Achieving and sustaining universal primary education : international experience relevant to India

Policy, Planning, and Research WORKING PAPERS Education and Employment Population and Human Resources Department The World Bank March 1989 WPS 166 Achieving and Sustaining Universal Primary Education: International Experience Relevant to India Nat J. Colletta and Margaret Sutton Funding, local accountability, relevant curricula and instruc- tional materials, emphasizing basic knowledge and reasoning skills, good teacher training, and upgrading poorer schools are all important. But promoting female enrollment may also require more organized day care- to free young girls from child care duties. The Policy, Plauning, and Research Complex distnbtnes PPR Working Papers to disseminate tht, findings of womk in progress and to encourage the exchange of ideas among Bank staff and al other interested in development issues. These papers carry the names of the authors, reflect only their views, and should be used and cited accordingly. The findings, interpretations, and conclusies are the authors' own. They should not be attributed to the World Bank, its Board of Directors, its management, or any of its member countnes.  Polac nning, and Research I Educadon and Employment To achieve its goal of full enrollment of children with state and central management) as well as aged 6 to 11 by 1990, and children aged 6 to 14 community partiJ ipation promises the best ad- by 1995, India must increase its enrollment from ministrative oversight of schools. less than 80 percent to more than 100 percent in 10 years. Sustaining universal primary educa- * Improve quality in the poorer schools and tion also means reducing the high dropout rate states to increase participation and retention (over 50 percent by grade 5) to near zero for rates in the primary schools. This requires well- eight grades. trained teachers - motivated both by better pay and in-service training (which the introduction To reach its targets, India must: of educational technology will faciLitate). Back up policy with funding - in particu- * Make curricula, instructional materials, and lar directing central govemment funds to areas exams relevant to local students (emphasizing and states with insufficient resources to improve basic knowledge and reasoning skills over facilities, quality of instruction, and local academic knowledge). language resources. * Make a special effort to provide organized - Encourage local accountability of teachers day care to release young girls from their child- and administrators to communities. Encourag- care care responsibilities. ing good local school management (coordinated This paper is a product of the Population, Human Resources, Urban and Water Operations Division, Asia Country Department IV. Copies are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact Mary Philiph, room D10-017, extension 75366. The PPR Working Paper Series disseminates the fndings of work under way in the Bank's Policy, Planning, and Research Complex. An objective of the series is to get these findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official policy of the Bank. Produced at the PPR Dissemination Center  IDA Achi.yvig and Sustaining Universal PrmaXry EAucAtion: A Studv of International Exoerience Relevant to India Table of Contents Executive Summary . . . . . . . . . . . . . . . . . . . L I. Study Purpose and Methodology . . . . . . . . . . . . 1 II. Background - Universal Primary Education in India . 3 III. World-Wide Trends in Achieving and Sustaining UPE . 5 IV. Select International Case Experiences . . . . . . . . 7 A. Kenya. 7 B. Tanzania .11 C. Indonesia .14 D. China .17 E. Sri Lanka .20 V. Analysis of Lessons From Successful International Experience .... . . . . . . . . . . . . . . . . . 24 VI. Implications for India .... . . . . . . . . . . . . 28 ANNEXES: 1. Worldwide Trends in Attaining and Sustaining UPE 2. School Enrollments in Kenya 3. School Enrollments in Tanzania 4. School Enrollments in Indonesia 5. School Enrollments in China 6. School Enrollments in Sri Lanka SELECTED BIBLIOGRAPHY This paper was prepared by Nat J. Colletta (Senior Educator), India Population and Human Resources Development Division, The World Bank, and Margaret Sutton (Consultant).  EXECUTIVE SUMMARY A,chieving and Sugtaining Universal Primary Education A Study of International ExIerience Relevant to India Background India's 1985 White Paper "Challenge to Education" advocates full enrollment of children aged 6-11 by 1990, and of those 6-14 by 1995. These goals mean increasing enrollment from less than 80% to more than 100% in 10 years, and reducing the currently very high dropout rate of over 50% by grade 5, vo near-zero for 8 grades. Although the Central Government is providing instructional inputs under "Operation Blackboard" the dependence on States to mobilize and finance school construction has met limited success due to the uneven nature of state revenue generating and community organizational cepacity. In order to reach the current goal of UPE by 1990, the Government of India will need to take strong and effective steps. Simply put, there must be an adequate supply of school places to serve all children, and there must be sufficient parent and child demand for primary education to keep children in school. On closer examination, both of these issues are quite complex. The purpose of this study is to identify the specific problems related to the supply of and demand for primary education in India, and to show from the experiences of other countries that have recently or rapidly attained UPE how these problems have been addressed. The lessons derived from studying international e2x zience in rapidlv achieving and sustaining UPE are: 1. While political will is necessary to attaining UPE, it is not sufficient unless backed by organizational capability and fiscal commitments to primary education. Reaching the last 10%-15% of students is relatively difficult and vostly. 2. Good micro-planning of school locations and instructional resources is necessary to provide equitable access to primary education. 3. With guidance and standard-setting, local communities can offset initial capital costs of schools through contributions of labor and material. 4. Increased community participation and professional school management provide the accountability necessary to raise school quality and increase the leve'l of cohort retention. - i -  5. Reaching and sustaining UPE requires quality upgrading in impoverished schools. Competent and motivated teachers are crucial to such quality improvements; professional support through in-service training erhances teacher quality, which in turn enhances participation and retention. Educational technology provides a rapid means to upgrade A.nd professionalize teachers at low cost. 6. Curricula and exams that emphasize academic over basic knowledge have low relevance to children who will not go on to higher educacion; low relevance, like low quality, adversely affects participation and retention rates especiLlly when the opportunity costs of child labor are high. Steps to address opportunity costs, particularly of female labor, such as the provision of day care for young children can affect increased female enrollment. The implications of these lessons for India's drive for UPE are: 1. For India to achieve her targets for UPE, resource allocation will have to reflect the priority given to primary education in policy statements. Additional support for UPE from the Central Government should be focused on areas and States with insufficient resources to meet special needs for facilities, quality improvements, and local language resources. 2. Encouraging local accountability of teachers and tdministrators to communities will enhance efforts for quality improvement. Good local school management through community participation coordinated with State and Central management promises the best administrative oversight of schools. 3. India needs to better target quality improvements of poorer schools and States in order to increase participation and retention rates in primary schools. Well-trained teachers are essential to improving school quality. When teachers cannot be motivated through wage increases, enhancing professionalism through in-service training can provide both quality improvements and motivational incentives to teachers. This can be done in a rapid, cost effective manner through the introduction of educational technology. 4. Promoting female enrollment in India will necessitate a special effort to address opportunity costs of young girls as "care givers." This may mean a greater emphasis on the provision of organized day care. ii  INDIA Achigving and Sustaining Universal Primary Education: A Study of International Exnerience Relevant to India I. Studv Puroose and Methodology Purpose of the Study 1. In order to reach the current goal of UPE by 1990, the Government of India will need to take strong and effective steps. Simply put, there must be an adequate supply of school places to serve all children, and there must be sufficient parent and child demand for primary education to keep children in school. On closer examination, both of these issues are quite complex. The purpose of this study is to identify the specific problems related to the supply of and demand for primary education in India, and to show from the experiences of other countries that have recently or rapidly attained UPE how these problems have been addressed. Studv Methodolorv 2. The background section of the paper reviews the current situation in India with regard to the goals, targets and constraints on realizing UPE. The next section provides an overview of international trends in achieving and sustaining UPE. 3. In order to better understand the international experience in moving towards UPE, the study then analyzes five select countries which have made rapid progress towards attaining and sustaining UPE. The study examines the experiences of these five countries and derives lessons relevant to India from them. All five case countries have high primary enrollment ratios and high retention rates. Three of the five case countries were selected on the grounds of recent quantitative success in attaining and sustaining UPE. Kenya, Tanzania, and Indonesia all fall into the first pattern of progress towards UPE. They have gross enrollment ratios higher than 90%, fifth grade cohort retention rates of 60% or more and primary enrollments have grown at an annual rate of 5% or higher, from 1960-1980. Because their progress towards UPE is recent, it has been attained in a global situatior. similar to that which exists for India today. Because their progress towards UPE has been rapid, they offer lessons in how UPE can be attained in a short time span, and what problems such a development poses. 4. Sri Lanka and China were chosen as cases on the grounds of special similarities to India. Both countries show high enrollment and retention rates like the first three cases, but have not grown at such a rapid rate in the same time period. Sri Lanka was selected for its sociocultu'ral similarity to India. In addition, Sri Lanka approached UPE nearly three decades ago, and has since worked on the problems of oustaining and reinforcing UPE. Sri Lanka's experience gives particular insight into the problems of sustaining UPE. The People's Republic of China was selected because it has approached provision of UPE for the largest, and among the poorest, child population in the world. Table 1 summarizes data on primary education for the five case countries, India, the developing countries, and industrialized countries.  2 Table 1 i tnioI_ total 1' A*96s *tmn:b m,epsab Guowth Inroltant GIrls rol1 Rate Rate Rate Ica" ,702,414 48 94 61X 5X 8.41 Tw15is 3.493469 49 104 64 1S 10.5X I10-m-to 26,567,686 48 115 in 91 AS CilSd IMM02.OO 45 96 n.f0 n.e. n.s. let Laft 2,13.522 48 103 951 93 0.O Irmo 116,12U.000 39 01 43S s.. Son Cw Ies 461,413,.000 4 n.o. n.l. Xe *1 Csmtulm W,248.000 49 1 n.la. n.e. 1 a. UNESCO fLUitea, latest ysa aVatlabl-. 1964 or 1985 for aatVL*eao 19U3 world f4uwe. b. X of bozt eteruas araa 5a e.1950, cal-ulted in Komena (1957). S. Following a detailed account of the policy, program and financial interventions made in the five select cases, lessons are analyzed from this International experience. Finally, these lessons are applied to the Indian context indicating the major implications.  