Mexico - The initial education strategy

Report No. 101 29-ME Mexico The Initial Education Strategy Octobey 23, 1992 Count,y lDepartment II Human Resources Operations Division Latin America and the Caribbean Regional Office FOR OFFICIAL USE ONLY MICRFlUTE COPY PepTt o. OI2-4~ Tyqpe (SEM lei THE INITIA., EDUCATION STRATEGY i .,Ti :\~tiQr:PEEIRA, A. 3 pO39( ioQm:17 125 Dept. ;LLHR Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties, Its contents may not otherwise be disclosed without World Bank authorization. I~~  FOR OFFICIAL USE ONLY THE INRIIAL EDUCATION STIATEGY TABLE OF CONTENTS EXECUTIVE SU9MMARY . ............................................ i Recommendation 1: Priority to Non-Formal Initial Education .... ...... vi PART ONE: RATIONALE FOR DNiIAL EDUCATION .............. ....1...... 1. THE IMIAL EDUCATION CONCEPT ....... ............ .......... I A. Terminology ................. ........................ I B. Approaches ............................ 2 H. THEORETICAL FRAIEWORK ................................. . 3 A. Scientific Support ......................... 3 B. Socioeconomic Arguments ................................. 6 1. Efficiency ....................................... 6 2. Social Equity . 10 EmI. EVIDENCE FROM INrTIAL EDUCATION PROGRAMS ...... .............. 11 A. Evidence from Industrialized Nations ............11.............. I B. Evidence from Developing Nations ..... I....... ............... 14 1. The Education Approach .............................. 14 2. The Health and Nutrition Approach ....................... 18 C. Conclusiors .......................................... 19 TV. INTERNATIONAL FRAMEWORK ............................... 19 A. The ConveLtion on the Rights of the Child ....... .. ............. 20 B. Education for All ................. ..................... 20 1. The World Declaration ............................... 20 2. The Framework for Action ............................ 21 C. The World Summit for Children .............. ................ 21 V. BANK EXPERIENCE WITH EARLY EDUCATION ....... .. ............ 22 PART TWO: MEXICO'S NON-FORMAL lNlTIAL EDUCATION PROGRAM .... ...... 24 VI. NON-FORMAL DIIAL EDUCAIION IN MEXICO ................ ... 24 A. Socioeconomic Framework .............. .................. 24 B. Overview of the Iitial Education Sector ........ .. .............. 25 1. Description ............. ......................... 26 2. Public Expenditures ......... ........................ 29 C. Results of an Evaluation ................ .................. 29 D. Issues in Non-Formal Initial Education ......... .. .............. 30 Issue 1: Limited Priority Profie ............................. 32 Issue 2: Insufficient Institutional Capacity ....................... 33 Issue 3: Deficient Targeting Mechanisms ....................... 34 Issue 4: Incomplete Educational Inputs ........................ 35 Issue 5: Loosely Defined Supervision ......................... 36 This report was based on inputs from Messrs./Mmes. Antonio Pereira, Carmen Hamann, Louise Cord, Manuel Vera, David Warren, Darren Dorkin, and Vinh Nguyen (LA2HR); Monika Huppi (A2C2); Eduardo Ve. ~z (LATHR); Robert Myers (CONS). Peer reviewers for the report were Mmes. Michelle Riboud (LA2HR) and Eleanor Schreiber (LA3HR). The managing SOD Chief was Mr. Kye Woo Lee (LA2HR). The managing CD Director was Mr. Rainer B. Steckhan (LA2DR). This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Issue 6: Unsatisfactory Incentives for Community Educators ........... 36 Issue 7: Inconsistent Community Participation ...... .............. 36 Issue 8: Sporadic Linkages with Complementary Programs ....... ..... 37 E. Government's Strategy for Initial Education .......... ............ 37 F. Recommendations ...................................... 39 Recommendation 1: Priority to Non-Formal Initial Education .... ...... 39 Recommendation 2: Institutional Strengthening ................... 39 Recommendation 3: Improved Program Targeting .................. 40 Recommendation 4: Improved Program Inputs ................... 40 Recommendation 5: Supervision Redefined ...................... 41 Recommendation 6: Community Educator Incentives ..... ........... 41 Recommendation 7: Promotion of Community Participation .... ....... 42 Recommendation 8: Linkages with Nutrition and Health Programs .... .... 42 BIBLIOGRAPHY ................................................... 43 Annexes Annex I - Issues in Basic Education in Mexico (excerpted from World Bank Report No. 8930-ME, dated June 1991) Annex 2 - SEP Organizational Structure Annex 3 - Emollment Trends in Initial, Preschool and Primary Education in Mexico Annex 4 - Participants by Age Group in CENDIs by Type of Provider (1991) Annex 5 - CENDI Centers, Enrollment and Unit Costs by State (1991) Annex 6 - Evolution of SEP Education Spending Annex 7 - Breakdown of 1991 Actual SEP Iitial Education Budget Annex 8 - Initial Education Unit (IEU) Organizational Structe Annex 9 - Regional Staff for Non-Formal Initial Education Program Annex 10 - CENDI Employees Annex 11 - Enrollment and Coverage Rates for Non-Formal nitial Education (1991) Annex 12 - Projected Non-Formal Program Coverage and Enrollment Annex 13 - Non-Formal Enrollment and Unit Costs by State (1991) Annex 14 - Composition of Marginality Index - IBRD No. 23547  MEXICO'S INITIAL EDUCATION STRATEGY EXECUTIVE SUMIARY PART ONE: RATIONALE FOR INITIAL EDUCATION Initial Education Concept 1. Initial education (IE) promotes the physical, emotional, intellectual and social development of infants and toddlers, between birth and the age of four. It revolves around four interrelated and simultaneously occurrfihg processes: child survival, growth, development and care. There are two principal types of initial education: a formal modality (used by traditional kindergartens and nursery schools) and a non-formal approach. In the past two decades, the non-formal method has received increasing attention, as countries have sought to provide early education on a larger scale and to draw on the local culture by incorporating parents and communities. The Case for Initial Education 2. The scientific, economic and equity arguments in favor of initial education are compelling. Scientific evidence highlights the link between early sensory stimulation and the development of intelligence. Initial education also reduces societal inequities as its low unit costs and community orientation permit it to effectively reach children from low income families. Empirical evidence indicates that initial education raises the return on education investments, contributes to human capital formation and reduces public expenditures. 3. Scientific SuDDort. There is strong scientific support indicating a link between initial education and the child's intellectual development and academic performance later in life. Research reveals that sensory stimulation affects the structure and organization of neural pathways, which are formed during the first two years of life, and which in turn, are linked to the child's cognitive and social development. Furthermore, deficits caused by inadequate intellectual stimulation and affective care, as well as early malnutrition, are likely to result in irreversible damages on the physical and emotional capabilities of human beings, which are crucial for fuirther learning. In this sense, elementary education and even kindergarten can be too late to develop these capacities in children. 