SAGA Progress Report - Senegal/West Africa

SAGA
B16 MVR Hall
Ithaca, NY 14853
(607) 255-8931
Fax (607) 255-0178
saga@cornell.edu

SAGA Progress Report
April, 2003

I. RESEARCH
4. Senegal/WARP

Most of our efforts have gone into the education and related thematic research in Senegal where we are investigating the household, community, and school-level determinants of the following education outcomes in Senegal: primary and secondary enrollment, transitions from primary to secondary school, grade repetition and dropout, and learning—both academic (math and French test scores) and non-academic ("life-skills"). We will do this by supplementing an existing five-year cohort survey of school children with additional comprehensive data collection on households, communities, and school alternatives in areas where the original data were collected. Combining the data sources, a range of econometric approaches will be used to explain variation in these education and cognitive outcomes. In addition, the research will examine related factors such as entry into the workforce after school, children’s time allocation, and schooling differences among siblings and between boys and girls, migration and health and their relationships to education, the impact of shocks on health and education outcomes, and the role of indigenous and formal institutions in affecting household welfare.

The actual survey commenced in the end of March. This followed many months of working with our partners on developing the questionnaires, sampling procedures, data entry programs, and other procedures. In addition, we spent two months training enumerators and supervisors. Furthermore, we needed to arrange complementary financing, which came from the USAID Mission, UNICEF, the World Bank, the French Ministry of Development Cooperation and the Ministry of Education. We anticipate that we will complete the survey at the end of June. A first workshop on the plans for analysis of the data, sponsored by CREA and Ministry of Education, will take place at the end of April. A series of workshops will then take place over the next year, as the data analysis begins to yield results. These workshops are designed to ensure maximum integration of the results into the policy-making process.

We also continue our efforts focusing on HIV/AIDS. As noted in the previous progress report, we had devised extensive plans for evaluating HIV prevention interventions, particularly Voluntary HIV Counseling and Testing, in the high-prevalence context of Côte d’Ivoire. As also noted in that report, the political crisis and violence in Côte d’Ivoire, which at that time has only recently begun, put these plans into doubt. At this point, it is clear that Côte d’Ivoire will be unsuitable for the research for the foreseeable future as the conflict is far from being completely resolved.

Our main response in the last several months has been to turn our efforts, for the time being, to the analysis of secondary data. Owing to the availability (primarily) of Demographic and Health Survey data from many countries, there is a great deal of analysis that can be done. For many countries, these surveys have been fielded more than once in the past decade, with detailed modules on HIV/AIDS knowledge, testing, and sexual behavior. The existence of multiple rounds makes it possible to examine how behaviors and knowledge have been changing over time as the epidemic, and general awareness of it, has expanded. Understanding how these changes are related to socioeconomic variables as well as location (e.g., rural, urban) will help better pinpoint gaps in knowledge and behavior that prevention and education programs can target. This type of cross-country, over-time analysis has yet to be done. Data from multiple years are—or soon will be—available from the following countries: Benin, Malawi, Tanzania, Zimbabwe, Uganda, Ghana, Kenya, Madagascar, Mali, Nigeria, and Zambia.

Preliminary analysis on the data for Tanzania (1996 and 1999 DHS) suggests that even over a relatively short period of 3 years the population became more knowledgeable about HIV/AIDS and the methods to prevent the risk of infection, in both urban and rural areas, and among both men and women. However, geographical and gender gaps in the knowledge of the HIV prevention methods remain. There is also evidence of positive behavioral change: for example, condom use increased across all age groups and educational levels. Further, behavioral changes (in particular, the probability of condom use) have been the largest in youngest cohorts, most educated, and urban residents.

It is important to replicate the analysis for other African countries. This includes not just high-prevalence countries like Kenya and Malawi but low prevalence countries like Madagascar and Benin: identifying where or among which groups the gaps in knowledge or safe behaviors are largest in these countries, and where they are failing to change or are changing too slowly, will be key to preventing the epidemic from taking hold.

Madagascar will be a particular focus of the research, consistent with the interest of the USAID Mission and their new strategic objectives (discussed above). This work will make use of DHS surveys from 1997 and 2003 (the latter is soon to be put in the field). What makes the Madagascar analysis of special interest is that we will be able to integrate the DHS data with other sources, such as the recent commune census and the population census. This will provide a better understanding of how factors outside the household, such as the availability of community health or fertility centers as well as roads and other infrastructure (or more generally 'remoteness'), affect the speed of diffusion of HIV knowledge, testing, and sexual behavior. In contrast, the DHS surveys themselves are limited to information about individuals and their households.

Previous Section | Next Section

Return to SAGA Progress Report (April 2003) Table of Contents