3 II. Backeround - Universal Primary Education in India 6. In the Seventh Five Year Plan (1985-1990), the Government of India has reaffirmed its commitment to attaining Universal Primary Education (UPE). The legal basis for UPE was established in India's constitution of 1949, which promises free and compulsory education for all children up to the age of 14 by 1960. In 1960, along with other .'sian nations at the Karachi conference, India set 1980 as a target date for attaining UPE. In the early 1970s, the Planning Commission set out a program to r' ch UPE by the end of the Sixth Plan period in 1984. In 1986, the Pa.liament tpproved the New Education Policy, a response to the Challenge to Education docunent drafted the previous year. The Action Plan for the New Education Policy aims at enrolling all children aged 6-11 by 1990, and all 6-14 year olds by 1995. Gross enrollment in primary education in 1987 represented 97% of the appropriate age group, up from 79% in 1980. However, India's efforts to sustain UPE are being undermined by high dropout rates. In 1987, it is estimated that only 53% of the cohort had reached the fifth grade. India has yet to attain the high and sustained primary participation rates achieved in a number of other developing countries. 7. Overall expenditures on education in India grew in current prices from Ri- 13,880 million in 1973/74 to Rs 97,440 million in 1986/87 (budgeted). In constant prices, education spending increased by 7.6% per annum during this period. Elementary education (primary and middle school) expenditures grew more rapidly, at an 8.5% annual rate, with the result that the share of elementary education in total education expenditutre rose from 36% in .973/74 to 40% in 1986/87. 8. For India to realize its current goal of attaining and sustaining UPE by 1990, greater expenditure on primary education is almost certainly necessary. Where there are no school facilities, these must be provided; where facilities are inadequate, they will need renovation and equipment. Skilled educational administrators will be required to manage and monitor the system. Teacher salaries constitute the major portion of recurrent expenditures on primary education, so an increase in the size of the teaching force has long- term cost implications. Textbooks and other student materials, whether provided free of charge or not, require developing, printing and distribution. Curricular reforms and teacher upgrading, common concommitants to the drive to UPE, also draw on resources. 9. Equally important to the mobilization of resources for primary education is their judicious allocation to the areas of greatest need. Disaggregated enrollment data show several forms of inequity in primary participation which will need to be overcome if India is to enroll all of her children in primary school. There is a pervasive urban/rural difference, reflected in che lower literacy rates of rural than urban children: in 1981, 78% of urban children aged 10-15 were literate, compared to 50% in rural areas (Subbarao, 1987). 10. There are also strong caste and gender differences in school participation rates in India. The 1980/81 economic survey of Maharashtra State (reported in Subbarao) showed only 45% participation among "backward class"  4 children in rural areas, compared to 58% for children of other castes. In urban areas, 69% of backward class children were enrolled, compared to 81% of the other castes. Within each of these populations, girls' participation rates are lower than boys'. The differences are especially marked among rural children. In the backward castes, the participation rate for rural girls was 33% compared to 54% for boys; among other castes in rural areas, 65% of the boys and only 51% of girls were enrolled in schcol. The gender difference in participation declines somewhat in urban areas, but still is a cause for concern. Enrollment rates for boys in urban areas were 72% in the backwards classes and 84% for other castes, whila for girls they were 64% and 78%, respectively. 11. Household income is another determinant of school participation. In Maharashtra, the 1980 primary participation rate for children from households spending less than Rs 30/month was 33%, compared to 58% for households spending Rs 51-70/month, and 93% for those spending Rs 151 or more. Gender and caste biases exist at all household expenditure levels, as does the urban/rural difference in participation rates. 12. The persistent differences in primary participation rates between urban and rural areas, boys and girls, high and low income families, and backward communities as compared to other castes, all reinforce tLe point that attaining and sustaining UPE in India is not a matter of simply expanding the primary school system. Adequate school places in appropriace locations with teachers available to conduct classes constitute the necessarv, but not sufficient, conditions for attaining and sustaining UPE. Just as important is designing specific programs and policies that meet the needs of the groups currently marginal to schooling, for until these groups are drawn into the educational system. UPE will not be attained. And, as this study will suggest, until educational services are of sufficiently high quality and relevance, it is likely that high dropout rates will continue to undermine sustaining UPE. 13. Recognizing the need to eoualize the quality of education offered by resource-poor schools and States, and to emphasize enrolling girls in primary schools, India launched "Operation Blackboard" in 1985. This program aims to provide a minimum package of educational materials to all primary schools, to upgrade the quality of single-teacher schools by adding a second teacher, and to recruit more women as primary school teachers. It is not yet clear how well this program is working, but two short-comings are obvious. First, Operation Blackboard requires states to provide classrooms meeting certain specifications, and many states do not have the resources to do so. Second, the extremely complex implementation guidelines promi;e bottlenecks and delays that have probably already impeded its effectiveness. 14. Before turning to examine the experiences of select other countries which have attained and sustained UPE, and the lessons particularly relevant to India, a broader perspective is provided in the following chapter.  S III. World-Wide Trends in Achieving and Su5stainia UPE 15. It is no longer necessary, in 1988, to argue for the desirablity of UPE. Universal primary eaucation has taken on the character of a normative imperative (Boli et, al.. 1986). The social and economic benefits of primary education have been demonstrated by the generally larger rates of return tn primary than to higher levels of education (Psacharopoulos, 1981; 1985); by the relationship between literacy and productivity (Colclough, 1980; Lockheed It, &1, 1980); between mother's education and infant mortality, child health and lowered birth rates (Cochrane, 1979; 1980); and by the connection between primary education and nation-building. It is estimated that, in 1870, 24% of the world's children were enrolled in some form of primary education, and that this figure grew only to 35% in the five years preceding World War II. In the latter period, only Canada and the United States showed primary enrollment rates exceeding 90% (Benavot & Riddle n.d.). Since the end of World War II, primary enrollment has increased tremendously, with the world rate estimated at 91% in 1980. 16. India, along with other Asian nations, set 1980 as the target date for attaining UPE at the Karachi Conference of 1960. A similar goal was set by African nations at Addis Ababa in 1961, and by Arab states at Tripoli in the same year, while the Santiago Conference in 1962 called for Latin American nations, with relatively more developed educational systems, to attain UPE by 1970. Considerable progress has been made towards UPE since 1960; among 104 Asian, African and Latin American countries, 78 reported gross enrollment ratios of 90% or more in 1980.1 For developed countries, on average the gross primary enrollment rate was 107%, for developing countries, 86%, and for India, 783%2 17. The move towards UPE is thus clearly a world-wide phenomenon, but the methods and experiences of countries have varied according to local social, economic, political, and cultural conditions. On closer examination, the developing nations can be divided into five categories, in regard to their progress towards attaining and sustaining UPE (see - First, there are 10 countries that have made recent and rapid progress towards UPE, and have relatively high rates of cohort retention, evidence of sustained progress towards universal 1 The gross enrollment ratio represents the ratio of children of primary school age to children enrolled in school. When this ratio is 90% or higher, then the educational system is approaching the caDacity to offer UPE. Because of under- and over-age students, errors in estimating the size of the child population, and sometimes, exclusion of handicapped children from the child population estimate, a high gross enrollment ratio does not in itself indicate success in attaining UPE. 2 Figures are UNESCO's Gross Enrollment Ratios (National Structures), UNESCO, 1983. As noted, India's gross enrollment ratio had increased to 97% in 1987.  6 primary.3 In Annex 1, these have been further divided into countries with primary school populations above and below two million. - Second, there are 24 countries with high enrollment and retention, but lower growtb rates. These are countries that either expanded their primary education system prior to 1960, or grew at a slower rate.4 - The third category, containing 34 nations, are those with low gross enrollment but high growth eates.5 - Fourth, 27 countries showed high initial enrollment for primary education, but low retention rates, indicating a problem in sustaining UPE.6 - Finally, India in 1980 was among the 11 nations which showed both low gross enrollment and low growth.? 18. While the general statistics point to those countries wit' eatest success in achieving UPE, a more detailed look behind the numbers s equired to understand the unique policy and programmatic interventions which produced such results. 3 Gross Enrollment Ratio 89%; Retention Rate (% of cohort reaching Grade 5) 59%; Repetition Rate 10%; Annual Growth Rate 1960-80 5%. 4 Gross Enrollment (1980) 89%; Retention Rate 59%; Annual Growth Rate (1960- 80) 5%. The Growth Rate can be lower because: (a) the primary education system had a relatively high capacity already in 1960; and (b) birth rate, and thus the size of the primary school cohort, has been low. ' Gross Enrollment Ratio (1980) 90%; Annual Growth Rate (1960-80) 4.9%. These are nations that had extremely low capacity for primary education in 1960, and often, high birth rates. 6 Gross Enrollment Ratio (1980) 80.? Retention Rate 60%. ' Gross Enrollment Ratio (1980) 90%; Growth Rate (1960-80) 5%. Note that high retention with low gross enrollment indicates simply that the few who enter primary school, tend to remain in the system.  7 IV. Select International CAse E,Reriences 19. For each of the five countries there is a brief description of the growth of primary education, and special steps taken to move towards UPE. There are fou- issues common to attaining and sustaining UPE in developing countries today: A. In all cases, an initial capital outlay for school construction is required. B. By definition, as each country moves towards UPE, social groups previously outside of or underrepresented in the school system must be provided educational services. C. The underre;?resented always include the poor, which raises the issue of ofisetting rather high opportunity costs for child labor. D. Progress towards UPE also raises the inevitable trade-off between providing mass education - - similar services to all --and providing speeialized academic education leading to the development of highly trained manpower. The case countries have all dealt somewhat differently with these four issues. Each case study therefore highlights two or three policy or programuatic steps that helped move the country towards UPE by addressing some of these basic educational issues. Ah. Kenya 20. In the twenty years following Independence in 1963, the number of students in Kenya's primary schools increased nearly five fold, from less than 900,000 to almost 4,500,000. This in=rease reflects both the very high rate of population growth in Kenya reaching over 4% per annum during the period, and an extraordinary quantitative development of the primary education system. In 1963, the primary gross enrollment ratio was around 508, whereas in 1983, it stood at 100%, indicating a capacity to provide primary education to all of Kenya's children. As enrollment grew, the proportion of girl students also increased, from about 358 in 1963, to 48% in 1983. Annex 2 summarizes the increases in primary students and teachers, as well as gross enrollment ratios for secondary and tertiary education in Kenya. 