38191 ABBREVIATION A-NET Advanced National Educational Test BMR Bangkok Metropolitan Region CGD Comptroller Generalûs Department CYS Children and Youth Survey DGE Department of General Education ESAs Education Service Areas FDI Foreign Direct Investment GAT General Aptitude Test GDP Gross Domestic Product GER Gross Enrollment Rate GNI Gross National Income GPP Gross Provincial Product HSEP High School Equalization Policy ICL Income Contingent Loan IEA International Association for the Evaluation of Educational Achievement LGO Local Government Organization LLECE Latin American Laboratory for the Assessment of Quality in Education MDGs Millennium Development Goals MOE Ministry of Education MOI Ministry of Interior NEA National Education Act NFE Non-Formal Education NER Net Enrollment Rate NESDB National Economic and Social Development Board NIETS National Institute of Education Testing Services NSO National Statistics Office OBEC Office of Basic Education Commission OEC Office of Education Council OECD Organization for Economic Co-operation and Development ONEC Office of National Education Commission ONESQA Office of National Education Standards and Quality Assessment O-NET Ordinary National Educational Test ONPEC Office of National Primary Education Commission PASEC Programme dûAnalyse des Systemes Educatifs de la CONFEMEN PIRLS Progress in International Reading Literacy Study PISA Programme for International Student Assessment , , PROGRESA Programa Nacional de Educacion in, Salud y Alimentacion RTG Royal Thai Government SACMEQ Southern and Eastern African Consortium for Monitoring Educational Quality SAT Scholastic Aptitude Test SES Socio-Economic Survey SPR School Participation Rate TIMSS Trends in International Mathematics and Science Study UNESCO United Nations Educational, Scientific and Cultural Organization UNICEF United Nations Childrenûs Fund ThailandSocialMonitor:Improving SecondaryEducation 3 ACKNOWLEDGEMENTS The Thailand Social Monitor series Thomas (Sector Manager) and Tamar was conceived as a tool to reflect on Manuelyan Atinc (Acting Sector Director). current situations about selected social The team wishes to especially thank issues in Thailandûs health, education and Khunying Kasama Varavarn (Permanent social protection sectors. The first Social Secretary) and the Ministry of Education Monitor, Challenge for Social Reform, was staff for their inputs and invaluable com- launched in 1999 in response to the 1997 ments as well as to Ana Revenga and economic crisis. To date, five Social Khuankaew Varakornkarn for the initial Monitors have been issued. development of this Social Monitor. The team benefited from background papers This issue of the Social Monitor focus- prepared by Dilip Parajuli and Deon es on Thailand secondary education. The Filmer, Niels-Hugo Blunch, and Ana Monitor first attempts to highlight recent Revenga and Chaiyuth Punyasawatsut. accomplishment in the advancement of The team would also like to extend its Thailand secondary education. It appreciation to the peer reviewers, addresses the issues of access and equi- Ernesto Cuadra, Daniel Mont, Carmen de ty, quality and efficiency and considers Paz Nieves, Omporn Regel, Kaspar Richter, challenges Thailand is facing with regards Norbert Schady, Christopher Wheeler and to secondary education. Finally, it pro- Charlotte Vuyiswa McClain-Nhlapo for vides some recommendations for policy their insights and suggestions to strength- consideration. en this study. Finally, the team would like to thank Keiko Inoue, Achariya This Social Monitor was led by Luis Kohtbantau, Rachadawan Pasugswad Benveniste under the overall guidance of and Juliana Williams for excellent research Ian Porter (Country Director), Christopher and administrative assistance. 4 CONTENTS Abbreviation 3 Acknowledgements 4 Contents 5 ExecutiveSummary 8 Background to Education in Thailand 9 Access to and Equity in Secondary Education 9 Quality of Secondary Education in Thailand 10 Efficiency of the Thai Education System 11 Addressing the Next Generation of Challenges 12 I.EducationinThailand: AnOverview 14 Introduction 15 Education Reform in Thailand: Historical Context 15 Bridging Basic Education, Tertiary Education and the Labor Market 18 A Key Tool for Closing Equality and Equity Gaps 23 II.AccesstoandEquityinSecondaryEducationinThailand 26 Access to Secondary Education 27 Equity in Secondary Education 31 Reaching out to the Excluded 36 Interventions to Stimulate Equitable Secondary School Expansion 44 III.QualityofSecondaryEducationinThailand 52 Quality of Thai Education: Lessons From International Data 53 Factors Affecting Quality of