Human Development in Sub-Saharan Africa at the turn of the Millennium

Nathaniel Ayewah

Spring 2006

Introduction

In this project, we study the development of countries in Sub-Saharan Africa using various economic and social indices. We are particularly interested in the impact of industrial and corporate activity as well as the benefits of government responsibility. With all the focus on some of the conflicts and diseases in this part of the world, it is easy to forget the economic activities and ignore their potential.

This project combines datasets from the World Bank and Transparency International. Figure 1 shows all the indices studied grouped by areas of interest. Data Analysis was done using Spotfire (http://www.spotfire.com/) and TreeMap (http://www.cs.umd.edu/hcil/treemap/).

Human Development Indices Mortality rate, infant (per 1,000 live births) Mortality rate, under-5 (per 1,000) Literacy rate, adult female (% of females ages 15 and above) Literacy rate, adult male (% of males ages 15 and above) Life expectancy at birth, total (years) Malnutrition prevalence, weight for age (% of children under 5) Poverty headcount ratio at national poverty line (% of population) Immunization, measles (% of children ages 12-23 months)	Foreign or Corporate Investment Foreign direct investment, net inflows (current US$) Industry, value added (% of GDP) Services, etc., value added (% of GDP) Aid per capita (current US$) Aircraft departures
	Economic Development GDP (current US$) GDP growth (annual %)
Government Responsibility *Corruption Perceptions Index School enrollment, primary (% net) School enrollment, secondary (% net) Roads, paved (% of total roads) Cash surplus/deficit (% of GDP)	Technological Development Internet users (per 1,000 people) Personal computers (per 1,000 people) Fixed line and mobile phone subscribers (per 1,000 people)
	Debt Short-term debt outstanding (DOD, current US$)
*Source: Transparency international (http://www.transparency.org/policy_and_research/surveys_indices/cpi/), Corruption Perceptions Index is the collective assessment of experts and business people surveyed on the perception of the level of corruption in a country. Values range between 10 (highly clean) and 0 (highly corrupt). All others indices are from World Bank Development Index (http://devdata.worldbank.org/data-query/) Bold: The indices visualized in this report.

Figure 1: Indices Analyzed for this report

Countries with Less Government Corruption have Lower Child Mortality Rates

Our analysis showed that the corruption index is more strongly correlated to the child mortality rate than economic performance measures like GDP or debt. In Figure 2, both South Africa and Namibia have low mortality rates and high trust scores even though they have vastly different gross domestic products and debt concerns. Furthermore, Nigeria, a country with a large GDP has high mortality rates and high corruption.

Figure 2: Scatter plot of Mortality Rate and Corruption

In the TreeMap, the size corresponds to the Corruption Perception Index with larger boxes indicating higher trust scores, while brighter boxes indicate low mortality rate. It is a little harder to see this relationship in the TreeMap. However, certain points stand out better. It is a lot more noticeable that Mauritius has low child mortality while Angola has very high mortality. Botswana is also more conspicuous as the least corrupt country in this region.

Figure 3: TreeMap of Mortality Rate and Corruption

Some Countries Rich in Natural Resources have Weak Service Industries

We stumbled upon an interesting observation while experimenting with Spotfire’s Heatmap. In Figure 4, we map the Value Added by Industry beside the Value Added by Services (as a % of GDP). The rows are ordered by industry value from low value (dark green) to high value (bright green). The trend in the services column seems to go from dark (low Services value) to bright (high Services value) and back to dark again. This last group is most interesting because they represent countries that get high value from their industries but do not have a developed service industry.

Figure 4: Trends in the value added by Industry and Services

We look to a scatter plot to illuminate this trend. In Figure 5, we see a curve trend that corresponds to the shading of the heat map, and observe that the interesting group includes Nigeria and Angola (the two largest oil producers in Africa), and other countries rich in mineral resources.

One of Spotfire’s strengths is that it can quickly add up to 6 variables to the scatter plot, each represented by different characteristics including x-axis, y-axis, color, size, shape and rotation. In Figure 5, we noticed a correlation between the value of services and the availability of the internet and computers, suggesting that as African economies seek to be more service oriented, the internet and computers will play a significant role.

Figure 5: Trends in the value added by Industry and Services

Countries with a Higher Services Value Perform Better in Health Indices

The previous scatter plot (in Figure 5) also suggested some interesting relationships between services and health indices. To illuminate these relationships, we created the scatter plot in Figure 6. This shows that the mortality rate goes down for countries for which services contribute a greater percentage of the GDP.

The nodes are colored by the percentage of children with measles immunization and show that higher rates of immunization occur in countries with higher Services value. The correlation between these health indices and Services is stronger than their correlation to GDP suggesting that the development of the Services industry is a better measure of overall human development.

Figure 6: Relationships between Services Value and Health Indices

We compared the visualization in Figure 6 with the Tree Map in Figure 7. Here, size corresponds to the value of Services while brightness corresponds to the percentage of children receiving measles vaccinations. Again it is harder to spot the trend with a TreeMap but easy to spot large values. Here Botswana and Ghana have the highest vaccination rates. In the scatter plot in Figure 6, these two have the brightest green nodes but this is hard to spot because both nodes are small, buried in the trend and obscured by Spotfire’s markings and labels.

Figure 7: Relationships between Services Value and Health Indices

Comparing Spotfire and TreeMap

Spotfire offers more options for visualizing trends but sometimes the complexity can be overwhelming. We initially started our analysis by using TreeMap to try and quickly identify potential trends. We soon ran into the limitations of TreeMap when we wanted to look at many variables.

TreeMap also has specialized coloring schemes which made it easier to spot some trends. This was necessary in our dataset because some very large values skewed the coloring in Spotfire, which uses a linear color scheme. In TreeMap, values can be arranged on a logarithmic scale to bring them closer together before coloring. They can also be arranged in buckets of equal density which ensures values clustered together get close colors. To improve coloring in Spotfire, we had to transform some of the data columns using logarithms. For example, Figure 2 is colored according to the following logarithmic transformation of GDP:

Max(Log10([GDP (current US$)]), 9)

Conclusions

Our analysis suggest that reducing corruption and encouraging the growth of Services both play a role in improving the human development of countries in this region. Nigeria and Angola are noticeable in this analysis because of high corruption and low Services value. Yet both of these countries are financially prosperous, in part because of oil revenues. International investment in natural resources will not by itself be the answer to the region’s human development needs. More emphasis on services is needed.