4. Social Eguity. Initial education can help modify inequalities rooted in poverty and discrimination, by giving children from disadvantaged backgrounds, especially those living in raral and urban-marginal areas, girls and other children at risk, a fair start. Poverty is by far the most discriminatory of all social factors. A large part of cognitive achievement differentials between lower socioeconomic and higher income groups can be attributed to malnutrition, lack of sanitation, and to low levels of psychological stimulation among poor children. S. Efficiency. Early childhood education increases the return on primary and even secondary school investments, raises participants' productivity and income levels, and has been shown to lower public expenditures (i.e., lower welfare, health and education costs). In the Perry Preschool Project in the U.S., a US$1 investment yielded US$6 in tax-payer savings due to lower welfare and educational expenditures and higher productivity among participants. Similarly, in the U.S. Head Start Program, US$1 spent is worth US$4.75 in costs saved. 6. Innumerable empirical studies reveal that initial education improves the school readiness skills of children. Thirteen studies have traced the effects of initial education on the child's primary school performance. These studies, carried out in Asia, the Middle East, and mainly Latin America, suggest  ii that initial education also promotes timely school enrollment, lowers primary repetition and dropout rates, and improves overall academic skills. The beneficial impact of initial education is particularly pronounced among the traditionally disadvantaged groups (e.g., girls and children with rural, indigenous, and lower socioeconomic backgrounds). In five out of seven studies evaluating school readiness, there were significant improvements in the child's skills. Six of the nine studies found a positive relationship between initial education and lower dropout asnd repetition rates. In seven of the eleven studies with available information, children participating in early childhood programs had a better primary academic performance than the control group students. 7. Initial education programs often involve nutritional supplemenation and health care services. Evidence suggests that health and nutrition-related programs are also positively associated with physical growth, basic cognitive abilities, school readiness and positive classroom behavior. A ten year study in the state of Puebla (Mexico) shows that supplemented children walked at an earlier age and demonstrated language superiority from the 20th week onward. In the early years of schooling the supplemented children were more active and performed significantly better than unsupplemented children. Slightly more than one-third of the unsupplemented children failed the first year while there were no failures among the supplemented children. International Framework 8. In the past three years, three international events have demonstrated new concern for early childhood education and have addressed some of the institutional and organizational challenges facing early childhood programs at the national government levels: (a) the Convention on the Rights of the Child was adopted by the United Nations General Assembly in November 1989; (b) the World Conference on Education for All was held in March, 1990, in Thailand, under the sponsorship of the World Bank, UNDP, UNESCO, and UNICEF; and it placed special emphasis on the expansion of early childhood care and development, involving family and community interventions, especially for poor and disadvantaged children; and (c) the World Summit for Children, held at the United Nations in New York on September 30, 1990, brought together the second largest gathering of Presidents and Prime Ministers in history. These meetings have expressed a new awareness by the international donors and more practical commitments by a number of governments in favor of early childhood development. World Bank Experience with Initial Education 9. At the end of the 1970s two studies commissioned by the Bank (Grawe and Smilansky, 1979) recommended the implementation of policies designed to assist poor children by working directly with their families, and by stressing educational support to the family unit. In the early 1980s, the Bank provided funds to enhance child development as part of more comprehensive projects, and more recently, specific projects have been designed. In the Bank, child development activities have been interpreted broadly to include loans that affect the physical and psycho-social development of children between the ages of 0-6 (through education of caretakers or by providing infrastructure), and to include activities specifically focussed on reducing infant and young child mortality (breast-feeding promotion, immunization, etc.). 10. The bulk of activities until the mid-1980s were in the context of health and nutrition and urban development programs, and virtually nothing had been done within the education sector. The most typical activities were: health and nutrition screening services, food supplementation, physical infrastructure of child care centers and day care services. More recently, components on education for preschoolers have been added to this list.  iii 11. More representative of the new projects related to child development are the following: India: Integrated Child Development Services Project. 1990. The project was designed to improve the nutrition and health status of children under 6 years of age, with special emphasis on 0-3 year old children, and pregnant and nursing women. This should be achieved, through: increases in the range, coverage and quality of nutrition and health services; training for health and nutrition workers and mothers; community mobilization to increase local involvement in project services; support to provide educational toys and play materials to the villages' Integrated Child Development Service Centers; and, the improvement of preschool education through parental and caregiver training. This last action is an integral part of the training, communication, and community mobilization components of the project. Colombia: Community Child Care and Nutrition Project, 1990. The project was designed to strengthen an ongoing program of home-based child care. The program caters to nutrition, health, and specifically to early childhood development needs of children age two to six in low-income communities through "community mothers" who care for up to 15 children each in their homes. This Child Care Homes program combines the goal of day care for poor children (including stimulation), supplemental feeding, and community development. 12. Given the evidence regarding the benefits of early education programs, and in light of increased internati6nal awareness and national commitments to the concept, the time seems ripe for more ambitious Bank involvement in early childhood development programs. In this context, the Bank is supporting the improvement and expansion of Mexico's non-formal initial education program. PART TWO: MEXICO'S NON-FORMAL INrTIAL EDUCATION PROGRAM 13. To address problems of low human capital formation, poor primary school performance, and inadequate early childhood stimulation, Mexico introduced two modalities of initial education in the late 1970s. Both programs are supported by the Initial Education Unit (IEU) currently located in the Preschool General Directorate within the Elementary Education Sub-Secretariat of the Secretariat of Public Education (Secretar(a de Educaci6n Pablica -SEP).. The first is a formal program handled mainly through the Government's Child Development Centers (Centros de Desarrollo Infantil, or CENDIs), which care for children aged 45 days to 6 years. The program provides education, health and nutrition care to children of working mothers employed in the formal sector, who by law are entitled to day care services as a fringe benefit (although not all enterprises comply). The majority of the children (60%) attend government operated CENDIs, while the rest attend private centers. 14. 