Education in Thailand 59 Teacher Characteristics 59 School Characteristics 61 Household and Individual Characteristics 64 Quality Assurance Mechanisms in Thai Education 66 IV.EfficiencyoftheThaiEducationSystem 70 Public Spending on Education 71 Household Education Expenditure 77 Raising Efficiency of Secondary Education Financing 80 V.AddressingtheNextGenerationofChallenges 84 Increasing Access And Equity 85 Improving Quality 86 Assuring Efficiency 88 References 91 ThailandSocialMonitor:Improving SecondaryEducation 5 FIGURES Figure 1.A: Primary School Participation Rate by Income Quintile, 1994-2002 19 Figure 1.B: Education Transition Patterns 20 Figure 1.C: Comparison of Secondary Gross Enrollment Rate Trends, 1970-2000 21 Figure 1.D. Secondary Participation Trends by Income Quintile, 1994-2002 22 Figure 2.A: Thailand Educational Attainment Profiles for Ages 16 to 19, 1994-2002 28 Figure 2.B: Transition of Three Cohorts from Grades 1 to 12, 2001-2003 30 Figure 2.C: School Participation Rates, Gross Enrollment Rates and Net Enrollment 32 Rates by Sex and Gains, 1994-2002 Figure 2.D: Grade Completion by Urban/Rural Location, 1994 and 2002 32 Figure 2.E: Secondary Net Enrollment Rates by Region, 1994-2002 33 Figure 2.F: Survival Curve Estimates for Ages 6 to 15 in Thailand, 1994 and 2002 34 Figure 2.G: Trends in Secondary Gross Enrollment Rates across Asian Countries, 1980-2003 35 Figure 2.H: Percentage of Upper Secondary Graduates to the Population, 2003 36 Figure 2.I: Reasons Cited for Dropping Out, Grades 7-12, 2004 39 Figure 2.J: Reasons for not Continuing Lower and Upper Secondary Education, 2002 40 Figure 2.K: Average Annual Expenditure on Education per Person by Education Level, 2002 42 Figure 2.L: Share of Children and Youth at School or Work by Age, 2002 43 Figure 2.M: Ratio of Number of Students Attending Public to Private Schools at Secondary 49 Level,1992-2002 Figure 3.A: PISA 2003 Test Score Results in Mathematics Literacy by Income Group 55 Figure 3.B: PISA 2000 Test Score Results in Reading Literacy by Income Group 55 Figure 3.C: Trend Line of PISA Test Scores against Log GDP per Capita 56 Figure 3.D: Trend Line of TIMSS Mathematics Scores against 2003 GNI per Capita 56 Figure 3.E: Trend Line of TIMSS Science Scores against 2003 GNI per Capita 57 Figure 3.F: Average Mathematics Performance by Wealth, 2000 57 Figure 3.G: Between- and Within-School Variation in Mathematics Scores, 2000 58 Figure 3.H: Student Teacher Ratio and Class Size in Secondary Schools by Region, 2002 61 Figure 3.I: Cross-National Comparison of Average Mathematics Achievement and Class Size, 62 1999 Figure 3.J: Perceived Adequacy of Physical Infrastructure and Educational Resources 63 Figure 3.K: Mean Scores by Level of Mothersû Education 65 Figure 4.A: Share of Education Budget by Spending Category, 1997-2004 72 Figure 4.B: Sources of Revenue for Local Governments, 1997-2004 75 Figure 4.C: Incidence of Public Expenditure across Income Quintiles by Education Level, 2002 76 Figure 4.D: Distribution of Public Spending on Education by Level, 2000 and 2002 77 6 TABLES Table 1.1: Incidence of Public Spending for Primary Education by Income Quintile, 2000 19 Table 1.2: Incidence of Public Spending for Secondary Education by Income Quintile, 22 2000 Table 1.3: Gini Coefficient by Country and Share due to Differences in Educational 24 Attainment of Household Head Table 2.1: Average Years of Educational Attainment, 1999-2003 29 Table 2.2: Age Distribution by Grade, 2002 26 Table 2.3: School Participation Rate by Income Quintile and Provincial per Capita 34 Income, 2002 Table 2.4: Number of Disabled and Special Students by Type and Gender, Academic 38 Year 2004 Table 2.5: Private Expenditure Estimates on Education by Income Quintile (Real Baht), 41 1994-2002 Table 3.1: Percent of Teachers with Masterûs Degree or Higher, 2002 59 Table 4.1: Education Budget, 2000-2005 71 Table 4.2: Education Budget Allocation and Student Enrollment by Spending Category, 73 2002 Table 4.3: Total Secondary Education Expenditure as Percent of GDP by Source of 74 Funding, 2003 Table 4.4: Per Capita Educational Expenditure by Region (Baht), 2002 76 Table 4.5: Household Expenditure on Education by Region, 1994-2002 78 Table 4.6: Household Expenditure on Education by per Capita Income Quintile, 1994-2002 79 Table 4.7: Utilization Ratios by Education Level, 2002 82 BOXES Box 2.1: Educational Enrollment-GER or NER? 29 Box 2.2: The Path to Universal Secondary Education in Korea 37 Box 2.3: Summary of Government-Initiated Interventions 45 Box 2.4: Demand Side Financing Mechanisms 50 Box 3.1: Measuring Quality of Education across