21. In 1985, the primary cycle was extended from seven to eight years, which accounts for the drop in the gross enrollment ratio in that year. Students may enter Standard 1 at age 6. The primary school completion rate has been relatively high in Kenya. It was estimated at 74% in 1977, although there are indications that the rate may be falling with expanded intake in the lower grades. While the gross enrollment ratio now stands at over 90% for the 8 years of primary schooling, Kenya still faces difficulties in attaining and sustaining UPE.  8 22. For the Kenyan Government, educational costs have mushroomed beyond a sustainable level. Government spending on education exceeded 6% of the GNP in 1977. Reforms throughout the 1970s and 1980s, which will be discussed below, have shifted an ever greater proportion of the financial burden for primary schooling onto the Government. Despite its rather poor record in addressing the financial requirements of Universal Primary Education, however, the Kenyan Government has undertaken important policy measures to support UPE. In particular, Kenya reformed its primary school-leaving exam, the CPE, in the mid- 1970s. This reform aimed to increase both the quality and relevance of primary education. The CPE reform will be desc.ibed in greater detail below, after a consideration of how Kenya's current problems in financing education came about. Private and Local Contributions to Primary Education 23. Analysts of Kenyan education generally agree that the massive expansion of both primary and secondary education in Kenya has been largely demand-driven. Beginning in the colonial period and continuing into the contemporary Republic, education has featured as the single most important source of upward socio- economic mobility in Kenyan society (Sheffield, 1973; Court, 1979). Actual enrollment in primary ar,d secondary schools consistently exceeded government projections in the 1960s and ;970s, although in the poorer regions of Kenya, education is still not sufficiently attractive to offset the opportunity costs of schooling. This is reflected in the low participation rates in the arid region with its pastoral economy. For these children, the Government has provided boarding schools, which, however, do not appear to be attracting and retaining the children at whom they are aimed (Nkinyangi, 1982). 24. Kenyan parents and local communities have historically borne a high share of the costs of their children's education. The impetus for community provision of education began in 1929 through the founding of the Kikuyu Independent Schools Association, with the purpose of providing education to Africans without compromising their cultural values. Community initiative in education, particularly in the construction of facilities, has continued to be important after Independence, and is recognized in the concept of "Harambee" or "pulling together." In 1980, 700 out of the 1,200 secondary schools in Kenya were harambee schools; it is estimated in addition that Harambee or community contributions to primary education totalled over KSh 700,000 in 1979, and more than KSh 1,600,000 in 1981, compared to recurrent government expenditures of KSh 2,200,000 at the primary level Li 1986 (Olembo, 1986). Clearly local and private contributions have been significant to educational expansion in Kenya. 25. The demand for education and willingness of individuals and communities to act on it has proved a mixed blessing to the Kenyan Government. While private and local contributions have lessened costs of establishing schools, the Central Government has borne the major share of recurrent costs, particularly of teachers' salaries. As enrollment at primary levels outstripped government projections in the 1960s and 1970s, central government expenditure for primary education grew out of proportion to the Government's capacity to pay. Over the past 15 years, the Government of Kenya has devoted more than 20% of its current expenditure to education. The proportion of current education expenditure on the primary level rose from 54% in 1965 to 65% in 1975, and has remained above 60%  9 in the 1980s. The proportion of educational expenditure on the secondary level has correspondingly declined, from 30% of the current budget in 1965 to 18% in 1982. The high rate of population growth continues to compound the expense of providing primary education to all of Kenya's children in the future. 26. Budgetary implications notwithstanding, the Kenyan Government has made several moves towards making primary education free of private cost. In 1974, school fees were banned for the first four grades of primary, which accounts for the large jump in enrollment in that year. By 1980, fees for all seven years of primary were eliminated, and in 1985 Grade 8 was shifted to the primary level, and also provided without fees. Parents still bore a high financial burden, however, in the form of contributions to school building and equipment funds. The Government estimated that parental contributions to these funds averaged KSh 30 per student in 1976; .tther sources estimate that the private costs of primary schooling quadrupled after the abolition of school fees. A World Bank study estimated that in 1980 the per student cost borne by the Government averaged KSh 400 (US$53), while parents contributed around KSh 20 (Wolff, 1984). 27. The Government has taken further steps to reduce the private cost of primary education. In 1980, schools were forbidden to levy any fees directly on parents; contributions could come only from harambee or community projects. Wealthy individuals also provide harambee contributions to primary and secondary schools. 28. De facto dependence on private and local contributions has relieved the Government of some costs. It has also exacerbated two of the continuing problems in attaining and sustaining UPE in Kinya: equity in access and the quality of the education provided. The nation-wide primary gross enrollment ratio reached 108% in 1979, but as the system has expanded, observers here noted a general deterioration in the quality of education provided. Low quality of primary education is particularly acute in poorer schools and districts, which further detracts from equity in educational opportunity. Exam Reform 29. The high private cost of schooling is one major barrier to educational participation; selective examinations at different points in the schooling cycle constitute another. In Kenya, the Certificate of Primary Education (CPE) exam served until 1983 to select the 35%-40% of children who are eligible to pursue secondary schooling. With the structural reform to 8 years primary in 1984, the CPE was replaced by an exam more appropriate to eighth-graders, based on principles similar to the reformed CPE. 30. As in any system where access to the next level of education is controlled by examination, the form and content of the CPE exert a powerful influence on the characteristics of primary education, particularly in its last two or three years. By the early 1970s, it had become clear to Kenyan educators, politicians, and parents that the CPE enhanced two of the major problems facing primary education. First, the exam was seen as culturally biased, favoring the knowledge and skills associated with urban, middle class life over those of rural and poor children. A second, related bias was towards academic over practical knowledge. Since a relatively small proportion of children would be  10 going on to secondary schools, the exam distorted t.e content of upper-primary education, engendering an academic orientation where general and vocational skills and knowledge should more appropriately be emphasized. Beginning in 1973, the Kenyan Government undertook major reforms of the CPE, aimed at addressing these two issues. 31. The major thrust of CPE reform has been to reduce the number of items in the three multiple-choice papers -- English, mathematics and general -- which test factual recall and secondary-level knowledge, and replace them with items testing reasoning, and drawing on rural experience. Examination reform is an on- going process, based on evaluation and analysis of exam results. The Kenya National Examination Council, founded in 1980, continues to operate actively towards the goal of making the primary school-leaving exam an effective tool for fostering relevant and high quality education (Somerste, 1988). 32. Along with its influence on the content of education, the CPE relorm has included a system for providing feedback on performance by district and school, thus pointing to specific needs for quality improvement. Some encouraging results have ensued. Although performance differences between urban and rural and richer and poorer schools increased in the first years following reform, by 1979, some rural schools and districts were ghowing higher performance and less disparity with better-endowed sys..ems. The four lowest-scoring districts showed substantial improvement in exam outcomes, and the deviation in mean scores in rural areas dropped to a lower level than the first year of reform. These increases have been attributed to effective management strategies of employing diagnostic feedback for quality improvements. In the poorer districts showing greatest improvement, teachers were offered seminars and workshops on the new educational demands on their students, and low-scoring schools were given special attention (Makau & Somerset, 1980). 33. The feedback process brought out that school quality is even more important to success on reasoning items than on factual recall. That is, while the content of the examination has become less biased towards academic information, the disparity in performance between students attending poorer and richer schools increased. Exam reform in itself is no guarantee of equity in education; only when the quality differences between schools are reduced can the educational system hope to provide equal educational opportunity. Lessons from Kenya 34. The harambee schools and contributions to education in Kenya demonstrate how effoctive community action can be in providing initial capital for school facilities as well as for subsidizing their operation. It is very difficult to determine total harambee contributions to primary education in Kenya, but they have clearly played a major role in off-setting capital expenditures for the expansion of primary education. 35. Although it has added to the fiscal burden on the central government, Kenya took the step of banning parental contributions to primary education. Enrollment leaps in the years that school fees and private parental contributions were banned suggest that these expenses are indeed a barrier to school participation. So far, it seems, the only steps that the Kenyan government has  11 taken to reduce its fiscal burden for primary education is to encourage jurther community support for schools. 36. Kenya's experience with examination reform provides three important lessons concerning the relationship between examinations and education. First, if primary education is intended to provide a basic set of knowledge and skills to all students, then the terminal examinations must reflect that educational decision. Examination performance is a major influence on pedagogical practice. Teachers and administrators will direct their energies towards examination results, so those exams must themselves test for the knowledge and skills that are considered essential for all students, rather than those thought to be related to academic achievement at higher levels of schooling. Second, a well- designed examination backed by performance evaluation can provide an effective tool for quality improvements in individual districts, schools, and classrooms. With a properly designed and analyzed examination system, districts, schools, and teachers receive meaningful feedback on their educational successes and failures. Thirdly, the use of distance teaching techniques such as radio forums and correspondence provided a cost effective means of improving pedagogy in connection with feedback from examination performance to teachers. B. Tanzania 37. In the first 15 years following Tanzania's independence in 1961, primary enrollments grew steadily at a rate of around 6% per year. Assuming steady growth at this rate, the Second Four Year Plan of 1969 projected that UPE would be attained in 1989. Instead, government leaders decided at the Musoma conference in 1974 that UPE would be attained in the space of three years, that is, by 1977. Annex 3 summarizes educational enrollment in Tanzania during the period 1960- 1984. 