'he second approach is the non-formal, "Zero-to-Four Year Old Child Development Program through Paents and Community Members." The objective is to educate and "empower" parents, mainly mothers, in such a way that they improve their care of and interaction with their children. The program emphasizes the family and aims to enrich the immediate environment in which child development occurs (the home), rather than to provide a substitute for it. The program is oriented toward low-income families in poor rural and urban marginal communities. The non-formal program rests upon a system of successive training. SEP professionals at the central level work with state secretariat personnel to train the zone coordinators, who in turn train module supervisors, who then train and supervise up to ten community educators. The community educators, together with 20 sets of parents and their children, form the core or nucleus of the non-formal initial education program. After decentralizing the basic education system to the States, SEP is currently redesigning its cential level strmctue.  iv 15. The relative size of and budgetary allocations to Mexico's overall initial education program are small when compared to the preschool and primary school programs. Within the overall SEP budget, the share allocated to initial education represents approximately 0.4% (1991) of SEP's total educational budget, of which 80% was allocated to the formal CENDI program, and only 20% to the non-formal program. SEP expenditures on initial education include all the costs of the non-formal program, but only a small share of the costs of the formal CENDI program, the overwhelming majority of which are borne by the employers. 16. In 1987, the Netherlands-based Bernard Van Leer Foundation flnanced an evaluation of the Mexico initial education program. The study indicates that the non-formal program was equally effective and in some cases more effective than the formal preschool program in improving the child's school readiness skills and primary school performance. The positive impact of the non-formal program was especially marked for indigenous children. Overall, students participating in the non- formal program: scored 8% higher on a school readiness test than control students; had a first grade repetition rate of 9% compared to 27% for non-program students; and scored 13% higher than control children on a test of academic skills. Issues 17. Mexico has undertaken a pioneering program in the field of non-formal initial education, and has built-up ten years of experience in the area. However, recently the program has faced the following issues, which have hindered its progress. 18. Issue 1: Limited Priority Profile. A small share of the SEP budget reflected the limited priority profile given to IE programs until recently. While focusing on poor rural and urban marginal families, the non-formal IE program still lacks a strong institutional identity and adequate budgetary resources. The cost per child of US$31 is the lowest of any similar program in Latin American countries. The CENDI program, which includes day care, physical and mental health care, and food and nutrition in addition to initial and preschool education, has a cost per child of US$1,406. Primary education costs per pupil are approximately US$100. 19. At the central level, only 6% of the IEU's 336 staff members are directly involved in the non- formal program compared to 19% which directly support the CENDI program. Furthermore, the planning, technical support, supervision and administrative sub-units in the IEU spend the majority of their time backstopping the formal CENDI program, and less than 25% is allocated to non-formal education. Although not directly comparable, in 1991, the child:educator ratio in the CENDI program was 8:1, compared to a parent:educator ratio of 26:1 in the non-formal program. 20. Issue 2: Insufficient Institutional Capacity. The IEUs at the central and state levels lack the capacity to effectively manage and supervise the program, evaluate its interventions and ensure its quality, due to the following factors: (i) inadequate numbers and qualifications of key flnancial management and planning personnel; (ii) inefficient flow of information between different administrative layers; (iii) absence of homogeneous and efficient information and management systems; and (iv) the lack of an evaluation and monitoring unit to provide feedback on the program. 21. Issue 3: Deficient Targeting Mechanisms. There is no overall targeting methodology to distribute program resources among and within the states. As a result, resource allocations and enrollment levels are determined haphazardly. Although the program is aimed at children from low- income families, the allocation of program resources among states and at the local level is currently not based on poverty criteria nor on the region's potential demand.  v 22. Issue 4: Tncomplete Educational Inputs. As a result of Inadequate attention and resource flows to the non-formal program, the quality of Its teaching strategies and materials have been undermined. Despite the fact that some community educators exhibit very positive approaches and seem to be in command of the main tools of their activities, the quality and level of community educators' skills has in general been declining. Furthermore, the weekly parent-promoter meetings, originally conceived as the focal point of Interaction within the nucleus (i.e., promoter/family) have almost disappeared as an educational strategy. As a result, the domestic visits, originally designed to reinforce the weekly meetings have become the main activity within the nucleus. Finally, the seven program training guides, prepared in 1981, have never been updated. While It is recognized that the guides were well designed and serve the program's goals, they require updating in terms of their content, written text and visual presentations. 23. Issue 5: Logsely Defined Supervision. As a result of inadequate staffing in the state and central IEUs, the non-formal regional supervisors are forced to assume the administration of the program, and have little time for supervision. Furthermore, limited budgetary allocations for travel expenses restricts the mobility of the regional supervisors, which reduces the frequency and the quality of supervision and educational services. 24. Issue 6: Unsatisfactory Incentlves for Community Educators. The program's weak institutional capacity has led to problems in delivering compensation to the widely dispersed community educators in a timely fashion, undermining the program's ability to atract and retain qualifled community educators. Furthermore, their monthly compensation of Mex$80,000 Is very low even when compared to other "volunteer" community educators with similar backgrounds and responsibilities. 25. Issue 7: Inconsistent Community Participation. Without a clear Government strategy for non-formal initial education and reflecting inadequate program diffusion efforts, the program has been unable to engender sufficient community support. The community based "Pro-Childhood Committee" has not been able to play the expected supportive role for the program. Furthermore, the program has not been able to dzvelop a reliable social marketing strategy to increase communities' knowledge and acceptance of the program. 26. Issue 8: Sporadic Linkages with Complement Programs. Given the complementary relationship between psycho-social development and health and nutrition, it is important that malnourished children also benefit from nutritional support. Currently, the non-formal program has no official links with other nutrition and health programs that operate in poor communities. Furthermore, because of recurrent cost constraints, program leadership has been unable to finance food and/or micro-nutrient supplements. Government Strategy 27. The Mexican strategy for initial education is articulated in two policy documents: the Education Modernization Program (1989-1994) and the SEP proposal submitted in early 1991 to international donors requesting assistance for the sector. The Government seeks to expand coverage of the non-formal program nationwide to reach a total of 3.7 million children over the next decade. It also seeks to improve program quality, expand the treatment of hygiene and environmental issues, reinforce the program's institutional capacity, and expand program promotion efforts through adult education courses and mass media programs. Recommendations  vi 28. The following recommendations are offered which could be incorporated in a coherent strategy for non-formal initial education that addresses the above issues. Actions to implement these recommendations have been included by the Mexican government in the design of the lE project submitted for Bank financing. The proposal is based on the assumption that with some speciflc improvements the existing program can maximize Its accumulated experience and multiply its far- reaching effects among the poorest children of Mexico. 