2016 Global Youth Development Index: Results and Analysis all domains except Political Participation, where the region scores just a fraction above the global average. Compared with other regions, volatility in YDI scores in the five-year period was greatest in Sub-Saharan Africa. The YDI scores in 83 per cent of countries improved, the most significant gains being in the Education and Health and Well-being domains. The two countries in Sub-Saharan Africa where youth development seems to have worsened the most are Angola and Chad, which declined by 11 per cent and 7 per cent, respectively, between 2010 and 2015. Other countries that deteriorated did so by less than 5 per cent. Angola registered a dramatic decline (of 70 per cent) in the domain of Civic Participation, and a moderate fall was also recorded in the domain of Employment and Opportunity. Health and Well-being was the only domain in which Angola made notable progress, thanks to gains in the indicators of youth mortality, adolescent fertility rate and alcohol abuse YLL. Chad’s score deteriorated in both the Civic Participation and Political Participation domains. Indicators that contributed to this deterioration include the NEET rates, and numbers who volunteered time or voiced an opinion to an official. Kenya’s overall YDI score increased by 22 per cent between 2010 and 2015, the biggest improvement not just in Sub-Saharan Africa but also globally. Improvements were recorded in all domains, the largest being in Civic Participation (61 per cent), Health and Well-being (39 per cent) and Political Participation (38 per cent). Indicators that contributed the most to this progress are volunteered time, voiced an opinion to official, helped a stranger, youth mortality, alcohol abuse and mental disorder. Kenya scores above the SubSaharan African average in all domains. South Africa’s YDI score recorded 20 per cent progress between 2010 and 2015, making it the second biggest improver in the world. Gains were the largest in the domains of Civic Participation, Political Participation and Health and Well-being, respectively. Indicators that improved the most include volunteered time, helped a stranger, alcohol abuse, drug abuse, mental disorder and youth mortality. Despite making significant progress in the Health and Well-being domain, South Africa continues to perform very poorly in this domain, with a score well below the SubSaharan African and global average. This is explained by the country’s high youth mortality rates and high HIV prevalence among young people, especially females. The prevalence of HIV is nearly twice as high in young females as in young males. Niger’s YDI score has improved by 19 per cent since 2010, helped by a dramatic increase in the scores for the Political Participation (61 per cent) and Civic Participation (18 per cent) domains. Moderate improvements were also recorded in the Health and Well-being and Education domains. These gains are made possible by improvements in the indicators of voiced an opinion to an official, existence of youth policy, volunteered time, youth mortality, and mental disorder YLL. Niger scores above the Sub-Saharan African average in the domains of Health and Well-being and Political Participation. Other Sub-Saharan African countries in the region whose YDI scores have improved by at least 10 per cent since 2010 include Senegal, Republic of the Congo, Guinea, Sierra Leone and Mozambique. Despite being in the lowincome category, all of these countries increased enrolment in secondary education and/or literacy rates for their young people and also managed to bring down their adolescent fertility and youth mortality rates. Box 2.3 Correlation versus causation Correlation is a statistical measure that reveals the extent to which two variables are associated. The correlation coefficient indicates the strength and direction of the relationship between two variables. Correlation simply means that change in one variable corresponds with change in a second variable. For example: the statement ‘where countries have large youth bulges we find they also have low YDI scores’ means that, in more cases than not, a country that has a youth bulge is also likely to have a low YDI score. It does not mean that the youth bulge has led to poor youth development or that poor youth development has led to larger youth bulges. This would be causation, where variation in one variable cause or create variation in a second variable. Source: Urdan 2010. \ 55