Countries 59 Box 3.2: The Office for National Education Standards and Quality Assessment 67 Box 4.1: Private Participation in Education: Examples from Korea, Africa and Chile 77 ThailandSocialMonitor:Improving SecondaryEducation 7 EXECUTIVESUMMARY Secondary education has the potential to labor market. Yet secondary education serve as a pathway for studentsû progress can also act as the main bottleneck pre- and advancement. It can offer skills devel- venting the equitable expansion of educa- opment for producing a workforce with tional opportunities. Thailandûs secondary expertise that matches the needs of the education is currently at a crossroads, with labor market. Indeed, investment in sec- the potential to improve opportunities for ondary education reaps great rewards young people or become a binding con- when it acts as the critical link between straint to economic growth and competi- basic education, higher education and the tiveness. It is thus an opportune time to 8 take stock of recent accomplishments in the development. Decentralization of educa- advancement of Thailandûs secondary edu- tional management has been implemented cation, as well as consider the challenges through the creation of Education Service that lie ahead. Areas (ESAs) that are meant to increase community-level participation. Chapter 1 provides a background to edu- cation in Thailand, including the historical During the 1980s, the RTG focused on context of reform and recent trends at the expanding primary education. Such efforts primary and tertiary levels. Chapter 2 are reflected in the achievement of near reviews the current state of the Thai second- universal primary education, regardless of ary education system as a whole and income, geographical location or sex. across different groups of the population. During the same time period, however, sec- Additionally, comparisons are offered across ondary education enrollment lagged. regions. Based on these results, an analysis Efforts by the Thai government to expand of the demand side of secondary educa- access at the secondary level have pro- tion is explored, such as the effects of duced dramatic growth in secondary enroll- household decision-making in sending chil- ment in the 1990s and onwards. In 1997, 70 dren to school. Chapter 3 looks into the percent of the total labor force had only issue of the quality of secondary education received elementary education or less, while and suggests possible avenues for improving 17 percent had obtained secondary educa- the quality of educational services in tion and 8 percent had a university degree. Thailand. Chapter 4 investigates options to But investments in secondary education in enhance efficiency in the utilization of finan- the 1990s began to pay off. By 2004, the cial resources. Finally, policy recommenda- labor force with more than primary educa- tions are suggested in Chapter 5, taking into tion reached close to 40 percent. While consideration the goals of the Royal Thai recent achievements put Thailand ahead of Government (RTG) and the distinct historical most East Asian countries, it still falls behind 1 evolution of its education system. Asian Tiger and Organization for Economic Co-operation and Development (OECD) BACKGROUNDTOEDUCATIONIN countries. InorderforThailandtotakethe THAILAND next step in advancing its secondary edu- cation system, a better understanding of Education has been a priority area since how key issues such as access, equity, qual- Thailand shifted to a constitutional monar- ity and efficiency play out in the Thai con- chy in 1932. National Education text is crucial. Development Schemes (NEDSs) have guided major education reform strategies and iden- ACCESSTOANDEQUITYIN tified targets and priority areas at the SECONDARYEDUCATION national level. Most recently, the 1999 National Education Act (NEA) and the 2002- The RTG has set a goal in its first national 2016 National Education Plan have respec- Millennium Development Goals (MDGs) tively raised compulsory education from 6 to report to achieve universal lower secondary 9 years and introduced a balanced focus education by 2006 and universal upper sec- on both human-centered and economic ondary education by 2015. Both gross and 1 Asian Tiger countries refer to Hong Kong, SAR, Singapore, Taiwan and South Korea. ThailandSocialMonitor:Improving SecondaryEducation 9 net secondary enrollment rates have vital in order to address access and equal- increased significantly in the past 15 years. ity gaps in Thailand. A cost benefit analy- Currently, secondary education is at reach sis shows that the main reason preventing for most children. This