29. Recommendation I: Priority to Non-Formal Initial Education. Initial education, regardless of the modality utilized, deserves higher priority in Mexico's social policies and educational strategies. Policy statements must be as comprehensive and pervasive as possible in recognizing the role played by IE as an efficient preparation process for preschool and primary school. As a consequence, budget allocation for IE must be commensurate with its recognized importance and the commitments made by the Government at national and international forums. 30. Reqommendation 2: Institutional Strenghenina. As resource flows to, and the size of the non-formal program expand, the IEU will need to be strengthened and reorganized at the central and state levels. It is recommended than a computer based information network be set up nationwide. To evaluate and monitor the impact and activities of the program, an evaluation system should be designed and implemented. 31. Recommendation 3: Improved Program Targeting. Within each state the program should select rural and urban marginal communities according to an index of marginality. 32. Recommendation 4: Improved Proeram Inputs. SEP needs to define new mechanisms for identifying excellent community educators and rewarding them with economic incentives and opportunities for personal and professional progress. It is recommended that the role of the weekly parent's meetings be reinforced and clarified. The current guides and manuals should be reviewed (for content, text and presentation) and new educational materials should be developed to supplement the existing material and encourage a more active and participatory training process. 33. Recommendation 5: Supervision Redefined. A more precise and clear description and distribution of educational, supervisory and administrative functions need to be established for zone coordinators, module supervisors and community educators, in order to ensure the effectiveness of program norms, expedite the flow of information throughout the system, and improve efficiency in the use of program resources. 34. Recommendation 6: Community Educator Incentives. The states need to be more innovative in identifying the best community educators available and in rewarding them with economic and social incentives and opportunities for both personal and professional progress. 35. R m tin-7 Promotin f omuni ParticipatioLn. More community involvement should be stimulated through an enlarged membership in the Pro-Childhood Committee and expanded program promotion efforts (such as multimedia communication techniques). 36. Recommendation 8: Linkages with Nutrition and Health Programs. It is recommended that the non-formal program intensify its relationship with other poverty alleviation programs that function at the local level.  This section provides .he just:flcation for Bank involvement in non-formal Initial education. After a description of the Initial education concept, theoreticalframework and empirical evidence for the non-formal Initial education are evaluated. A brief review of Bank Involvement to support the educational development and health and nutritional status of children under four is also presented. I. THE INITIAL EDUCATION CONCEPT 1.1. The concept of initial education for children is a newcomer to educational theory, and has been enlarged to include the early years and a variety of psychological and social elements. From a 19th century romantic concept of kindergarten limited to just a few children, early education efforts have diversified and, in more recent decades, have evolved Into a scientific strategy of education deemed necessary for all children. Although the earliest forms of young children's education, beginning in about 1800, were directed toward the children of the working class, the kindergarten developed by Friedrich Froebel (1782-1852), a German educator, provided an appeal to the middle class and was the prevailing model spread worldwide in the late nineteenth century (White & Buka, 1987). In 1837, Froebel created a special environment for the education of young children before they reached the ordinary school age (Boyd, 1952). Today, the idea of early education has been extended downward to comprise learning from birth. So understood, initial education has come to play a pivotal role in modern education and has started to make its presence visible in education systems worldwide. 1.2. The newly discovered value of initial education derives from its capacity, in general, to set the basic conditions for fiiture psychological, sensor-motor and social development of children and, in particular, to prepare their future behavior and performance at preschool and primary school, not to mention later stages of education attainment and labor productivity. A. Terminology 1.3. Initial education strategies revolve around four basic notions: child survival, growth, development and care (Myers, 1992). The complex meanings and interrelationships of these terms warrant clarification. 1.4. Sulvjyal is something more than simply avoiding death at an early age. Indeed, death seldom occurs instantly, and surviving children fall along a continuum running from near death through sickness to a state of satisfactory physical, mental, and social health and welfare. Mortality rates are still high in early childhood in most parts of the developing world, but as strategies have succeeded for reducing infant mortality, the importance increases of both attending to the quality of children's survival and to the effect of their balanced and healthy development (McGuire & Austin, 1987; UNICEF, 1991). 1.5. Growth occurs as cells are added to the body or as cells increase in size, and depends not only on the quantity and quality of the food a child eats, but also on how well that food is assimilated and used by the body. The measures of growth most commonly used are weight and height, both relatively easy to obtain, if compared with measures of social and psychological development, for instance. Growth is positively associated with greater potential productivity in adulthood, through Improvements in both physical and mental status and improved receptivity to education (Hill, 1990).  2 1.6. DP iimm , is not the same as growth, although the two are inter-related and often used interchangeably. Whereas growth is the increase in size, development is characterized by changes in complexity and function. Development therefore is a multi-dimensional process which includes: (a) a physical dimension, or the ability to move and coordinate; (b) a mental and cognitive dimension, or the ability to think and reason; (c) an emotional dimension, or the ability to feel; and (d) a social dimension, or the ability to relate to others. It is worth noticing that survival, growth and development are simultaneous (not sequential) processes. By attending to developmental needs it is possible to have a positive effect on survival and growth. 