38. The two large jumps in primary enrollment, from 1974 to 1975 and from 1976 to 1977, correspond to the announcement and implementation of UPE according to the Musoma declaration. Declining enrollment ratios from 1979 to 1984, however, indicate some problems in sustaining the impetus towards UPE. Besides school enrollments, another impressive quantitative leap is in the number of primary teachers, which doubled from 1976 to 1978. A large proportion of this increase in the primary teaching force resulted from an innovative in-service teacher training program, which will be discussed below, after considering the overall picture of Tanzania's progress towards UPE. Primary ExDansion for Eguity 39. A comparison of school enrollment figures for Tanzania and Kenya points to one of the major differences between the two countries' educational policies. In Kenya, secondary school enrollments expanded alongside primary, with the secondary school gross enrollment ratio growing from 4% in 1965 to 20% in 1984. In the same period in Tanzania, the secondary gross enrollment ratio rose only from 2% to 4%, before dropping back to 3%. The restriction on growth at the secondary level is one of the most tangible expressions within the education sector of Tanzania's policy of putting equity goals first. The other major expression of equity in Tanzanian education has been the emphasis on adult basic education. As a result of campaigns and on-going adult education programs, adult  12 literacy in Tanzania increased from 31% in 1967 to 85% in 1983. The literacy campaigns have had beneficial effects for primary education. Many of those mobilized as literacy teachers later became school teachers. When the UPE goal was announced, there was already a social infrastructure emanating from the literacy campaigns for community action in education. This infrastructure was effectively utilized for delivering in-service teacher education as well as mobilizing community self-help for school construction. 40. Tanzania's primary education system has expanded at relatively low costs, to Government and to parents. By 1969, the Government had banned school fees and private primary schools and instituted a free textbook policy, so that virtually all recurrent costs of primary education are borne by the Government. Per student government expenditure was estimated at US$34 in 1981, compared to a mean of $71.5 and median of $47 for all of East Africa. Parents contribute T Sh 20 (US$2.2) per child, and supply uniforms (Wolff). Like Kenya, Tanzania has depended heavily on community effort in the construction of school facilities. 41. The low unit costs of primary education in Tanzania undoubtedly rest on low teacher salaries, which were estimated at $851 in 1981, compared to an average of $1,859 for Eastern Africa. Since salaries consume a relatively small portion of its budget for primary education, only 58% in 1979, compared to 85% or more in virtually all developing countries, Tanzania is also able to devote a greater share of expenditure on primary education to teaching materials, 29% as compared to an average of 3.5% for Eastern Africa. Tanzania devotes a lower, but not insubstantial, portion of its recurrent expenditure on education to primary education than does Kenya. In 1983, 47% of recurrent education expenditure went to primary education, and 30% to secondary. 42. Despite the strides that Tanzania has taken towards equalizing educational opportunity, the low levels of expenditure seem to be taking their toll not only on the quality of primary education, but also on the sustainability of UPE, as reflected in the decline in gross enrollment ratio from 100% in 1979 to 83% in 1984, and decline in absolute numbers from 1983-84. Some of this is certainly related to Tanzania's political situation in the early 19.80s, but educational observers also point to low quality and relevance of primary education to rural life as a constraint on sustaining UPE. In-service Teacher Training 43. One of the most innovative and successful programs that Tanzania has undertaken in support of UPE has been the Distance Education program for in- service teacher training. With very small numbers of secondary school graduates, Tanzania faced major constraints in teacher supply when the UPE drive began in 1974. To meet the quantitative need for primary teachers, the Government offered a special in-service training program using distance-education techniques, to approximately 45,000 primary school graduates, who were admitted in three batches Zrom 1976 to 1979. Altogether, some 37,000 teachers had been trained and recruited through the distance education program by 1981. 44. Under the in-service training program, selected primary school graduates taught under the supervision of experienced teachers for a period of three years. The program stipulated that the trainees should teach 15 periods per week,  13 although studies of the program showed that many carried a much heavier load. The trainees attended seminars taught by 2,400 tutors for 3-4 hours a week, followed radio instruction and correspondence programs, and completed their basic training in a 6-week course at an established teacher training program. 45. The distance-education program was highly effective in providing primary teachers at low cost. Whereas in traditional teacher-training colleges, net annual costs per student average T Sh 27,334 (approximately US$3,000), for the distance-education program these costs were T Sh7,201 (less than US $800), not including the savings represented by trainees' teaching services. 46. Moreover, eveluations of these teachers' performance in the schools compare favorably to the quality of pre-service trained teachers. Two studies of distance-education and pre-service trained teachers show no appreciable difference in subject-matter knowledge, and some pedagogical superiority among those trained in the in-service program (Carr-Hill 1984). In particular, these teachers were found to adapt lessons more to the local environment, and to be more supportive of students than their traditionally trained colleagues. It should also be noted, however, that both groups of teachers showed need of improvement in teaching skills, and most expressed a desire to attain greater mastery of the subjects they were teaching. Thus, the studies recommended expanding the in-service program for knowledge and skills upgrading. 47. Good organization was central to the success of the distance-education program. Building on prior and successful mass delivery systems for health and li,racy campaigns, the Government was able to identify teacher training candidates and deliver the training program to them. Tanzania's service delivery system rests on vertical coordination of local, district, and central governments. Another example of its use was for taking a census of the child population, at the beginning of the UPE drive. Teachers were asked to survey the child population in their school catchment areas, to report these figures to local officials, who compiled district data to send on to the central government. Lessons from Tanzania 48. Tan-ania's experience in attaining UPE, like Kenya's, demonstrates the effectiveness of community mobilization for provision of school facilities. What is distinctive about Tanzania's self-reliance efforts has been the on-going development of mass delivery systems for social services. Coordination between community, district, and central leadership, essentially through the national political party machinery, has enabled Tanzania to provide primary education and teacher training, as well as other social services, to the large majority of her population. 49. Building on the organizational network of the earlier mass literacy and health campaigns, Tanzania's experience with distance education for in-service teacher training also provides evidence that traditional pre-service training is neither necessarily better than in-service training, nor perhaps even as relevant to the realities of teaching under difficult conditions. As with primary education itself, one of the greatest short-comings of the program was seen to come form a shortage of materials. The major factors contributing to its success  14 seem to have been the employment of experienced teachers as trainers, and good admlnistration. C. Indonesia 50. Indonesia's primary education system grew steadily following Independence in 1949 although the rate of growth slowed in the early 1970s, with enrollment ratios nearing 80%. The final push towards achieving full primary enrollment came with the SD Inpres program initiated by President Suharto in 1974. Annex 4 summarizes the information on school enrollments in Indonesia. 51. Indonesia has nearly reached its goal of UPE for all children. It is estimated, however, that 5%-10% of the school-age population are currently not in school. Government efforts in the area of UPE are now focused on closing this non-schooling gap, and on improving the overall quality of education. SD Inpresa 52. The jump in international oil prices in 1972/73 provided a sudden increase in revenues to the Indonesian Government. A substantial portion of theue were diverted to primary education through the program SD Inpres (Presidential Order on Primary Schools). As a result, the development budget for -_ication under the Second Five Year Plan - Repelita II (1974-79) was 12 times higher than that of Repelita I. The major portion of Inpres funds went into school building construction and renovation. Three types of standard buildings were constructed: multi-story buildings in urban areas with land scarcity; 6- room, 6-grade schools in moderately populated areas, and 3-room, 3-teacher schools in sparsely populated areas. Over 10 years, Inpres allocated over Rp. 2 billion, or around U.S. 2 million, to school construction and renovation. By 1978, nearly 25,000 schools had been built, representing 30% of all secular schools (USAID, 1986). 53. The school construction program was preceded by a pilot project in provincial educational planning, conducted in East Java and West Sumatra. Through this program, school mapping techniques were developed. These techniques were then applied to the complex task of mapping school locations nationwide, for construction of Inpres schools. The provincial management project also raised overall levels of educational management. 54. An associated stimulus to primary enrollment was the abolition of primary school fees in 1977. The 10% jump in gross enrollment from 1977 to 1978 can be attributed to the combined effects of greater capacity for primary education from the construction of facilities and removal of fees as an obstacle. 55. One potential problem of rapid primary expansion that Indonesia did not face was the availability of teachers. Although teacher supply and retention were seen as a major educational issue in the early years following Independence, 8 SD Inpres: Presidential isntruction earmarking oil profits for primary school contruction and quality improvements.  15 the Inpres push in fact occurred at a time of teacher surplus. Salaries for primary teachers, which had previously been so low as to lead to a widespread practice of double-jobbing by teachers, were raised substantially in the early 1970s. Private teacher training schools grew up with little government oversight. Finally, a few years before Inpres was launched, a hiring freeze was placed on teachers. When Inpres made teaching positions available, there were ample numbers of candidates. In 1984/85, less than 10% of primary teachers were underqualified, a relatively low proportion for a country recently reaching UPE. There are indications now of a teacher surplus in Indonesia, which hcwever has not yet resolved the problem of regional imbalances in teacher supply. 