is a result of a con- children from attending or staying in schools certed effort from the RTG to redress is lack of financial support. Educational inequities in education participation. cost poses a larger burden to poor families, who must contribute a greater share of the Equity in access to secondary education household income after paying for food has improved. Participation rates in second- and other basic needs. The opportunity ary education demonstrate that access costs of forgone income are compounded across gender, the rural-urban divide, for these families as well. In addition, while regions and socioeconomic status groups the RTG committed to provide 12 years of has expanded, benefiting all groups in some free education, non-tuition costs such as measure. However, while the urban/rural library fees, exam levies, meals and trans- gap has grown smaller, on average the portation serve as substantial financial gender gap has grown significantly larger obstacles for many poor families. Despite and to the benefit of girls. It is also worthy high rates of return to education in of notice that barriers to secondary school- Thailand, many poor families are still unable ing narrowly due to household income to meet the direct or opportunity costs of remain an important concern in Thailand. sending their children to secondary school. There are large differences in secondary school enrollment between the poorest and Thailand has implemented several key poli- the richest population groups. Although cy interventions in an attempt to boost sec- these differences have narrowed over time, ondary enrollment. Such interventions they have remained quite substantial. The include loan and lunch programs, scholar- gap in school participation between the ships and a bicycle lending project for rural wealthiest and the poorest population quin- students, which have indeed contributed to tiles in 1994 was 24 percent and stood at the recent surge in secondary education 17 percent in 2002. Finally, international access. However, the interventions have comparisons show that Thailand made dra- fallen short of meeting the needs of very matic improvements over the past two poor families, whose children continue to decades. Thailand started out in the late drop out of the education system. In par- 1980s with one of the lowest secondary ticular, targeting disadvantaged students at gross enrollment rates in the region, but sub- the lower secondary school level will ensure sequently picked up pace in later decades a higher transition rate from primary to sec- to position itself in the top tier due to a ondary level. concerted Government effort to expand access. QUALITYOFSECONDARY EDUCATIONINTHAILAND But despite important gains, much work remains to be done. While 98.6 percent of An area of great concern for Thailandûs children were estimated to complete pri- secondary education system is student mary school in 2002, only 88 percent trans- achievement. An important conclusion ferred to lower secondary and 69 percent stands out from benchmarking the perform- continued to upper secondary. ance of Thai students internationally: Understanding both demand and supply Thailand has higher scores than other coun- side factors to educational attainment is tries at similar income levels, suggesting that 10 it has been generally successful at providing higher student achievement. Greater educational services of certain quality equi- investments in basic school resources in tably. And not only Thai students on aver- order to provide a minimum set of materi- age perform well relative to their peers in als for effective use by teachers to support other countries at similar income levels, but instructional content could well buttress stu- the distribution of knowledge across dent performance in Thailand and translate Thailand is fairly equitably distributed. into higher learning outcomes. Socioeconomic status accounts for a mod- est share of the total variation in student EFFICIENCYOFTHE achievement scores overall. THAIEDUCATIONSYSTEM Yet, while the Programme for International The RTG allocates more than one fifth of its Student Assessment (PISA) and the Trends in total budget on education, which accounts International Mathematics and Science for about 4 or 5 percent of the countryûs Study (TIMSS) results suggest that Thailandûs Gross Domestic Product (GDP). This level of performance is acceptable given its income allocation was maintained even through the level and knowledge is rather fairly distrib- economic crisis in the late 1990s. More than uted, they also underscore that there are two thirds of the education