1.7. Care falls into a somewhat different category from the three previous concepts. Child care consists of the actions necessary to promote survival, growth and development. Caring for a child. means responding to basic physical, psychological, emotional and social needs. The basic needs of development go beyond protection, food, and health care to include the need for affection, interaction and stimulation, security provided through consistency and predictability, and play allowing exploration and discovery. 1.8. Initial education, in a very positive perspective, is mainly concerned with the integrated practice of child development and care, to enable participant children to reach their maximum human potentials. It supposes survival and involves growth, and has an effect on both, but the key purpose of initial education is to make the most of children's resources and to channel children into healthy and productive lives. B. Approaches 1.9. Initial education can be carried out through a diversity of strategies. Two basic forms have been discerned: formal and non-formal, each with a multiplicity of strategies at hand. Historically, the traditional early education institutions, especially kindergartens and nursery schools, have adopted a formal -- or norms- and curriculum-based - structure, similar to those observed at the primary level of educational systems. Such formal interventions have been portrayed as requiring special facilities, extra materials, a rich range of equipment, and above all a quality in teachers beyond that provided by normal teacher training institutions. 1.10. The past two decades, however, have witnessed an increasing consideration throughout the world of the special contributions that non-conventional initial education could provide to the formal school system, and could make to the early educational process, especially in light of evidence that conventional strategies have failed to reach the populations most in need, have lacked flexibility, and have been very costly. Non-formal strategies have the advantage of being closer to the populations who are the intended beneficiaries, involving parents and communities directly by drawing on their own culture, language and local know-how. Moreover, non-formal strategies allow for the development of early education programs without the necessity of heavy investments in physical infrastructure, professional staffing and support systems (Carvalho & Arango, 1984; Leighton & Ldpez, 1991). 1.11. The non-formal approach has assumed two basic forms: a center-based form which takes place in a collective place where community mothers volunteer to care for a small group of neighborhood children with eventual professional support; and a home-based form carried out through parental support and education. 1.12. A distinctive trait of the new trend toward non-formal initial education is its linkage with an earlier tradition of "community development" whereby local community groups come together around  3 their own perceived needs and organize to meet these needs, largely by drawing on their own strengths and idlosyncracies (Paul, 1989). Once small-scale experiments, such community Initiatives have grown to maturity and include regional and even national programs. Early childhood education of this sort has been a major area of action in countries as diverse as India ("Integrated Child Development Services," World Bank, 1990b; Colombia (the "Hogares Comunitarios" or Community Homes, World Bank, 1990a; Kenya (the national "Harambees" - meaning "Let's pull together" -- movement, Njenga, 1989; Mwiria, 1985); Brazil (the "Creches Comunitarias" or Community Nurseries, Franco, 1988); the United Kingdom (the "Playground Movement"); and Venezuela (the 'Hogares de Culdado Diario" or Daily Care Homes, Wor!d Bank, 1991b; Ministerio de la Familia, 1991). These projects tend to encourage autonomy and decrease unnecessary dependence on outsiders, providing services and education in ways that motivate parents to take control of their own growth and development, and make people feel that they are participants in the program rather than beneficiaries. These advantages have had ready appeal to planners, economists, anthropologists and finance ministries in societies looking for practical ways of achieving rapid human resource development. 11. THEORETICAL FRAMEWORK 2.1. Research concerning the key importance of early childhood education to learning and success in life for low-income children began a few years before the advent of the Head Start program in the United States (see para. 3.9), in a series of specially designed and controlled research projects. Later, evaluation studies were funded to study the impact of the Head Start program, and these two streams of work converged into a remarkable demonstration of long-term effectiveness (Hubbell, 1983; Lazar, 1989; Lazar & Darlington, 1982; Lazar et al. 1982). Their basic flnding is that early childhood education can improve the lives of low-income children and their families, and that it can enhance the quality of life for the community as a whole (Berrueta-Clement et al., 1984). It is important, however, to understand the scientific and socioeconomic bases upon which actual initial education programs are founded. A. Scientific Support 2.2. In the early 1960s, a working hypothesis was developed that human intelligence in general, and in particular the ability to do well in school, could be improved during the early childhood years. A growing body of empirical evidence in support of this educational hypothesis has been derived mainly from two sources: (a) studies conducted with animals and designed to test their "general intelligence" (Krech, Rosenzweig, & Bennet, 1960; Scott, 1962); and (b) psychological studies, particularly by Jean Piaget (1969) and his associates (Piaget & iaihelder, 1969) on the development of the thinking process in young children. 2.3. Heralded by Hebb's (1949) theorizing on the importance of early experience for intellectual growth, a series of studies examining the effects of early sensory restriction or stimulation on later performance were conducted in the 1960s and 1970s (Bronfenbrenner, 1968; Hunt, 1961; Rutter, 1974; Thompson & Grusec, 1970), all making evident that a sensorially restricted education impairs intellectual development. Conversely, studies providing specific kinds of sensory stimulation demonstrated that enriched experience -- in particular, in the areas of seeing and doing -- leads to intellectual enhancement in a variety of species (Held & Hein, 1963; Hymovitch, 1952; Meier & McGee, 1959).  4 2.4. For some characteristics there is as much quant.tative growth in a single year of a child's life as there is in eight to ten years at other stages in his/her development. During the nine months from conception to birth, growth is of the same absolute magnitude as it is during the nine years from three to twelve; and from conception to the age of about two and a half years, growth equivalent to that of half the full adult stature is attained. Hence, one would expect that it is during pregnancy and early infancy that the environment has the greatest effect, for better or for worse. 