56. As with Kenya, Indonesia may soon face some difficulties in continuing to finance its large primary school system. When the eevelopment and regular budgets for education are combined, 50% of the total education expenditure is on primary education. In 1981/82, only a fraction of the SD Inpres, or development, budget goes to meeting operating costs. However, there are some indications that this figure may be rising, and it is not certain how much will be allocated to SD Inpres in the next five-year development plan (USAID, 1986). Closing the Non-Schooling Gap 57. With the construction of facilities, hiring of teachers, and abolition of school fees, it is now estimated that 90%-95% of Indonesia's 7-12 year olds are in school, leaving 5%-10% still not receiving formal education. As in other developing countries, these are the handicapped, children in remote areas, or those whose labor is required to support the family. Because providing adequate numbers of accessible schools does not in itself guarantee that these children will be educated, Indonesia is now focusing on providing the special services required to meet the needs of those currently out of school. The PU Wajar, or universal compulsory education, program, housed within the Directorate General of Primary Education, monitors progress towards UPE and designs programs to reach out-of-school children. 58. For the estimated 300,000 handicapped children, Indonesia, like other countries nearing UPE, has been expanding special education services. In 1984, more than 550 special education schools and 74 special education programs in regular primaries served nearly 20,000, or over 6%, of Indonesia's handicapped children. Six teacher training schools graduate approximately 1,000 special education teachers a year. The current targets are to increase service to 30% of the handicapped by 1990, and 100% by 1995. At present, these goals seem overall optimistic. 59. The SD Inpres program has provided schools in many areas where thetq were none previously, thus incorporating an increasing number of students. Being an archipelagic nation with a wide diversity of cultures, however, the problem of reaching all children in remote areas remains. This is in part a problem of teacher distribution, which has not been successfully resolved. To compensate for the lack of sufficient teachers in remote areas, Indonesia has developed self-instructional material designed for small-group interaction.  16 60. The self-instructional program, known as SD PAKONG, grew out of Project IMPACT, a program developed at the SEAMEO research institute, INNOTECH. IMPACT/PAMONG was originally conceived as a low-cost alternative delivery system for primary education. In Indonesia, the potential for its use in isolated areas became evident. SD PAMONG allows one teacher to supervise a large number of students at different grade levels through the use of peer teaching. It thus provides a structure for multi-grade teaching in isolated areas. 61. The number of primary school dropouts in Indonesia has declined over the past decade, but remains non-negligible. From 1975 to 1984, the percentage of students reaching the fifth grade increased from 56% to 72%. However, a substantial number of children still drop out before completing six years of primary education. It is estimated, for example, that some 780,000 students dropped out of primary school between 1983/84 and 1984/85. These dropouts constitute one of the major barriers to sustaining UPE. 62. As in all of the countries considered here, the problem of primary school dropouts in Indonesia can be traced to two major sources: the opportunity costs of schooling for poor children, and low quality of education. Recognizing the link between quality and the retention of children in school, the SD Inpres program has recently placed major emphasis on quality improvements in the existing schools, including provision of texts, equipment and personnel. Additional efforts related to quality improvement include work on assessing educational achievement and its correlates, conducted in the research branch of the Ministry of Education. Writing, printing, and distribution of textbooks has been a major undertaking for the past ten years. Textbook development is the one area of Indonesian primary education that has depended on substantial external support, notably in the form of a World Bank education loan for textbooks. What is not yet present is a systematized approach to linking survey and assessment findings to educational planning. It may be that the Management Information System currently being constructed can provide the organized information base necessary for directing efforts at quality improvement most effectively. Lessons from Indonesia 63. It may seem that Indonesia's recent success in attaining and sustaining UPE are based on economic good fortune. Certainly, the sudden increase in oil prices provided a source of funds not available to many countries. It should be noted, however, that both an initial decision and an on-going commitment to UPE had to be made concerning the use of those funds. In fact, for a variety of reasons, Indonesia's oil revenues have not been as substantial as hoped, and the SD Inpres program has been increasingly financed from the Government's general revenue sources. 64. As Indonesia began to implement its UPE policy, the nature of the problems changed. The building construction program and abolition of fees made primary education available and accessible to the vast majority of Indonesian children. As enrollment ratios approached 90%, however, it became clear that physical and financial accessibility in themselves do not guarantee that all children will attain a primary education. Handicapped children and those in particularly remote regions require specially designed educational programs and servic s such as peer teaching and modular materials used by Project PAMONG.  17 Moreover, there remains a proportion of children who drop out before completing their primary education. 65. Indonesia has made some efforts to understand the social and educatidnal determinants of primary school retention and completion, and has concluded that the quality of education provided is the central educational factor affecting primary school retention. While these efforts could benefit from greater systematization, Indonesia has be6un to address the problem of improving quality for sustaining UPE by increasing critical inputs such as textbooks, teaching materials, in-service teacher training, and adequate facilities. D. China 66. In 1949, the People's Republic of China faced massive educational probl.ems. Around 85% of the adult population was illiterate; there were school places for fewer than 40% of the children. In the early years of the Republic, military spending absorbed much of the government budget, with less than 4% going to education, culture, and public health. Despite these difficulties, however, educational development took place. Local communities built and staffed schools for their children, on the principle of minban gongz, or management by the people with government assistance. Since independence, local responsibility with government guidance and assistance has continued to underlie the development of primary education in China. 67. Since the early 1970s, gross enrollment ratios at the primary level have exceeded 100%, reflecting the large number of adults engaged in some form of primary schooling. Estimated net enrollment ratios show that age-specific enrollment is also very high, exceeding 90% since the early 1970s. The decline in net enrollment following 1977 is thought to be connected to the introduction of the responsibility system on collectives, which raises the opportunity costs of schooling rural children. The uneven and still limited growth of female participation illustrates a particularly recalcitrant problem concerning the schooling of girls. The available data on educational enrollments in China are summarized in Annex 5. 68. Educational policy in China has shifted several times along with shifts in fundamental political orientation. Both within China and among Sinologists, the basic tension in Chinese social policy is often characterized as the trade- off between being red and being expert, where "red" stands for mass orientation, political consciousness and social equity, while the "expert' orientation is towards efficiency and productivity. In education, the "expert" orientation implies an emphasis on technical and higher education, and an academic emphasis at the lower levels, while "red" schooling emphasizes primary and adult education, and political education within the schools. At present, the Chinese Government is pursuing a "red and expert" policy line, with mass participation in basic education, and renewed emphasis on higher and technical education. One of the reasons that China is able to do this is the vast expansion of schooling that has already taken place. This will be examined in the next section.  18 UPE Through Local uontrol 69. Attaining UPE in China has meant above all developing education in the rural areas, where the majority of the population live and educational services have historically been weak. The vast educational needs and limited resources to meet them have led China to follow, except during the Cultural Revolution, a two-track educational system since 1949. Urban schools, inherited from the Nationalist period, have been for the most part supported by the central Government and developed to meet the need for highly trained manpower, whereas rural school development has been primarily the responsibility of local communities, with assistance and oversight from the central Government. The mission of the rural schools has been to impart basic literacy, numeracy, agricultural, and political knowledge. 70. Minban schools became the cornerstone of primary school development in the rural areas. Their major growth period followed on the collectivization of land in the early to mid-1950s. By 1956, primary school intake was 23 million, and the primary gross enrollment ratio stood at 62% (Lofstedt, 1980). Currently, local governments are in the process of assuming teacher salaries and other costs of minban, in order to upgrade and equalize quality. However, these locally constructed and operated schools have relieved the government of major capital outlays in the countryside, and continue to subsidize recurrent costs through community contributions. In one province, for example, commune, brigade and other community efforts accounted for 72% of the total capital investment in basic education, from 1979 to 1981. In Gansu province, for example, it was estimated in 1984 that nearly half of the recurrent expenditure per pupil, 15.50 yuan of a total 33, came from community contribution, and less than one quarter from public sources. This compares to an 89% public expenditure of 89 yuan per urban primary student (The World Bank, 1986). 71. In 1977, China took further steps to decentralize control of basic education, which will ultimately encompass nine years. Financial reforms in 1980 that separated state from local revenues placed both fiscal and administrative responsibility for primary education on local governments, who in turn encouraged community action. In 1978, central and local government expenditure on primary education had declined to 61% of total investment, from 86% in the 1950s. Contributions from communes and production brigades had risen to 24%, from less than 5% in the 1950s, and 13% in 1962 (Lofstedt, 1980). 72. By decentralizing control, the Government has encouraged local initiative in identifying and meeting the special educational needs of the local area. A variety of innovative educational programs have been undertaken by local governments. Spare time education for workers and peasants and their children, boarding schools in nomadic areas, travelling teachers in sparsely populated areas, are some of the options that local authorities have adopted to meet the educational needs of all children. Central Oversight for Ouality and Equity 73. While local effort and control have been the key to expanding educational services in rural China, disparities in educational quality are an obvious outcome. The central Government supports local efforts in primary education  19 through strategic investment, oversight of key educational inputs like teachers and texts, and through a variety of intangible supports. Many of these interventions aim at promoting equitable access to education, especially for groups which were previously marginal to the system, and at raising the quality of education in the rural areas. The central Government invests heavily in the education of minority students, which is much more costly per student than for the Han majority, owing to the need for special texts and teacher training, and the fact that many minorities live in underdeveloped parts of the country. In 1981, per capita expenditure for minority students was 290 yuan, or more than 10 times that for a Han student (Wang, 1984). The central Government also supports directly some of the locally-initiated special programs like pastoral boarding schools or travelling teachers, required to serve hard-to-reach populations. 