budget is allo- urgent problems with education quality that cated to basic education, with pre-primary demand urgent attention. Very few Thai and primary levels receiving the largest pro- children score in the top proficiency levels. portion. Thailand allocated approximately Furthermore, a very large share of students 28 percent of its total education budget is performing below acceptable proficiency (1.13 percent of GDP) to secondary educa- levels. Thailand had roughly 40 percent of tion in 2003, falling behind what countries students performing at or below the PISA with strong secondary education sectors level one in literacy and over 50 percent of typically spend. Lower-middle income students performing at or below the PISA countries allocated on average 40 percent level one in mathematics. This contrasts of their total education resources (1.86 per- greatly with the upper income countries cent of GDP) to the secondary level. Most where only around 10 percent of students strikingly, the unit cost for secondary educa- score at or below level one. In summary, tion in Thailand is lower than the unit cost a vast proportion of students are function- for primary education. The limited resourc- ing at or below the most basic level of lan- ing of secondary education through public guage, mathematics and science ability. financing is further compounded by very low shares of private resources coming to What factors affect student achievement? secondary education. In Thailand, private Given Thailandûs low levels of academic sector contributions amounted to 0.06 per- achievement in absolute terms, policies to cent of GDP, an equivalent of 5 percent of raise the overall performance level of the the public sector financing. general student population are imperative. Teacher quality is an important factor. Local governments are reliant on subsidies Enhancing teacher professional develop- from the central government to finance ment could potentially translate into signifi- education. The RTG has encouraged the cant improvements in student flows and decentralization of educational manage- learning. There is also a shortage of ment in order to improve local participation resources for learning in Thai schools and and ownership. Decentralization also this is generally perceived as a constraint to includes increasing the share of local gov- ThailandSocialMonitor:Improving SecondaryEducation 11 ernment resources spent on education. ADDRESSINGTHENEXT However, most local agencies still rely on GENERATIONOFCHALLENGES subsidies from the central government to finance education. In general, resources In order to include the excluded and from local government only comprise 20 to increase school participation, efficient data 30 percent of education spending. collection and analysis is a prerequisite. While education receives the largest share Data and management information systems of the national budget across sectors, should be able to timely estimate children whether those resources are equally and outside the formal education system, efficiently distributed among different enabling the design of suitable programs income groups is debatable. The poorest and providing strong evidence for policy 40 percent of the population receives 56 decision making. For instance, although the percent of total spending, reflecting a pro- gender gap in secondary education parti- poor allocation of resources. The quintile cipation has been increasing, to the bene- distribution for secondary level is distributed fit of girls, there is a lack of clear under- relatively equally; however, spending for ter- standing about the reasons that are driving tiary education is clearly regressive. The this phenomenon and virtually no policy dis- wealthiest 20 percent of the population cussion as to how to redress this situation. receives 53 percent of total spending. The role of alternative education service The NEA articulates a vision for free basic modalities can be strengthened. Existing education during 9 years of compulsory programs, although encouraged by the education. In addition, it proposes ambi- NEA, are small and do not seem to satisfy tious targets on education service provision, the potential demand. Flexible education including a student teacher ratio adjusted arrangements can play a key role in provid- to 25:1, a new teacher compensation struc- ing opportunities to disadvantaged children ture, increased and better integrated use of in accessing secondary education. Greater information technology, and additional flexibility in terms of learning sites, class funds to encourage more children to enroll schedules and