School attendance and the perceived value of formal education: Evidence from Tanzania

The benefits of universal primary education (UPE)—ranging from increased personal wellbeing to socially important outcomes such as lower population growth and improved maternal and child health—are widely documented, and donor organizations have invested significant amounts of money to reduce barriers to education. However, there are still many children—and girls, in particular—who do not attend school. Countries in sub-Saharan Africa (SSA) have not attained rates of primary school attendance as high as countries in other parts of the developing world. Authors of the Millennium Development Goals (MDGs) listed UPE in SSA as a core need for achievement of the MDGs. Economic models of educational choice have provided an important framework to think about determinants of school attendance (Becker, 1975). These determinants include benefits, such as higher wages, and costs—both the direct costs households incur in paying school attendance fees, and indirect costs to the household, such as lost labor (e.g. Glick, 2008; Handa, 2002). Pairing these theoretical models with data on school attendance has yielded important insights into decision-making. These insights have been used to guide policy (Fiszbein, Schady, & Ferreira, 2009), and school attendance rates in SSA have increased significantly over the past thirty years (Lincove, 2015). However, as attendance rates have increased, the children who are not in school tend to live increasingly in marginalized, rural communities that do not have the data necessary to understand parents’ schooling choices with the standard model. In many of these communities formal labor markets are thin or non-existent, requiring students to migrate to urban areas to take advantage of the financial benefits of increased education. Additionally, these communities are much less likely to have a history of formal education, reducing opportunities for parents to learn about benefits of education through experience or social networks. Gender disparities in school attendance are frequently also higher than in the general population, and may be due to cultural beliefs or household reliance on female labor. Cumulatively, these factors may engender more inter-household heterogeneity in the conceptualization of the benefits of education. To address gaps in the evidence, this paper introduces data on household leaders’ ideas about the positive or negative effects of education into a model of schooling choice. We collected data on current education decisions and expenditures for children (both children attending and not attending school), educational levels achieved for adults, and male and female assets and income from 196 households of three sedentarized pastoralist and agro-pastoralist tribes living in rural south-central Tanzania. Further, we gathered data separately from male and female heads of household on perceptions of the effects of education for male and female children and on whom in the household or community makes decisions about school attendance for the household’s children. We also have, among other variables, household-level data such as distance to water sources, agricultural and livestock holdings (a factor in the household’s opportunity cost of sending their children to school), and distance to school. There is marked variation in education choices among the tribes, ranging from a low of 25 percent of school-aged children (with 15 percent of them female) attending school to over half of school-aged children (and 49 percent of them female) attending school at the high end. Interestingly, the tribe investing the most in schooling is also on average the poorest in terms of livestock and agricultural holdings, which comprise the bulk of these communities’ wealth. Household leaders’ perceptions of the effects of education encompass a range of benefits and costs. Among these are beliefs that would fit with the motivations commonly assumed to be drivers of education—the ability to get a job, earn wages or a salary, or improve their (and their family’s) material standards of living. However, other households viewed education as a safeguard against exploitation (more frequently listed for females than males), as a way to help their families adapt to a changing environment, as a public good for the entire community, or as a benefit to the students’ abilities to manage the household’s livestock. A significant minority of households expressed ambivalence (“there is no value to education”) or opposition, citing the potential for moral decay, to education; non-positive sentiments tended to be expressed more frequently about educating female than male children. Using data on household composition, school attendance, and education perceptions, we estimate models of whether a household chooses to educate any students, and of the number of students currently being educated, for the entire sample and for males and females separately. Our findings confirm previous results from work on education, while adding new insights. The education levels of household leaders—and of mothers in particular—are an important determinant of children’s schooling. Households in which fathers alone make the schooling decisions educate fewer females and fewer children in general. More female income is associated with higher school attendance. New insights stem from perceptions of education. We find that perceptions of the educational benefits are important in understanding the schooling decisions. Interestingly, however, there are differential effects among the benefits. Households indicating that the opportunity to get a job and earn wages is an important benefit of education are more likely to educate children than households that do not view this as a benefit, while households stating that education is important for children to become better herders are less likely to educate children than those who think this is not an important effect. The novel integration of decision-makers’ perceptions of the benefits of education into the analysis of education choice yields interesting findings. While some of these findings support the standard assumptions of human capital accumulation models and corroborate previous findings, there are new insights into households’ educational choices that will stimulate healthy discussion about the nature of education in marginalized populations, and implications for the achievement of MDGs in SSA. References Becker, G. (1975). Human Capital. New York: National Bureau of Economic Research. Fiszbein, A., Schady, N.R., & Ferreira, F.H. (2009). Conditional cash transfers: Reducing present and future poverty. Washington, DC: World Bank Publications Glick, P. (2008). What policies will reduce gender schooling gaps in developing countries: Evidence and interpretation. World Development, 36(9), 1623–1646. Handa, S. (2002). Raising primary school enrolment in developing countries: The relative importance of supply and demand. Journal of Development Economics, 69, 103–128. Lincove, J.A. (2015). Improving identification of demand-side obstacles to schooling: Findings from revealed and state preference models in two SSA countries. World Development 66: 69-83