74. In recent years, China has also targeted girls' enrollment as a high- need area. In Gansu province, there are 20-30 schools to which girls can bring their younger siblings, rather than caring for them at home. The Chinese Women's Federation has held informational meetings with local parents, and goverrnment publicity campaigns emphasize the importance of educating girls. 75. Training of primary teachers, both pre- and in-service, is largely conducted at the local level. Rapid expansion of education has depended on a make-shift teaching force of under-educated and untrained teachers. To bring all teachers up to an acceptable level of education and training, the Chinese Government has encouraged a wide range of teacher training programs. The results of these efforts are impressive: while the size of the primary teaching force increased over 500% from 1949 to 1980, the proportion of seriously underqualified teachers (those with education below the level of junior normal or middle school graduates) has steadily decreased, from 58% in 1953 to only 10% in 1980 (Wang, 1984). Over half of the remainder, those with the equivalent of senior middle school or normal school education, have been upgraded through in-service training. In-service programs include spare-time schools in urban areas, and correspondence, radio, and other forms of distance-education in the rural areas. In 1980, nearly 140,000 teachers graduated from spare-time training schools, while nearly a million and a half were enrolled. The Government also sets salary levels for teachers. 76. The central Government also exercises orersight in the area of textbook development. Primary curricula are determined by the Ministry of Education. Most textbooks are designed under Ministry of EducatiLr: oversight, and the forms sent out to provincial publishers for printing. At .be same time, local textbook design and publication are encouraged, in order to irovide materials specifically suited to local conditions. 77. Finally, the central Government continues to actively support efforts for basic education through a variety of informational campaigns. The directives issuing from national conferences on education and pedagogy were widely publicized, in order to give direction to the immense and little-trained teaching force. An example of these directives is the exhortation for teachers and students to balance work with rest, since ill health and exhaustion from heavy academic work loads were clearly a problem in the country-side. In addition, exemplary efforts of local communities are publicized, to serve as examples and  20 encouragement to other communities. Among their many other effects, China's literacy and adult education campaigns also create a context for valuing education, which surely encourages both parents and communities to provide education for their own children. Lessons frcm China 78. The delicate balance that China has struck between local and central control of primary educatiort provide several lessons on how UPE can be attained in a large, poor country which is geographically and ethnically diverse. Local control has been necessary not only for financial reasons, but also because variations in local social and cultural conditions create specific educational needs, which can best be identified at the local level. The central government encourages accountability at the community level as providing the best form of oversight for the massive educational system. 79. At the same time, where special needs generate an excessive financial burden on local communitles, the central government provides assistance. China's experience in educating minority students and those in remote areas underline the point that reaching the last ten percent or so of students is much mort costly per student than reaching the first ninety percent, and the need to make special efforts to reach these students. In order to equalize the necessarily uneven quality of locally provided schooling, the central government exercises oversight and encourages local accountability for key in-puts to educational quality such as teacher training and curriculum and textbook design. Finally, although this is difficult to measure, the intangible support and encouragement that the government gives to education must be considered as one of the factors contributing to the progress China has made towards attaining UPE. E. Sri Lanka 80. Sri Lanka's educational system expanded and developed several decades earlier than those of the other countries being considered here. The Education I of 1944 made education free at all levels from primary school to university, and gross enrollment at the primary level had reached 95% by 1960. Free education has come to be perceived as a social right in Sri Lanka, which no government has seriously attempted to abrogate. Annex 6 summarizes school enrollment information for Sri Lanka. 81. The school cycles and entrance ages have changed twice, once in 1972 and again in 1978. Sri Lanka's primary cycle now comprises one year of lower kindergarten plus grades 1-5. About 70% of the cohort completed primary school in 1981, and some 70%-90% of these went on to lower secondary, facing their first selective exam, the 0 levels, at the end of Grade 10. 82. From 1977 to 1981, the non-university education budget increased 125%, the university budget rose 360%, while overall government spending increased by 200%. There has been in other words a relative decline in funding for primary and secondary education (Enquist et. al., 1981). Because primary and secondary education have been financed under a single budget since 1975, it is difficult to determine their relative share in the total education budget. One of the major budgetary problems in Sri Lankan education, however, has been a pattern  21 of over- and under-expenditure of budget allocati ns. This may reflect weaknesses in educational management, a problem that Sri Lanka has been trying co address in recent years. 83. Despite Sri Lanka's impressive record as a poor country providing basic education, there are still children in Sri Lanka who do not enter or complete primary school. Tamil children in the estate sector have had particularly limited educational services, though this picture has been changing as the government began to assume responsibility for these schools in 1977. Nation-wide, government studies showed around 20% of 6-9 year olds out of scho.l; in the estate sector, 41% of this age group are not in school (Lofstedt at, al., 1985). Problems concerning the quality and relevance of primary education have continued to generate concern, and indeed, have been linked to the persistence of primary school dropouts and non-entrants. In recent years, Sri Lanka has undertaken administrative and curricular reforms aimed at raising the quality of primary education, in order to increase retention at the primary level, and thus to sustain UPE. Raising Oualitv in Primary Education 84. In Sri Lanka as elsewhere, the opportunity costs of schooling are certainly a factor influencing primary school participation. This effect is particularly strong for girls in the estate sector; a 1970 study showed primary participation rates for these girls rising from 47% to 63%, from the lowest to highest income groups. For rural boys and urban children of both sexes the effect of family income was a much weaker determinant of non-participation, with participation rates varying by 10% or less between the lowest and highest income groups. The relatively weak influence of income on participation for most groups in Sri Lanka points to other factors as determinants of non-participation. 85. As in Kenya, and from the same colonial heritage, education in Sri Lanka has been historically exam-driven and academically biased. Although primary and secondary enrollments have grown tremendously since 1945, university admissions have remained highly restricted. With no fees acting as barriers to higher education, exams provide the single mechanism of selection for entry into higher education and thus the highest echelons of Sri Lankan society. Such an evidently meritocratic system generates a certain public perception of education: that the purpose of each level is to prepare one for the next. When schools are of low quality, such as those that serve the poor in Sri Lanka, children and parents can lose hope that the child will succeed in entering the next level, and so loose interest in continuing or completing the primary cycle. 86. Sri Lanka has taken several steps to equalize quality between schools and enhance the relevance of primary education for all children. Among these have been pedagogical and curricular reforms of primary education. A pedagogical reform initiated in 1972 aimed at making primary education a child-centered process that dealt with: the student's real world. The pilot project began with training for a group of teachers from rural schools. Along with educational theory, these courses emphasized ways and means of improving the classroom environment using local resources. Major curricular reforms proposed in the 1981 White PaRer on education derive from the same interest in making primary education the first stage of basic education for all of Sri Lanka's children.  22 87. An important feature of the in-service training session undertaken in 1972 was that the Circuit Education Officers (CEOs), the highest-ranking local education administrators, were invited to attend the sessions as observers. In this way, the CEOs learned the educational philosophy underlying the new pedagogy, and so were able to provide informed support to the teachers when they returned to their posts. In the second round of teacher-training, many CEOs nominated schools in their district that were most in need of the program. The in-service training for teachers also had an indirect effect of enhancing community involvement in the schools. The enthusiasm of the teachers to make improvements in the classroom environment led them to request parent assistance, for example, in building play grounds and simple equipment (Perera 1983). Decentralizing Administration 88. Historically, there have been extreme disparities in the quality of education offered by different schools in Sri Lanka, related to the ability of each school to mobilize community support. Although education is fee-free, many of the inputs for teaching materials and facilities improvement come from parents. The quality problem is particularly acute for Sri Lanka's many small schools -- some 25% of schools had fewer than 100 students in 1983 -- probably because the parertcal resource base is smaller. Sri Lanka has recently taken several steps to decentralize school planning and administration, in order to raise quality overall, and to equalize quality between schools. 89. Sri Lanka is now divided into 24 educational regions headed by a Regional Education Officer (REO). Each region contains several circuits of 30-40 schools. CEOs supervise schools and report to the REO. Since the early 1980s, REOs have prepared 3-year rolling plans for educacional development of their districts which are submitted to the Minister of Education, who then prepares an annual Implementation Plan. By bringing the planning process closer to the level of school operations, it is hoped that specific needs of schools for quality improvement can be identified and addressed. At the same time, local level administration brings the primary schools closer to the community, enhancing parental involvement in education. Although difficult to measure, anecdotal evidence suggests that parent and community involvement in schooling contributes to enhancing the quality of education and stimulates a positive parental attitude towards schooling, both of which are likely to raise enrollment and retention rates. 90. A related administrative move has been to establish School Clusters, each comprising a few secondary schools and several small primaries, for the purpose of sharing facilities, equipment, and expertise. The large secondary school serves as the core school of the cluster, with its principle exercising administrative oversight of all schools in the cluster. The pilot School Cluster project began in September, 1981. By mid-March of 1983, 21 clusters of 225 schools in 10 educational districts had been established. These schools served 105,255 pupils, or 3% of the students in government schools. 