Emit Fr~~~~~~~~~~~- F - A joint publication by the Policy Research & the Poverty and Social Policy Departments of the World Bank No. 2 April 1996 Investing in Education Igt. brnn the Education qu~~~~~~tyof~~~~i1 s~chool up.. y toftelvlo h aising primary school comple- 25% of primary school-aged children do e o e the R tion rates is a clear priority for not even enter the school system, and less ...... .Lvdeveloping countries. Despite than 50% reach the last year of primaiy a d net enrolment rates of 90% or more, school. particularly in lower- and upper-middle Such dropout rates are of concern m tW l S, m 9.3% income countries, completion rates re- for two reasons. First, lower grade at- main substantially lower. On average, tainment impies a loss of potential t 3..2%. only 59% of the school age population in human capital, an important determinant Two recent World Bank studies, drawing on household and community Primary School Net EnroHment and Completion Rates in surveys in Ghana and Egypt, address the Dleveloping Countries (1990 or latest available year) question of how primary school enroll- oment and completion rates can be raised. 100E School enrollment in 90 .. ...... 80 - . ;. low-income countries: 70- Percent of 60 f new evidence from school-age 50 R. llment Ghana children 40 i Enrollment 30- E~~~~~~~~~~ Completion When decidinig on whethier to enroll 20l children in primary school, parents take 10 into account the cost of all future Lo Low e Uppe schooling-not just the cost of attending Income Middle Middle primary school. This is the main con- Income Income clusion drawn from an exhaustive study of the determinants of primary school enrollment in rural Ghana. The net enrollment rate represents the proportion of school-aged children currently enrolled The study is based on data from the inprimaryschool. The completion rate shows the proportion of this population that success- 1987 Ghana Living Standards Survey fully finishes the primary school cycle (sources: World Development Report, 1993; (GLSS), which covers 1,850 rural Lockheed & Verspoor, 1991). households and 1,902 primary school- age children. A complementary com- lower-niddle income countries and 76% of economic growth. Second, high drop- munity survey provides information on in upper-middle income countries reach out rates reduce the efficiency of the the schools accessible to each child. This the last grade of primary school (see public education system. If the goal is to wealth of data allows the study to control chart). get a certain number of students through for access, quality and costs of schooling, In low-income countries the situation a given level of schooling-say primary as well as individual student and house- is even more acute-enrollnent as well school-then having students drop out hold characteristics, when exploring what as completion rates lag, so that more than raises the cost of education per graduate. determines primary school enrollment. In rural Ghana, access to primary Keeping children in reduce the observed dropout rate from schools is much higher than the observed . 9.3% to 3.2%, a two-t1iirds reduction in enrollment rate. While 85% of house- schook the evidence the annual dropout rate. holds had a primary school within one- from Egypt The implication for policy is that half nile of their home, only 54% of . raising the quality of schlooling may be school-age children in the sample were In Egypt, school attendance is com- the most effective way of achieving lower enrolled in primary school. pulsory by law through the sixth rade. dropout rates in countries where school Access to higher levels of schooling However, in practice about 10% of pri- completion lags significantly behind en- is far more restricted: only 46% of mary school students drop out each year, roliment rates. households had a middle school in the so that fewer than 60% of students who community, whle the rest had to travel at enroll in first grade actually complete Summar least 5 miles to reach such a school. primary school. Further, only 8% of households had a A second World Bank study--using Raising primary sci iool enrollment in secondary school in their community and the Egyptian Retention Study-explores low-income developing countries is an the average distance to the nearest school the reasons that children drop out of important policy goal. HowNever, achiev- was 15 miles. school. The sample of 1,800 clildren ing this goal will require more than Since 95% of students in Ghana at- was drawn from the population of pri- simply improving access to primary tend free public schools, the mamin dcir mary school students in the 1978/79 schools. Because access to post-primary costs of schooling are the time and cost academic year and followed these stu- education is an importa it detenninant of of travel. The results show that the costs dents-including the 9.3% who dropped primary school enrolliment, it is equally of post-primary schooling, represented as out-through to 1979/80. important to expand anid improve access the distance needed to travel to middle In evaluating the determinants of to middle and secondarv schools. and secondary schools, are among the dropping out, the study takes into ac- In regions where eii rollrnent rates are strongest determinants of primary school count each student's ability, fiumily already acceptable, the focus of policy enrollment and completion. Further, the background, earmings opportunities, and needs to be on improving school quality. costs of post-primary school can offset school quality. The quality of available schooling is the any positive efiects on enrollment which A simple set of conclusions stand single most important factor detemiining improved access to prinmary education out. First, higher sked children-those prnmary school completibon. might have, who perfor,med well on achievement e E \ 1 ,)4 Sdi4w1% The study suggests that even when tests-tend to be the ones who stay in Adi,. iMA pvo 5 Ds the focus of policy is expanding phmary school. But holding ability constant, a kzt'W a Nl7 'IkXki3k Wi school enrollment and completion, it is at student attending a lower quality school least as important to improve access to is more likely to drop out and complete vIirRy shw Iiis-,, c fi*t middle and secondary schools as it is Wiewer grades . " WaBr Wdi ,36ni access to primry schools. In Egypt, bringing the quality of all schools up to the level of the best would Poverty Lines is prepared by Gillette Hall with assistance from Jim Shafer. This issue was edited by Brian Culhane and Cynthia Orlando. For information orfor a subscription, contact: Visit our World Wide Web siies: LSMS Administrator (Attn: Poverty Lines) PSPPovertyHome Page: http://hco.worldbank.org/html/heovp/povcrty/contents.html World Bank, N8-058 LSMS Home Page: http://www.worldbank.org/htnl/prdph/lsms/1slsslsome.htmliI 1818 H Street, N.W. Washington, D.C. 20433 Fax: (202) 522-1153 Internet: Isms@worldbank.org