Happiness, defined as ‘the degree to which an individual judges the overall quality of his life-as-a-whole favorably’ (Veenhoven, 2008), is a very subjective measure and therefore difficult to measure. Sociologist Dr. Ruut Veenhoven of Erasmus University Rotterdam has created the ‘World Database of Happiness,’ organized according to nation, with results derived from empirical research on factors associated with subjective appreciation of life. Through this, Veenhoven has produced a list of nations ranked according to their happiness index, a numerical measure ranging from zero (least happy) to ten (most happy). According to Veenhoven, Denmark leads world happiness with a happiness score of 8.2. Togo and Zimbabwe, both African countries, are at the bottom of the list, each with a score of 3.3 (Veenhoven, 2008). In this report, a choropleth happiness index map was produced in order to determine any geographical patterns of happiness. Then, using data from the International Monetary Fund, nominal GDP/capita data was collected and a multivariate choropleth map was created to determine whether a correlation between happiness and money (GDP / capita) exists.
Comments Off on Introduction
Posted in Introduction
Happiness Index figures were derived from Dr. Ruut Veenhoven ‘World Database of Happiness.’ (Erasmus University, Rotterdam). In Arcmap, a worldwide, country level shape file was imported. In this layers attribute table, a new feature, ‘happinessindex,’ was created. In an editing session, numerical happiness scores for each country were added. Under the properties symbology tab, the data was categorized into four groups and classified using a graduating color scheme. In layout view, standard map elements were added, including a title, author, source, date and legend.
In order to create a bivariate choropleth map, mapping the relationship between GDP and happiness, GDP/capita and happiness index data for each country was listed and sorted in an excel document. Three categories for happiness index and GDP / capita were defined by dividing the excel list into three even sections. The three categories are outlined in Table 1, which sorted each country into one of nine categories. A three color choropleth color scheme was then derived using the ‘color brewer,’ website to represent the relationship between the two variables. In Arcmap, each country was selected and their color properties individually altered to be represented by the correct color. In layout view, standard map elements were added including a legend, title, source and author. Using the Excel document, a scatter plot chart was created plotting the same relationship between GDP/capita and happiness.
Figure One is the final cartographic product showing worldwide happiness categorized into four groupings. The countries displaying the highest happiness indexes are found throughout Europe, in particular Scandinavian Europe, North America, and Oceania. Countries found in the lowest happiness category can be found throughout Africa, Asia and Eastern Europe. Figures Two through Five are versions of the same map, yet zoomed into various regions for more detail. The ‘Americas’ map, in particular the area zoomed into Central America, highlights the relatively high happiness ranking among countries within this region, with the exception of Nicaragua and Haiti, a Central American country which ranked very low on the United Nations Human Development Index (2006). In Africa, Senegal, Ivory Coast and Nigeria all faired much higher in terms of happiness in comparison to other countries in the same region. Another interesting geographical trend in happiness was among former Soviet Union countries which faired low on the happiness scale. In particular, Armenia, Ukraine, Georgia, Bulgaria, Moldova and Russia fell into the lowest of four happiness categories, while Bylarus, Estonia, Kazakhstan, Latvia, Lithuania and Azerbaijan, well into the second lowest of four categories. Turkmenistan is the only former Soviet Union country with a happiness ranking above the bottom two categories. A few countries in the Middle East, Saudi Arabia and Kuwait, were categorized into the highest happiness category, yet other countries within this region did not follow this trend.
Figure six is the final cartographic product created to show the correlation between nominal GDP/capita and happiness. Each country falls into one of nine categories and is given a color scheme to reflect this. Figure seven is a scatter plot demonstrating this same relationship. In general, the scatter plot demonstrates a generally positive relationship between GDP/capita and happiness index. Therefore, on average, countries with a high GDP/capita have higher happiness rankings. Using the map and the scatter plot however, some interesting outliers were noted that did not correspond with this relationship. For example, Botswana, Africa was the only country to fall into the high income, low happiness category, shown in light pink on the map. Another very interesting pattern noted was among Latin American countries, in particular, Mexico, Columbia, Guatemala, El Salvador, Honduras, Costa Rica, Venezuela, Brazil, Argentina, Dominican Republic, Chile and Uruguay, which made up 12 of the 14 countries which fell into the medium income, high happiness category. More specifically, in the scatter plot, which does not divide GDP and happiness into three categories but instead individually plots each country, it is interesting to note the Latin American countries with very low GDP’s and very high happiness indexes. One such country is Columbia with a nominal GDP/Capita of $4264 USD and a very high happiness index at 8.1, ranking just behind leader Denmark (8.2). Please note that all images are found here in attached files.
In this project, it is very important to note the difficulty of quantifying happiness as it is a very subjective measure which must take numerous cultural, religious, social, political, personal and economic factors into consideration. Figure Eight shows the various data collection methods Ruut Veenhoven utilized to create his database of happiness. His selection of empirical studies involved scholarly information about the measurability of happiness, happiness indicators and theories and various other topics. His selection on ‘valid measurements’ involved citizen answers to numerous questions about past and present feelings of happiness and sadness. The validity of this particular collection method could be easily skewed if a group of people properly representing the population was not chosen. If these questions were asked by phone or internet, countless people with no access to these communication methods would likely be miscounted. Further, if these questions were asked in person, people living in remote areas, particularly those in lesser developed countries, may not be included. Further, the concept of happiness is very dependent on comparison of one’s lifestyle to that of others. In many developing countries, particularly within remote regions, people may be more content with their lifestyle as they have nothing to compare it to. With the internet becoming more and more assessable, even to those in developing countries, this may change the way people feel about their own lifestyles as they become exposed to the more lavish lifestyles of those living in developed countries.