91. The pilot clusters have shown some encouraging outcomes. Within many clusters, teacher assignment and the composition of classes has improved. Cluster principals have surveyed and requested assistance in upgrading facilities and  23 materials. Some anticipated problems, such as the core school receiving greater benefit than others or resentment to the core school principle, have not resulted). 92. According to the evaluations of the pilot project, what seems to make the most difference to the success of clusters are the quality of leadership exercised by the cluster principal, and of supervisory and inspection services (Perera, 1983). For that reason, the government has been paying special attention to deploying well-trained and highly motivated administrators to the core schools. Lessons from Sri Lanka 93. Sri Lanka's long experience with near-UPE, the problems it has faced and str:ategies it has devised in response, provide insight into the importance, and difficulty, of addressing issues of quality and relevance in primary education. Recognizing that one of the reasons children leave primary school is that the experience lacks obvious relevance to their future has led to continuing efforts to provide education that is meaningful both to those who continue beyond the primary level, and those who don't. Efforts at improving the quality and relevance of primary education are an important means of increasing enrollment and retention rates, and thus, of sustaining UPE. 94. Similarly, if the school environment and educational process are of low quality, there is less incentive for parents and children to forego earnings in exchange for schooling. Curricular and pedagogical reforms are one aspect of raising quality. Better administration at the local level is another, and is particularly suited to identifying quality needs at the school level.The introduction of regional planning, if well-conducted, should enable the government to provide educational inputs where they are most needed. 95. The Cluster School concept provides an inexpensive means of raising quality levels in small szhools, which do not have resources to provide a complete range of educational activities. Joint use of facilities and appropriate deployment of teachers within clusters allow an almost immediate benefit to small, poorly staffed or equipped schools.  24 V. Analysis of Lessons From Successful International Experience 96. From the analysis of the five case countries' rapid and sustained progress towards UPE, several conclusions can be drawn about the policies and programs which have stimulated the advances in their educational systems. All five countries have had, in the first case, to create a supply of accessible school places sufficient to the size of the child population. The mobilization of community resources played an important role in school construction. Each has also had to stimulate parental and child demand for primary education sufficient to keep children in school. The centrpl educational strategy for increasing demand for primary education has been to raise the quality of education, particularly thru providing in-service teacher training and enhancing local accountability. Each of the five case countries has taken policy and programmatic steps to address the nexus of issues underlying rapid progress in attaining and sustaining UPE. Below are summarized the specific measures undertaken by the countries studied to attain UPE. Making a Political Commitment to UPE 97. With many competing demands for limited resources, governments have made a firm commitment to primary education in order to attain and sustain UPE. Such a commitment frequently takes the form of specific UPE goals and targets emanating from a major reform study such as the "Challenge to Education -- Policy Perspective' White Paper in India. In other instances political will is legally embodied in the passage of compulsory education acts such as that of Indonesia, or constitutional rights such as in Tanzania. China and Tanzania launched UPE campaignz, encouraging communities to provide primary education and providing examples, advice, and assistance to foster this goal. Indonesia's SD Inpres program represented a major public commitment to UPE. In Sri Lanka, an inherited tradition of free education accessible to all makes UPE a key government responsibility in the public eye. Increasing Social Demand by Abolishing School Fees 98. Large increases in enrollment following the removal of fees suggests that school fees are a major barrier to primary participation for some children, even when other private costs remain. Kenya, Tanzania and Indonesia all abolished school fees as part of their drive towards UPE, while in Sri Lanka the increase in primary enrollment followed provision of free education. The governments of all five countries have taken successive steps to eliminate or limit other parental contributions, such as to school building funds, shifting non-government revenue sources to the community, and often to in-kind contributions, and away from individual households. Meeting Initial Capital Costs through Community Participation 99. Attaining UPE requires an adequate supply of school places. In all the cases presented here, as in India, the physical system needed expansion. School construction and equipment constitute the major capital investment in UPE. In three of the five cases, Kenya, Tanzania, and China, facilities have largely been provided by local communities. Even in Sri Lanka, which has long provided  25 free educational services, parents and communities continue as sources for capital investment, providing new equipment and renovating existing facilities. Only in Indonesia, through the Inpres program, has the total initial investment in primary education come from the Government. Cost containment for the Inpres buildings came from using standard and simple designs, and from prior mapping of school locations according to local need. Increasing Recurrent ExRenditure thru Intra-Sectoral Budget Shifts 100. Information on recurrent education expenditures is sketchy and hard to compare between countries, due to differences in budgeting and accounting systems. In all of the case countries except China, the recurrent cost of primary education - - which are primarily teacher salaries - - are largely borne by central governments. The available data suggest that a major financial commitment must be made to primary education, in order to attain and sustain UPE. In Kenya, Tanzania, and Indonesia, the proportion of the recurrent education budget devoted to primary education increased with the drive to UPE. Primary education absorbs approximately 63% of the recurrent education budget in Kenya, 45% in Tanzania, and about 50% in Indonesia. In China, local governments are assuming teacher salaries from the communities. In all five countries, however, there are also some subsidies to recurrent costs from communities, in the form of facilities renovation and supply of teaching materials. Micro-Planning for Primary Schools 101. Providing primary education to all children requires knowing where resources are needed. All evidence suggests that needs for facilities and teachers must be assessed at the local level. Tanzania initiated a child census conducted by teachers at the outset of its UPE campaign. In support of its school building program, Indonesia conducted an experimental mapping program in two provinces, before initiating nation-wide school construction. Because the buildings have been largely constructed by local communities, Kenya, Tanzania, and China did not engage in school mapping exercises prior to the drive towards UPE. Rather, communities were encouraged to build schools for their children. Providing and Motivating Competent Teachers 102. China, Tanzania, and Kenya all faced a shortage of primary teachers when the drive towards UPE was undertaken. In these countries, the need for teachers was met largely through in-service training of educated youth and adults. Indonesia embarked on its UPE program with a pool of teachers to draw on, while in Sri Lanka, the well-developed secondary education system provides an adequate supply of teachers. All of the case countries have engaged in on-going efforts to raise the quality of teaching through various in-service courses and the effective use of distance teaching technologies. The greatest benefit of such programs may lie in motivating teachers by enhancing their sense of professionalism and commitment.  26 Sustainina UPE by Increasing Oualitx to Raise Retention 103. All five countries continue to seek means of sustaining their progress towards UPE by raising the level of cohort retention. All have come to recognize that the auality of services is the most important educational determinant of retention. Poor children, especially the urban poor, face increasing opportunity costs for schooling as they get older. The need for child labor reduces parental demand for education; resource poor and low quality education provides little attraction to off-set the opportunity costs of schooling the poor. Conversely, there is evidence to suggest that children will stay in school if they and their parents perceive that school provides the student with knowledge and skills. Thus, one of the rationales for quality improvement of poorer schools in Tanzania, Sri Lanka, and China has been to increase retention in primary schools by reducing drop-out and repetition. Sustaining UPE by Increasing Relevance through Curricular and Exam Reforms 104. In all five countries, curricular reform has accompanied the drive towards UPE. These reforms are in part shaped by international trends in pedagogy. All also address the issue of quality and relevance in a mass education system. China alone has developed a two-track educational system for urban and rural areas. In the other cases, curricular reforms aimed at making the content of education relevant to the rural environment where the majority of children live, and at making primary education less academically oriented and more directed to developing basic skills and knowledge. In Kenya, examination reform linked to in-service teacher training helped drive the implementation of curricular reforms aimed at raising quality and relevance. Localizing Control to Ensure Educational Accountabilitv 105. In all the cases except Indonesia, efforts to attain and sustain UPE have moved governments in the direction of local accountability for primary schools. In Kenya, Tanzania, and China, community contributions to providing facilities initiated community involvement in schools. In Sri Lanka, primary school administration is being decentralized in order to bring administration closer to the communities. In all these cases, community involvement in the schools has been an effective means to raising quality and participation rates. Community involvement increases the normative pressure on teachers and administrators to be present and competent in their work. Promoting the Enrollment of Girls 106. It is almost always the case that a disproportionate number of out-of- school children are girls. While four of the five countries showed near-parity in girls' enrollment, with 48%-49% of the primary students female, it should be kept in mind that even these numbers suggest that there may be more girls out of school than boys. Sex-ratios in the child population can vary across countries but tend to be nearly equal, so 48% female students suggests under-enrollment of girls. Little information is available on girls' participation, and the barriers to it, in the case countries. However, it appears that the opportunity costs of young girls as caretakers of even younger sibling is a primary factor in under-enrollment and early dropout. Only in China, where enrolling girls has  27 been a particularly acute problem, is there evidence of national attention to the issue. Built-in child care at places of employment have particularly improved female enrollment in China's urban areas. The Women's Federation has launched an information campaign aimed at the parents of girls, and some schools in rural areas provide day-care for younger siblings of girl students. Reaching Remote PoDulations 107. Kenya, Indonesia, and China have all initiated programs to serve isolated or remote populations. All three countries have constructed boarding schools for children of nomadic groups. The scant data available suggest that in Kenya, these programs have met with little success, and in all three countries, have been very costly. China and Indonesia also employ itinerant teachers in sparsely populated areas. Indonesia's project PAMONG, or self-instructional materials program, provides a suitable curriculum for both itinerant and multi-grade teaching in remote areas. Remote populations in many countries are also linguistic minorities, and so require teachers and teaching materials designed especially for them, which is a relatively costly enterprise. Promoting Efficiency and Equity through Ouality Imorovements 108. The central finding of the case studies has been that equity, quality, and efficiency in primary education are closely linked and simultaneously influence each country's success in attaining and sustaining UPE. UPE is inherently an equity issue. The children who are not in school at any point in time represent specific child populations -- the rural and urban poor, girls, ethnic minorities - - for whom educational services are not available, or economic constraints make less attractive. Low quality education increases repetition and drop-out rates, lowering the system's efficiency. Low quality education is a particularly acute problem for children from poorer and marginalized populations.  