2.5. Cumulative research evidence in the fields of psychology, physiology, and nutrition continues to indicate that the most rapid mental growth occurs during infancy and early childhood and that, on the whole, early years are critical in the formation and development of intelligence, personality, and social behavior. The reasoning behind these conclusions is that because infancy is a period of unusually rapid maturation and sensitivity, a high degree of environmental stimulation is needed for the development of secure conceptual structures and social relationships in later life (Clarke, 1984). Among the more influential contributors in the research literature were Hunt (1961), who suggested a greater modifiability of development than had earlier been accepted; and Bloom (1964), who undertook a marathon analysis of longitudinal correlations for physical characteristics, intelligence, achievement, and personality. Among Bloom's conclusions were the following: "'When a number of longitudinal studies are compared with each other, and allowances are made for the reliability of the instrument and the variability of samples, a single pattern clearly emerges . . . Both the correlational data and the absolute scale of intelligence development . . . suggest that in terms of intelligence at age 17, about 50% of the development takes place between conception and age 4, about 30% between ages 4 and 8, and about 20% between ages 8 and 17 . . . These results . . . reveal the changing rate at which intelligence developed, since as much of the development takes place in the first four years of life as in the next 13 years." (pages 88-89) 2.6. This evidence was supported by the discovery that brain cells are formed during the first two years of life, and has been supported by the evidence that cognitive and social development are affected by the growth of brain cells and the development of neural connections. 2.7. Along the same cognitive approach, it is evidenced that to be able to decode words and even the simplest sentences children need two kinds of basic support, tiamely sensorial equipment and a set of psychological aptitudes. Cravioto & De Licardie (1973) present some studies showing the effects of performance deficits caused by malnutrition on specific visual and auditory abilities that are crucial for further learning. These studies are summarized (Selowsky, 198Ia) in Table 1. 2.8. Moreover, recent research has strengthened the theory on the temporal unfolding of the relationship between heredity and environment by showing that sensory stimulation from the environment affects the structure and organization of the neural pathways in the brain during the formative period. Individual variation is the product of a child's special biological makeup and the particular environment in which it struggles to survive, grow and develop. This theory invalidates the question of whether a personal trait is inherited or the product of experience, since a trait such as intelligence is now best considered as a reciprocal interaction between a genetic code controlling the biological unfolding of a child and environmental conditions (Sameroff & Chandler, 1975).  5 Table 1 Effects of Nutrition on Learning Authors Deficit in Performance | Probable Effect on ____________________________________________________ -_____________-______-Learning Cravioto & De Licardie (Mexico) Auditory-visual integration Reading ability Cravioto et al. (Mexico) Visual-kinesthetic integration Writing and drawing abilities Champakam et al. (India) Visual identificadion Reading ability Cravioto et al. (Mexico, Guatemala); Kinesthetic-visual, kinesthetic- General learning Guthrie et al. (Philippines) haptic, haptic-visual, auditory- abilities visual integration _ Source: Cravioto & De Licardie (1973); Selowsky (1981a). 2.9. Opportunities, therefore, to engage in complex perceptual and motor experiences at an early age positively affect various learning abilities in later life and are even able to offset, in part, the deficiencies linked to early malnutrition (Myers, 1992). By crying for communication purposes, by engaging in numberless playful activities, by imitating adults and their roles, by trying to interact with toys, brothers, sisters, peers and parents, children differentiate themselves from others, high from low, many from few, approval from disapproval. They lay "a foundation for lifelong tendencies toward trust or mistrust, self-acceptance or shame, initiative or passivity" (Gordon, 1964). Most importantly, they develop language and knowledge of symbolic or semiotic function, the very basis for reading and writing, communication and planning (Bandura, 1977). 2.10. Equally opposed to the notion of relatively fixed intelligence and aptitude, the work of Jean Piaget (1896-1980) quite directly supports and influences the importance of modifiable bebavioral processes in adaptive education. The stages of cognitive development described in the Piagetian theory of inte!ligence, starting with birth, are identified with major qualitative changes in the modes of thinking available to the child and, consequently, changes in the kinds of specific learning of which the child is capable. According to this theory, the origin of each individual's intellectual operations lies in the initial stage of that individual's sensorimotor development. The intellectual and cognitive stages described by Piaget thus provide individual modes of performance available to different children which would have to be considered in educational design (Piaget and Inhelder, 1969). From this perspective, the educational environment, particularly instruction, should accommodate the existing modes and processes of a learner; in such an adaptive environment, the power of instruction to influence a child's development is optimized. Adaptive educational environments can take advantage of the fact that children show great subtlety in adapting their cognitive competencies to different situations, if the situation permits such adaptability (Glaser, 1977; Piaget, 1947, 1969; Wittrock, 1977). 2.11. Parents are the most important influence in the environment of the developing child (Brim, 1965; Hess et al., 1971; White, 1975). Surveys show that some parents such as those of lower socioeconomic status have not been successful in meeting their responsibilities to young children (Goodson & Hess, 1976). New scientific knowledge about child rearing is available which, if disseminated, will effect behavioral change (Brim, 1965; Hess et al., 1971). A few studies suggest that the combination of child care and family influences best predicts the social development of the infant (Arango & Nimnicht, 1990; Howes, 1988; Howes & Olenick, 1986; Howes & Stewart, 1987).  6 Early education strategies, materials and methodologies must reflect zhese contributions to modern development. 2.12. Scientific research has indicated that, given the decisive influence of children's early defined physical, psychological and social equipment, primary school and even kindergarten (4-5 years old) can prove to be too late an opportunity to counteract some physical, neurological, psychological and social factors closely associated with early deprivation and lack of adequate stimulation (Brotherhood, 1977; Ibuka, 1990; Selowsky, 1981b). Summarizing what has become accepted currency in scholarly quarters, French biologist Rene Dubos (1967) wrote: 'Whatever be the influences that act on modern man, they have their most profound effect when they occur during the very early years of life, . . . because the human body and the human brain are incompletely differentiated at the time of birth and develop as the infant respo ds to environmental stimuli." (page 9) B. Socioeconomic Arguments 1. EMciency 2.13. One of the argunments to invest in early childhood development is based on the fact that investments in education, health, and nutrition early in life can bring a high return by increasing productivity in later years. The argument that improved childhood development increases the productivity of participating children over their lifetime comes from many fields. There is a long chain of causation that links early childhood development to later productivity in life; it begins with family and environmental characteristics affecting a child's early ability, which in turn affects later ability, educational and occupational attainment - all of which affect adult productivity. Currently there is enough empirical evidence to support this causal chain. Basically research findings indicate that a large part of the cognitive achievement differentials between lower socioeconomic and higher income groups can be attributed not only to malnutrition and lack of sanitation, but also to low levels of psychological stimulation among poor children. He also summarized the empirical evidence on earnings functions showing that the marginal product of additio..al schooling depends on the level of preschool abilities of the child. This implies that the productivity and effectiveness of future investments in schooling could be highly sensitive to present public policies to reach preschool age children in poverty. These actions, by increasing the level of early cognitive development of poor children could (a) increase the productivity of a given amount of schooling, and (b) generate additional benefits to the extent it induces additional schooling (Selowsky, 1981a). 