curriculum can provide a in the system. These are worthy goals. But more suitable environment to fit the needs these commitments require substantial fund- of children who cannot participate in tradi- ing upfront, either from the public or the pri- tional school settings, such as rural migrant vate sectors. Current spending in second- workers. ary education is insufficient to fulfill these commitments. Raising the cost effectiveness In order to improve student outcomes, inter- and performance of the education sector nal and external quality assurance mecha- will be imperative to enhance outcomes nisms have been established. The Ministry and contain costs. Without serious consid- of Education (MOE) has adopted a cautious erations to efficiency measures, many of the approach to manage the çstakesé of this commitments promulgated under the NEA assessment, in order to encourage school could remain elusive. staff to approach it openly as an opportu- nity for self-improvement. On the one hand, recommendations made by external evaluators must be tangible and achiev- able. And the quality of external independ- ent evaluators itself needs to be monitored and evaluated for the process to be mean- 12 ingful. On the other hand, school staff will uncertain and, according to current esti- need to demonstrate action to turn around mates, likely to be low. At present, the pri- ineffective practices in order to make a dif- vate sector plays a small role in general ference in student performance. secondary education, accounting for 11 Mechanisms to provide systematic rewards percent of student enrollments in lower sec- for improvements in academic or institution- ondary and 20 percent in upper secondary al outcomes could be weaved into the cur- education. Its overall share has either rent system. Performance-based incentives remained largely stagnant or diminished could provide the necessary impetus to fuel over the past decade. In terms of financial administrative and instructional behavioral contributions, the private sector accounted changes. for approximately 5 percent of overall Significant citizen participation is expected domestic secondary education resources. to take place in the management of ESAs. Mobilizing private resources can be an Additionally, pilot programs examining differ- important source of secondary education ent types of school boards to explore ways financing and could free up public to enhance the relevance and responsive- resources for improved targeting to disad- ness of education service delivery to local vantaged populations or service delivery needs are under way. As administrative quality enhancements. and service delivery functions are being devolved, a strong accountability system Thailand has achieved remarkable improve- must be actively nurtured in order to foster ments in education secondary provision and a service-oriented culture that is responsive participation. Much has been accom- to local aspirations and needs. plished in the last decade. The RTG has now embarked on finding solutions to the The NEA specifies that the financing system next generation of challenges: consolidat- will be restructured by providing block ing equitable access, improving quality and grants to ESAs and schools on the basis of enhancing efficiency. An ambitious educa- a standard capitation formula in addition to tion reform program is gathering momen- other per capita top ups according to tum. Dedicated efforts and a continued poverty levels and other provisions for disad- focus in redressing existing systemic imbal- vantaged students. ESAs will also be ances can realize the potential to fulfill the responsible for raising additional funding, but goal of a high quality universal secondary the level of local revenue generation is education for all Thai children. ThailandSocialMonitor:Improving SecondaryEducation 13 I.EDUCATIONINTHAILAND: ANOVERVIEW 14 INTRODUCTION EDUCATIONREFORM INTHAILAND: Secondary education has the potential to HISTORICAL serve as a pathway for studentsû progress CONTEXT and advancement. It can offer skill deve- lopment for producing a workforce with Since Thailand shifted from an absolute to a expertise that matches the needs of the constitutional monarchy in 1932, education- labor market. Indeed, investment in sec- al policy has been regarded as an integral ondary education reaps great rewards component of national development plan- when it acts as the critical link between ning. The first NEDS was devised the same basic education, higher education and the year, formally recognizing every individualûs labor market. Yet secondary education educational ability. During