28 VI. Im2lications for India 109. The findings of the case studies suggest certain policy and program directions that India might take to realize her goal of attaining UPE by 1995. These are summarized below. Focus on Opportunitv Costs of Child Labor and Ouality Improvements to Increase Retention 110. Poor children face powerful disincentives to schooling. It is probably worthwhile to explore and try to directly address some of these environmental constraints, such as the need for girls to fetch water and care for younger siblings. At the same time, the case studies all show that the single most effective educational measure for increasing primary school retention is to raise the quality of schooling provided to the poor. High quality education can and has offset the opportunity costs of schooling, leading to higher participation and cohort survival rates. Adequate facilities, a supply of teaching aids, and competent, motivated teachers are basic determinants of school quality that can be upgraded through well designed and administered policies and programs. SuDDort Key Oualitv Improvements such as In-Service Teacher Training through the Use of Educational Technology 111. Among the several inputs to quality improvement identi.ied in the case studies, in-service teacher training is most central. The majority of India's one and a half million primary teachers work under difficult conditions that include professional isolation. In lieu of greater financial incentives, professional support through in-service training may increase motivation and improve classroom performance of primary teachers. The greater sense of professionalism that comes from increased knowledge and competence is in itself a reward to teachers, and a gift to the children they teach. As demonstrated in other countries, mass in-service teacher training through distance teaching technologies is most cost-effective. Encourage Community Participation and Accountability for Primary Schools 112. The case studies demonstrate that, with a modicum of encouragement, local communities can and will contribute to the construction and renovation of school buildings. Furthermore, they show that when communities do become involved in local schools, they exercise oversight of quality. Strengthening local-level management brings two benefits to the schools. Local management is more accessible to parents, thus enhancing community involvement. When there is solid vertical integration between local and central management, specific educational problem areas can more effectively be identified and targeted for improvement. 113. Both the principle of local oversight and the sheer size and diversity of India's population dictate the need for strong community governance and local management of India's primary schools. Strong panchayat involvement in schools and development of local boards can create channels of influence and normative pressure for quality education. For local management to effectively address  29 educational needs, two things are required. First, managers must be competent and motivated, able to comprehend and deal with the particular problems of their schools. This implies a need for management education for local administrators. Second, local managers must have access to resources for meeting needs that cannot be covered by local resources. This requires establishing good channels of communication and accountability between local, state, and central administration, perhaps backed by a management information system. Target SRecial Needs for Central Government Intervention 114. Enrolling and retaining the last 10%-15% of children is generally more costly per child than providing education to the "first" 85%-90%. Lower quality schooling has little holding power for children who could be contributing to the household economy. Linguistic minorities and remote populations are a subsection of the poor who will require special teaching materials and specially trained teachers. Providing special services and raising the quality of village and slum schools both cost money that may not be available from local resources. Furthermore, the Central Government has heretofore expected the States to bear the burden of costs for classroom construction while instructional inputs like blackboards, chalk, etc., would be provided under "Operation Blackboard" by the Center. However, poor States have been unable to meet this financial burden. For India to see all her children through the primary cycle, selective support of school construction by the central government will be necessary. Putting Budgetary Commitments in Line with Policy StaLements 115. A comparison of India's allocations on education with those of countries which have been more successful in realizing UPE suggests that current levels of funding for primary education in India may not be sufficient to the task of attaining and sustaining UPE. There are no hard and fast iules for how much is finance is enough, but in all of the case countries for which data is available, the proportion of recurrent education expenditure assigned to primary education rose with the drive towards UPE, and has remained at 45%-60% of recurrent education spending. In India, the proportion of recurrent eduction expenditure has fallen to only 25% of the recurrent education budget. The relative fiscal allocation to primary education does not promise success in attaining UPE, nor does it reflect in financial terms the priority UPE is assigned in policy statements.  30 ANNEL 1 Page 1 WORLDWIDE TRENDS IN ATTAINING AND SUSTAINING UPE 18. Populous. Recent Rapid Growth. High Retention Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate ALGERIA 97 3,185,900 85 7 7.1 CHINA 116 139,720,000 - - INDONESIA 100 22,904,600 72 9 5.4 KENYA 98 3,520,500 61 5 8.4 MEXICO 121 14,619,900 60 10 5.7 U.R. TANZANIA 104 3,450,300 84 1 10.5 lb. Less Populous. Recent Rapid Growth. High Retention Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate BAHRAIN 102 48,000 -- -- -- BOTSWANA 102 171,900 83 6 8.1 QATAR 105 29,700 -- 10 -- SYRIAN A.R. 100 1,556,400 85 8 6.0 ZAMBIA 95 1,031,500 87 2 6.6 2. Earlier or Gradual Growth. High Retention Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate ARGENTINA 112 3,841,000 71 9 1.6 BARBADOS 120 32,100 -- -- -- CHILE 120 2,234,500 84 7 3.2 COSTA RICA 107 348,700 79 11 2.9 CUBA 112 1,516,400 99 4 1.8 GUYANA 99 135,900 -- -- -- HONGKONG 111 550,400 101 2 1.6 JAMAICA 99 358,200 97 5 1.9 MAURITIUS 103 129,600 64 0.7 PANAMA 116 347,000 69 12 3.7 PERU 112 3,177,700 66 14 4.3 PHILIPPINES 107 8,274,900 70 1 3.3 PUERTO RICO 104 461,600 86 -- -- SINGAPORE 106 289,800 103 1 0.1 SOUTH KOREA 109 5,658,000 97 -- 2.3 SRI LANKA 100 2,081,400 95 9 0.2 THAILAND 96 7,370,800 81 8 3.2 TRIN&TOBAGO 93 165,500 94 5 0.0 TUNISIA 103 105,400 91 21 4.4  31 ANNEX1 Page2 TURKEY 106 5,930,800 85 9 3.5 URUGUAY 104 328,500 96 15 0.2 U.A.E. 116 88,600 128 6 -- VENEZUELA 104 2,513,300 67 10 3.7 3. High Enrollment. Low Retention Country Gross Total Retent. Repeat Growth Enrol' Enrollment Rate Rate Rate ,. ... ... -- -- -- - -- - -- - - . - - ..... BRAZIL 93 22,447,400 38 21 5.7 COLOMBIA 133 4,318,200 41 12 4.6 DON. REPUBLIC 102 1,066,700 32 14 4.0 D.KAMPUCHEA 91 989,500 -- ECUADOR 108 1,432,100 59 9 4.8 FIJI 109 129,600 -- -- -- GUINEA-BISSAU 98 81,500 -- -- KUWAIT 98 153,500 -- 6 3.5 LAO P.D.R. 98 4,809 20 -- 8.7 LEBANON 113 385,900 -- -- 2.1 LESOTHO 104 243,100 46 24 3.0 MADAGASCAR 102 141,800 31 -- 6.9 MALAYSIA 92 2,012,900 -- -- -- MONGOLIA 105 143,700 -- 2 3.5 MOZAMBIQUE 97 1,387,200 -- 29 6.2 NICARAGUA 90 424,900 23 18 5.5 PARAGUAY 103 516,800 52 11 2.7 SURINAME 104 83,400 -- 24 -- SWAZILAND 103 109,100 -- 11 U.R. CAMEROON 104 1,342,800 57 29 5.6 VIETNAM 116 8,108,300 57 -- -- ZAIRE 92 4,278,800 58 19 5.1 4. Low Enrollment. High Growth Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate -------~~~. ------. . .......... - - - - - - - ........ .. AFGHANISTAN 31 1,062,600 27 6 9.7 ANGOLA 66 516,100 -- 29 13.5 BENIN 61 379,100 53 26 7.5 BHUTAN 15 25,000 -- 13 12.2 BOLIVIA 84 978,200 47 -- 5.0 BURKINA FASO 18 195,800 69 14 6.5 C.A.R. 72 259,800 42 35 6.6 D.YEMEN 83 261,200 97 -- 12.6 EL SALVADOR 80 906,800 36 8 5.1 ETHIOPIA 39 1,944,000 35 -- 11.9 GAMBIA 43 38,900 -- -- -- GUATEMALA 69 803,400 31 15 5.4  32 ANNEXL1 Page 3 GUINEA 36 277,200 58 24 5.0 HAITI 63 606,600 34 13 5.1 HONDURAS 87 567,900 34 15 5.5 IVORY COAST 78 407,300 79 26 7.6 LIBERIA 64 219,000 50 11 7.0 MALAWI 61 827,900 31 16 5.4 MALI 29 333,700 58 31 7.8 MAURITANIA 31 96,600 86 17 11.1 4. Low Enrollment. High Growth. continued Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate -- - - ---- . . ........ ....... ........ .... ...... MOROCCO 77 2,189,300 88 22 5.3 NEPAL 89 1,047,700 31 -- 11.3 NIGER 25 217,100 82 14 11.3 NIGERIA 88 11,683,500 -- *- 8.1 OMAN 62 91,900 90 11 -- PAKISTAN 58 7,382,500 42 -- 5.0 PAPUA N.G. 63 299,800 84 -- 9.0 SAUDI ARABIA 65 929,900 74 13 11.0 SENEGAL 45 416,700 87 15 6.1 SIERRA LEONE 40 264,800 55 -- 6.7 SOMALIA 43 303,500 46 -- 13.7 SUDAN 53 1,510,400 68 -- 7.9 UGANDA 61 1,545,800 -- 10 6.6 YEMEN 47 460,600 41 20 -- 5. lpw Enrollment. Low Growth Country Gross Total Retent. Repeat Growth Enrol% Enrollment Rate Rate Rate ~~~~~~~~~~~~~~~.. ... ...... -- - - - - ...... - - - - BANGLADESH 64 8,219,300 23 19 4.5 BURMA 85 4,050,500 29 -- 4.9 BURUNDI 27 168,600 95 15 -- CHAD 37 254,100 26 37 -- EGYPT 76 4,662,800 89 2 2.8 EQ. GUINEA 78 36,000 -- -- -- GHANA 73 1,449,400 74 2 3.9 INDIA 78 71,617,900 43 -- 3.8 RWANDA 65 653,400 58 12 -- ZIMBABWE 79 1,107,300 83 1 4.8  33 Annex 2: School Enrollments in Kenva #Primary Gross % #Primary Gross Gross Students Enrol Female Teachers Enrol Enrol Ratio Second Tertiary 1950 365,061 -- 27 7,692 -- -- 1955 432,678 -- 26 10,255 -- -- 1960 781,295 47 32 18,624 2 0.14 1963 891,533 -- -- 22,722 -- -- 1965 1,010,889 53 37 30,592 4 0.28 1967 1,133,179 -- 39 35,672 7 -- 1968 1,209,680 -- 40 37,923 -- -- 1969 1,282,297 -- 41 38,312 -- -- 1970 1,427,589 58 41 41,479 9 0.79 1971 1,525,498 66 42 49,396 10 -- 1972 1,675,910 71 43 53,536 11 -- 1973 1,816,017 74 44 56,343 11 -- 1974 2.734,398 107 45 78,340 12 0.23 1975 2,881,155 109 46 86,107 13 -- 1976 2,894,617 105 46 89,074 16 -- 1977 2,974,849 104 47 89,773 17 -- 1978 3,232,000 -- 47 92,046 18 0.70 1979 3,698,246 108 47 92,762 18 -- 1980 3,926,629 110 47 102,489 19 -- 1981 3,981,162 105 48 110,911 20 -- 1982 4,184,602 104 48 115,094 17 -- 1983 4,323,822 100 48 117,705 19 -- 1984 4,380,232 102 48 122,738 20 -- 1985 4,702,414 94 48 138,374 20 -- Source: UNESCO Statistical Yearbooks, 1969-1986.  34 Annex 3: School Enrollments in Tanzania #Primary Gross % #Primary Gross Gross (ages) Students Enrol Female Teachers Enrol Enrol Ratio Secondary Tertiary 1960 6-13 455,293 25 34 10,059 2 -- 1965 795,719 32 38 14,716 2 0.05 1968 833,898 37 39 17,893 3 0.1 1969 829,169 -- 39 17,529 -- -- 1970 7-13 922,083 38 39 20,094 3 0.2 1971 1,017,942 37 39 21,926 3 0.2 1972 1,126,165 40 40 23,167 3 0.2 1973 1,260,103 47 41 25,688 3 -- 1974 1,592,396 58 42 29,735 3 0.2 1975 1,954,443 70 43 39,245 3 u.2 1976 1,956,320 70 -- 38,199 3 0.2 1977 2,973,487 99 45 -- 4 0.2 1978 2,990,361 96 46 64,822 -- 0.4 1979 3,211,568 100 46 77,329 4 0.3 1980 3,367,644 93 47 81,153 4 -- 1981 3,538,183 94 48 81,659 3 0.4 1982 3,512,799 90 48 88,370 3 0.4 1983 3,552,923 87 49 85,308 3 0.4 1984 3,493,469 83 -- 91,359 3 -- Source: UNESCO Yearbooks. 1969-1986.  35 Annex 4: School gnrollments in Indonesia . .................................................... #Primary Gross % #Primary Gross Gross Students Enrol Female Teachers Secondary Tertiary Ratio Enrol% Enrol% ........ .....