The use of average figures can also be misleading as they often fail to represent the data’s range. For example, Botswana has a very high nominal GDP/capita, lending to impressive economic gains since independence in 1966 supported by the diamond industry which accounts for more than 1/3 of the country’s GDP (CIA World Factbook, 2008). However, their extremely high unemployment rate, 23.8% in 2004 with unofficial estimates as high as 40% (CIA World Factbook, 2008), suggests that there is a wide disparity among the rich and poor, making GDP/capita a misleading figure. In the case of Botswana, a country with the second highest HIV/AIDS infection rates in the world, happiness indexes, which in this study were very low despite the high GDP, therefore may be a more appropriate measure of human development than GDP. This is exactly what Bhutan’s King Jigme Singye Wangchuck did when he announced in the 1980’s that Bhutan would no longer follow popular economic indicators of gross national product but instead would use a “gross national happiness” measure to gauge development progress (Harris, 2004)
Inaccuracies surrounding the use of choropleth maps and average measures were another concern in this project. As stated by Longley et al (2005), choropleth maps can be highly misleading as the same color is applied uniformly to each part of an area, yet we know that the mapped property cannot be true of each part of the area. Therefore, choropleth maps are best suited for displaying geographic phenomena evenly distributed within each boundary and not skewed by outliers. In this case, outliers play a strong role in each countries data, yet given the large scope of this project, where the maps encompass the entire globe, showing variations within each country is not possible. In order to show such variations, it may be possible to focus on one country or region and create a dasymetric map, which uses the intersection of two datasets to obtain more precise estimates of spatial distribution (Longley), however, this would require a much more extensive data source.
Another concern surrounded the number of classification categories to include in the bivariate map. When constructing figure 6, the original intention was to construct a four by four, or sixteen variable, choropleth map. This would have divided the happiness index and GDP/capita into four categories, instead of three, constructing a more accurate result. Table 2 highlights how the results would have changed if four GDP and happiness categories were constructed. Unfortunately, as described by Leonowicz, in her 2006 paper on ‘Two Variable Choropleth Maps as a Useful Tool for Visualization of Geographical Relationships,’ the number of classes displayed in a bivariate choropleth map should be possible to deal with by the reader. Therefore, the number of possible combinations is limited to 4 (2 X 2) or 9 (3X 3) classes which are easily interpreted as readers can intuitively select low and high values or low, medium and values. Leonowicz references Olson (1981) who conducted a study that proved maps with more than 9 classes are too complex for the users. Therefore, figure seven is a very useful supplementary figure as it does not categorize countries into groups, and gives the reader an excellent idea of the correlation between the two variables. The division of classification categories was another step which greatly affected the outcome of the maps. In this project, it was decided to divide GDP and Happiness so that an even amount of countries fell into each category. This was perhaps not the best way to do this, as it resulted in a very large range in the top GDP category ($16,000 – $103,000). This resulted in Luxembourg, arguably the richest country in the world, being grouped with countries such as Trinidad and Tobago, a country with a nominal GDP/capita of $16,042 USD, just over the $16,000 cut off. If this project were to be repeated, the categories would likely have been divided using a different method to reflect this problem.
As presented in the results section, some interesting ‘outliers’ were observed in figures six and seven. Latin American countries, in particular El Salvador, Mexico, Guatemala, Columbia and Puerto Rico, faired high on Ruut Veenhoven’s happiness scale, despite mass poverty and corrupt governments and low GDP/capita’s. This is a trend sociologists have dubbed the ‘Latino Bonus.’ Eric Weiner, author of ‘The Geography of Bliss,’ explains how social scientists believe this trend may be the result of strong family relationships which enact as sense of belonging in Latin America (Ross, 2008). Former Soviet Union countries also represented an interesting trend in happiness and GDP. Table Two shows numerous former Soviet Union countries in the $4,001-$20,000 USD GDP/capita range and how they rank much lower in happiness rankings compared to their Latin American counterparts within the same GDP range. Many social scientist point to communism as a common theme within these countries. Ronald Inglehart and Hans-Dieter Klingemann, two of the leading academic researchers into the data on happiness, note that virtually all societies that experienced communist rule show relatively low levels of subjective well-being, even when compared with societies at a much lower economic level (Wright, 2008). This makes many question whether people were happier during the communist regime? However, a multitude of factors including the toll the fall of communism took on citizen’s health, rising poverty rates and rates of inequality and high levels of corruption among post communist governments could all play a role in decreasing happiness rankings in these countries (Wright, 2008).
To conclude, it is important to note that countless factors, many which were not discussed in this report, play imperative roles in determining happiness. Through the creation of cartographic products mapping happiness among countries, using figures from Dr. Ruut Veenhoven’s ‘World Database of Happiness,’ interesting geographical happiness patterns were observed, particularly among Latin American and former Soviet Union countries. Correlating this data with nominal GDP/capita figures from the International Monetary Fund, it was determined that a generally positive relationship exists between GDP/capita and happiness rankings. However, numerous outliers were also observed which did not correlate with this observation, again stressing the importance of considering the countless factors which play a role in determining happiness.