represented here.

Figure 9: Outliers (Afghanistan, Botswana, Venezuela, Lebanon)

We can see that there are some outliers: Venezuela has a very high happiness score compared to its GDP per capita score, while the remaining outliers (Afghanistan, Botswana, and Lebanon) have high GDP per capita scores compared to their happines score. These discrepancies suggest that factors beyond economic output influence happiness levels. For instance, political instability and economic crises could have significantly impacted Lebanon’s happiness scores, which have declined over the past decade. Similarly, Botswana’s high income inequality may contribute to its lower happiness score despite a relatively high GDP per capita. These examples highlight the complex relationship between wealth and well-being, indicating that a multitude of social, political, and environmental factors can affect the happiness of a population.

Figure 10

In this analysis, we looked into the happiness levels of 17 countries that ranked highest in the World Happiness Report for the year 2020. The focus was on illustrating potential connections between these happiness ratings and the corresponding COVID-19 death tolls in each country during the same period.

Looking at the graphs, New Zealand is a country with the least number of COVID-19 deaths among the selected nations. Though their happiness rating was still relatively high compared to other countries, New Zealand did not boast the highest happiness rating. Conversely, the United Kingdom, experiencing one of the highest counts of COVID-19 deaths, displayed a relatively low happiness index. Additionally, Denmark and Finland had some of the highest happiness indexes as well as low COVID-19 death counts. However, other countries do not show too much correlation between covid deaths and happiness, such as Israel and Iceland.

While the overall analysis reveals a nuanced connection between COVID-19 deaths and happiness levels with a slight inverse relationship, this data demonstrates that various other factors may be at play in contributing to happiness levels, including socio-economic, cultural, and policy-related factors.

Focusing more specifically at happiness within cities, the visualization below investigates the factors contributing to the happiness or unhappiness of residents in a city, emphasizing the correlation between various metrics and the subjective feeling of happiness.

Figure 11

This study aims to uncover key factors influencing residents’ contentment and satisfaction with their urban environment. The dataset, obtained from a survey, assesses several elements including the availability of information about city services, housing costs, public school quality, trust in local law enforcement, the upkeep of streets and sidewalks, and the presence of social community events. These factors are analyzed in relation to individuals’ self-reported levels of happiness. The findings indicate that an improved perception of these factors generally correlates with higher self-reported happiness among residents, with a few exceptions. This confirms the link between social infrastructure systems and reported happiness levels.

Interestingly, individuals reporting a complete lack of trust in the local police were more likely to report that they were happy, challenging the conventional assumption that increased or stricter law enforcement universally contributes to residents’ well-being and happiness. Moreover, there appears to be a threshold in perceived public school quality where the proportion of happy residents begins to decline, despite an improvement in perceptions about school quality. This decline could be influenced by various factors such as other community issues, personal preferences, or a plateau effect where the perceived school quality reaches a level where it no longer significantly impacts overall happiness. Despite these anomalies, it is evident that social welfare plays a significant role in the happiness of urban residents.


Finally, we wanted to look at the relationship between job satisfaction and the happiness index for various countries using available data from 2019. Merging the World Happiness Report and the HubSpot Global Report, we were able to look at 34 countries along with their respective happiness indices and job satisfaction percentages.

First up we have the top countries and their job satisfaction (measured as a percentage).

Figure 12

And then we have the top countries and their happiness indices.

Figure 13

As we can see the top countries for both graphs aren’t exactly lining up one to one.

To see if job satisfaction and happiness are correlated we can create a scatterplot.

Figure 14

There doesn’t seem to be a strong correlation between the two variables as seen by the large amount of variation maong the middle points, which we can possibly attribute to other variables such as the ones discussed above significantly influencing happiness. The one outlier at the bottom is India with a high job satisfaction and low happiness index suggesting that there are definitely other factors at play that contribute to the lower happiness index.

The scatterplot seems to be a better fit for a cluster analysis (as shown below), but even then there isn’t a relationship we can observe that relates both job satisfaction and happiness in a way we can use for predicting.

Figure 15


In conclusion, our exploration into the factors affecting happiness has uncovered a complex web of relationships that shape the well-being of individuals and societies. From the impact of age and socioeconomic factors to the role of recreational activities, substance use, and career satisfaction, the intricate connections between these variables offer valuable insights into the pursuit of happiness.

The age-happiness relationship revealed a clear trend, with countries boasting higher median ages generally reporting higher happiness levels.

Exploring the significance of hobbies, our analysis highlighted an unexpected connection between shopping and happiness. Contrary to the notion that money cannot buy happiness, countries with higher happiness scores tended to spend more time shopping, suggesting a link between economic affluence and contentment.

Delving into the realm of substance use, our investigation into the impact of drugs on happiness unveiled intriguing correlations. While cannabis showed a positive but non-significant relationship with happiness, opioids and amphetamines exhibited positive correlations.

Alcohol consumption patterns also demonstrated a connection with happiness, with more wealthier nations favoring wine. The interplay between economy, social welfare, and happiness became apparent, emphasizing the strong correlation between GDP per capita and happiness scores.

The analysis further extended to the impact of external events, such as the COVID-19 pandemic, on happiness levels across countries. While a nuanced relationship between COVID-19 deaths and happiness emerged, it highlighted the influence of various socio-economic, cultural, and policy-related factors on overall well-being.

Zooming into urban settings, the study revealed the significance of social infrastructure systems in influencing residents’ happiness showing that factors such as trust in local law enforcement and perceptions of public school quality played crucial roles.

Lastly, the exploration of job satisfaction and its correlation with happiness unveiled a more nuanced relationship. While some countries exhibited alignment between high job satisfaction and happiness, outliers like India suggested the presence of other influential factors.

In essence, our comprehensive analysis underscores the intricate and multifaceted nature of happiness, woven together by a tapestry of interconnected factors. While age, economic status, leisure activities, substance use, and career satisfaction all play roles, the complexity of human well-being demands continued exploration and research to unravel the secrets of a happier life.