2.14. Psacharopoulos (1982), based on socio-psychological theories, argues that early education (in the form of kindergarten) together with family characteristics affect early ability. Based on Bloom's reports that 50% of the ability level of an individual is built by the age of 4, and that the ability level increases by 60% between the ages of four and six (Bloom, 1964), as well as on findings by Jencks et al. (1972) and Pagerlind (1975), Psacharopoulos proposes a causation model to explain the relevance of early education on earnings later in life. The model, displayed in Figure 1, assumes an effect from family characteristics and early education on early ability, which in turn affect late ability and educational attainment. A higher level of educational attainment can lead to a better occupation and finally to higher earnings over a lifetime. In summary, early education indirectly has an effect on earnings. 2.15. In the case of education, during the 1970s and 1980s a body of research began to accumulate showing that attainment of even a few years of schooling is associated with important changes of  7 Figure 1 FROM FAMILY BACKGROUND TO ADULT EARNINGS 9 Early Childhood School Adult Life Note: F=hFarny Background; K=Wndetgatten; A=Abjl;ly; ED=Educafion; OCC=OccuPalion. Srmiaw r. (3wqxwia. The Eooonmos ni Ey-ay Chldhnod EduArnMbn ad Day Carm. IntHatialoanl Reafew of Educaoon. 28:53 70. 1082. economic value in individual skills (Schultz, 1963). Schooling helps build such skills as the ability to organize knowledge into meaningful categories, to transfer knowledge from one situation to another, and to be more selective in the use of information (Rogoff, 1980; Triandis, 1980). Schooling facilitates greater technological adaptiveness (Grawe, 1979); it relates directly to both increased farmer productivity (Lock)'eed, Jamison and Lau, 1980; Moock, 1981) and productivity in the informal market sector (Colclough, 1980). In general, investment in education yields an economic return comparable to capital investment; in particular, primary education (associated with early stimulation and sensori-motor readiness) yields a higher rate of return on the money spent than does secondary or higher education (Psacharopoulos, 1985). 2.16. Exploring the concept of child and family readiness, Myers (1992) develops and illustrates (see Figure 2) a model for relating early childhood development to schooling and later achievements. The term readiness is commonly used to mean child's readiness to learn to read. However, children's general social development and intellectual backgrounds should also be taken into account in any consideration of ways to help children prepare for school (Kagan, 1990; National Association for the Education of Young Children, 1990). This orientation revitalizes consideration of the "whole child." Rather than short-term goals (achievement in subject areas, for instance), long-term goals (such as creative and critical thinking skills) must define the content of the programs (Williams, 1989). Thus, Myers' model projects both child and family readiness into school progress and further into a wide spectrum of adult readiness and social readiness.  8 Figure 2 ~~~x I CL. z , I.. 2~~~~~~~~~~~~~~~~~~I X N g u 3 8 ~~~~~~~~~cnva 5 b H! cr CS) 0~~~~~~~. a Cs 0.~~ !: ~ ~ ~ da n eq _ .: . __ o 09 X-~~~~~~~~~~~~~~~~~~~~~~~~~0 , a I __ 2.17. Another way in)which investments in initial education can bring an economic i IN ~ ~ ~ ~ ~ I through cost savings - by reducing work losses, by cutting the later need for social welfare programs, by improving the efficiency of educational systems through reduction in dropout, repetition, and remedial programs, and by reducing the needs for hea!th financing, law enforcement and crime prevention (Myers, 1992). Figure 3 illustrates how poverty and delinquency are negatively related to othier key variables in thie model. Thus, poverty depresses intellectual performance; but strong commitment to schooling and favorable scholastic placement make delinquency less likely.  9 Figure 3 TRANSACTIONAL MODEL OF INITIAL EDUCATION * INITIAL EDUCATION COMMITMENT -_ DELINQUENCY TO SCHOOUNG INTELLECTUAL PCrOHMAroR cc POVIRIY - SCHOLASIIC _... SCHOLASIIC _... SCHOLASIIC tMPLOYMENI PLACEMENT ADOLESCENCE ATTAINMENT EAllLY CHILDHOOD CHILDHOOD ADOLESCENCE ADULTHOOD ' The rwniple was drawn from a iinIvem nl rchiidren 3 4 years of Hge. " Variables neoaaveIy related to the other vauiables in the model. Source: Berruela-ClemenL el al. (rnand Lives, The EIreI- of lh Pg(U Sc,hool Proaram on Youths through Age 19. Ypsianti Mghgean: Ihe HighScope Press. 1984. 2.18. Two specific examples drawn from evaluations of early childhood development programs demonstrate how cost savings can be effected. In the first example, results from an evaluation of the U.S. Perry Preschool Project (see para. 3.5 ff and Box 1, page 12) provide unusually solid empirical evidence that an early intervention raising a child's ability level can have effects on school performance, on the demand for schooling, and on employment and earnings. When the preschool children were compared with another group of children who did not participate in the program, it was shown that the early education group was more likely to graduate from high school and obtain employment and less likely to require remedial education, be in trouble with the law, or become pregnant while a teenager. At age 19, 45% of the members in the program group were supporting themselves as compared with 25% of the control group. The mean annualized welfare payments to members of the program group was US$633, less than half the amount received by the members of the control group, US$1,509. Members of the preschool group also had higher (59% versus 32%) employment rates. The value of benefits both to participant individuals and to society far exceeds the initial cost of the program. In the case of the Perry Project, it was estimated that a US$1 investment returns US$6 in taxpayer savings due to lower education costs, lower costs of public welfare, and higher worker productivity (Berrueta-Clement et al., 1984). 2.19. The second example comes from Brazil, where the evaluations of the PROAPE (Ministerio da Sadde, 1983) showed that by reducing the extra primary schooling cost associated with repetition, a program of integrated attention to preschool children more than paid for itself. This evidence of cost savings is corroborated by sparse but revealing evidence from less developed countries. Myers' (1992) review of 17 longitudinal evaluations examining the effect of early interventions reveals that children who participated in early childhood programs experience lower repetition rates in primary education. Although difficult to calculate, assessments of rates of return for early education programs are supportive of such interventions. Using Latin American data, Selowsky (1981a) concludes that:  10 Yearly investment per child in programs that can induce a change in ability equal to one standard deviation can be justified if they cost between 37% and 51% the yearly wage of an illiterate worker (page 342). 