SJ WorldNews - шаблон joomla Авто

Working paper - 11 Working-Age Adult Mortality and Primary School Attendance in Rural Kenya

Author(s):  Yamano, Takashi; Jayne, Thomas


The onset of the AIDS epidemic in the past two decades raises serious questions about long term human capital development, including education, and the future quality of the workforce in areas where the disease is particularly acute, such as in many countries of Africa. There is growing concern over the effect of high AIDS-related adult mortality and illness on child welfare and the disease’s effects on their potential over the long run to support themselves as adults and to contribute to their countries’ development (UNICEF 1999; Bell, Devarajan, and Gersbach 2003).

Available empirical evidence indicates that school enrollment rates are lower among AIDS orphans compared with non-orphans (World Bank 1999; Ainsworth, Beegle, and Koda 2002; Case, Paxson, and Ableidinger 2002). A long list of anecdotal case studies also supports this evidence (Gachuhi 1999; Nyambedha, Wandibba, and Aagard-Hansen. 2001; Guest 2001; USAID 2002).

However, using 39 nationally representative data sets collected in the 1990s from 28 countries, mostly in Sub-Saharan Africa, Ainsworth and Filmer (2002) show that the difference in enrollment rates between orphaned and non-orphaned children varies greatly across countries and wealth levels within a country.

These studies focus on the education status of orphans, regardless of the timing of parental deaths. The impact of parental mortality, however, could depend greatly on the timing of such mortality. More importantly, most children faced with the death of adults in their households do not become orphans, and so the broader question of how children’s schooling is affected by the death of adults in their households remains largely unknown.

This paper, therefore, focuses on the impact of working-age adult mortality on child primary school attendance. A major difficulty in measuring the impact of adult mortality, especially mortality attributable to AIDS, is that it is caused by behavioral choices rather than by random events. Individuals and households incurring adult mortality are more likely to display certain characteristics. For example, especially in the early years of the epidemic in sub-Saharan Africa, evidence suggests that men and women with higher education and income were more likely to contract HIV than others because they were more likely to have numerous sexual partners (Ainsworth and Semali 1998; Gregson, Waddell, and Chandiwana 2001). If prime-age mortality remains correlated with individual and household characteristics such as social status, wealth, and mobility – which are also important determinants of school enrollment – failure to control for these characteristics may generate biased estimates of the impact of adult mortality on school attendance. We overcome this problem, to some extent, by estimating household- and child-fixed effects models using panel data. Evans and Miguel (2004) took a similar approach in a recent paper, using a five-year panel of 17,000 children in Busia district of western Kenya.



Working-Age Adult Mortality and Primary School Attendance in Rural Kenya






Subscribe to our newsletter

Style Setting




Template Widths

px  %

px  %