2.20. In terms of health expenditures, cost savings as a result of early child development programs can also be effected. On average, an estimated 70% to 85% of the developing world's total health spending, both public and private, goes for curative care. Between 10% and 20% is spent on preventive care and the remaining 5% to 10% on community services. Within the curative sector, hospitals often account for more than 80% of the cost. Yet it is well known that preventive and community services are far more effective in reducing morbidity and mortality (World Bank, 1990c). Through the education of parents and early care of children, initial education programs aim to increase the use of public health care, thereby preventing more expensive treatments. 2.21. In short, the investment in initial education has a clear economic return because it: (a) paves the road for a more efficient primary education level; (b) makes a significant difference to future adults' productivity and incomes; and (c) reduces the present and future costs of health and education and other public services, including lower repetition and dropout rates at school. Basic education for every child is also a fundamental prerequisite for environmentally sound development in the years to come. The choices that today's children will have to make, whether they be choices about family size or land use, energy source or waste disposal, can only be made if they are capable of absorbing new information and wisely responding to its consequences. 2. Sodal Equity 2.22. Poverty and discrimination generate unfair conditions and unequal treatment that inhibit human development. During the past three decades the developing world has made enormous economic progress. Against the backdrop of achievement, it is all the more staggering that over one billion people in the developing world are living in poverty. This is the number of people who are struggling to survive on less than US$370 a year (World Bank, 1990c). 2.23. Poverty is by far the most discriminatory of all social factors, but it frequently combines with other elements to exacerbate the plight of socially underprivileged children. This is the case of most rural and urban-marginal children who are forced to live in precarious environments, where education, health and sanitation services are either remote or altogether nonexistent. This is especially true for girls: their primary enrollment rates are less than 50% in many African countries. Initial education increases the chances for poor children to obtain their just place in society, especially those living in rural and urban-marginal areas, females and children at risk by addressing their educational, health, sanitary and environmental needs (see paras. 3.12-3.30). 2.24. Any attempt to overcome or prevent the devastating effects of malnourishment and the psychological impairment of people must necessarily include an educational intervention. Education has been viewed as the key influence in improving the conditions of children at risk, and is considered as offering good prospects for narrowing the gap between at-risk populations and the rest of society. There is ample evidence that investing in human capital, especially in early education, also attacks some of the most important causes of poverty. As one specialist expressed it, 'the principal determinants of any real opportunity in educational, occupational, and social terms are for most laid down in the years between birth and five or six years of age" (Heron, 1979). Great emphasis has been placed on strategies and interventions to avoid school failure and better prepare children for successful school performance and thereby create a foundation for life success.  11 mU. EVIDENCE FROM INITAL EDUCATION PROGRAMS 3.1. Over the past 30 years studies have been conducted on the effects of early childhood programs on primary schooling and other developmental variables. Early findings of a rather skeptical tone (Cicirelli, et al., 1969) have lately been superseded by a growing body of long-term studies which suggest a more positive influence of such programs on later indicators, confirming the hypotheses discussed in the preceding section. 3.2. The last two decades have seen an accumulation of evidence from many countries, both in the industrialized and in the developing world, which support the case for early childhood education and care. The argument has passed from an examination of whether early childhood education can pay off, to a much more complex argument about the form of early childhood attention that provides the best results a given society can afford. 3.3. Experiences in both developed and developing countries have demonstrated the positive effects of early education on children's mental development, school enrollment, better school achievement and grade completion, all of which enhance the educational efficiextcy of and economic return on primary schools. Following is a discussion -- based largely on Myers (1992)- of some of the evidence from around the world regarding the positive effects of initial education. The first section discusses experience under several programs in industrialized nations, while the second reviews several early childhood development programs in developing countries. A. Evidence from Industrialized Nations 3.4. In the 1980s, a set of evaluations looked at children in early or late adolescence, and showed that participation in well-implemented early childhood education programas can have significant effects on progress through the school as measured by promotion; needs for special education, and high school completion (Halpern & Myers, 1985; Lazar, 1989). The evidence provided by these studies shows positive correlations between participation in preschool programs and school readiness, higher achievement in terms of initial primary school grades, higher educational attainment, and more likelihood to graduate from high school, to obtain employment and to be a self-supported young adult. This section presents details of some of the key findings from such evaluations. 3.5. Perry Preschool Study, United States (see Box 1). Given the relatively recent introduction of early childhood education in developing countries and the complexity of implementing multi-year longitudinal studies, no long-term evaluations of developing country programs exist. Consequently, this section draws upon the Perry Preschool Study enacted in Ypsilanti, Michigan, which is the principal source of information on the subject (Berrueta-Clement et al., 1984). Between 1962 and 1967, the program provided two years of preschool, coupled with weekly home visits, to approximately 60 poor black youths, aged 3-4. Information on the participants, as well as on a 60 member control group, was collected annually while the children were between years 3-11 and again at ages 14, 15 and 19. 3.6. The results of the study, summarized in Table 2, indicate that the Perry Preschool program had significant positive long-term benefits not only for the participants (higher secondary school performance and income generating skills) but also for society in general Oower welfare payments and more law abiding citizens). On average, participants enjoyed a high school grade point average 20% greater than non-program students. Participants were also about 50% more likely to finish high school, and 80% more likely to attend college or vocational school.  12 _cx . . LONG-TERM AE TS-OF RLY CHLDHOOD EDUCATIO tw .:, : T: E PEtRRY PRESCHO POO OPRAM:: TheXic Pe0ty ?esoloWol studyts bs n p of ey oni lives of lowime ci whot wenit risk ofschool fallur Bewn 1961 ad l967 $.8 cilde. age 34 fron. YpnlanI 8Mihgan received wh yer of o .. . dy wi : wel h' :,on o i patticipnts, and a control grou was:collec annually whiletecildren w n yeors 31 antdg at ages 14,1 and 19. , pm i4a call tss,however tc iferenc between stdy rups imnihe oerti ndhr secndgrde was nolne ttsiclyugicn N.eveh leas eal ducaton led to inresd ct.aiocieaemas measured . by andd d si sb e t andmiddes chool Tchildten' sci and emo, onala,,e k m al vored preschool group. Bys who ha precool w4erele ikyt rhav, beenplcdin s:pecial education or retained than tdi countpoup'uiyp