the NEDS of can also act as the main bottleneck pre- 1960-1976, the RTG pledged compulsory pri- venting the equitable expansion of educa- mary education, with special provisions tional opportunities. Thailandûs secondary made for children with disabilities. The pri- education is currently at a crossroads, with mary school dropout rate was as high as 60 the potential to improve opportunities for percent in the 1960s, and secondary enroll- young people or become a binding con- ment was only 2 percent of the age group straint to greater economic growth and (World Bank 1998). The NEDS of 1977-1991 competitiveness. It is thus an opportune changed the structure of the education sys- time to take stock of recent accomplish- tem from 4:3:3:2 (4 years lower primary, 3 ments in the advancement of Thailandûs years upper primary, 3 years lower second- secondary education, as well as consider ary and 2 years upper secondary) to 6:3:3, the challenges that lie ahead. whereby six years of compulsory primary education is followed by three years each This chapter elucidates the importance of of lower and upper secondary schooling. developing the Thai secondary education The subsequent NEDS of 1992-2001 oversaw system, first by considering its distinct histori- the passing of a new Constitution in 1997, cal development. Education reform efforts which ensures the çequal right to receive since 1932 are considered in order to fundamental education for the duration of explore the trajectory that Thailandûs sec- not less than twelve years which shall be ondary education system followed. The provided by the State thoroughly, of quali- potential for secondary education to serve ty, and without charge,é paving the way for as the key bridging point between primary universal access to 12 years of quality edu- schooling, tertiary education institutions and cation for all Thai children. the labor market is considered. The advancement of secondary schooling is dis- In 1999, the NEA was promulgated to serve cussed with respect to its links to the allevi- as the cornerstone of education provision ation of poverty and income inequality. and administration. The NEA raised compul- Finally, an international comparison places sory education from 6 to 9 years, enforcing the current state of Thai secondary educa- all parents to enroll their children in schools tion in the context of accomplishments of until they graduate from the lower second- comparable countries. ary level. In particular, the NEA recognizes children with special educational needs. Those with physical, mental, intellectual, emotional, social, communication or learn- ing deficiencies as well as economically or ThailandSocialMonitor:Improving SecondaryEducation 15 legally disadvantaged groups are fully enti- encouraging a thirst for knowledge that can tled to government educational services lead to life-long learning. and basic education is to be specially pro- vided. Teacher reform has focused on improve- ments in the quality of pre-service and in- Under the NEA, the Thai education system service teacher training, professional stan- was divided into formal, non-formal, and dardization and personnel administration. A 2 informal sectors. All educational institutions, five-year university program for preparing regardless of the type of education provid- new teachers has been rolled out. During ed, are expected to allow the transfer of the 2003-2006 period, the MOE has set a credits both within and across different target to produce teachers and school types of institutions so that students are able managers with post-graduate qualifications to transition smoothly between school levels (higher than a bachelorûs degree). The and tracks as necessary. This eases the results of these reform efforts need to be transition for students who drop out of the carefully evaluated. While support for in- formal education system but elect to con- service teacher development has become tinue their learning through community- more available, training programs have based or non-formal programs. The policy tended to be conventionally provided also allows students to re-enter the formal through short-term workshops. A cascade education system later. This section of the model predominates where Ministry officials NEA also spurred the building of additional train ESA supervisors who subsequently train schools for special programs and learning teachers. Training sessions generally provide centers to enhance informal education. little time for teachers to practice what they However, the credit transfer across different are learning and seldom is any follow up types of educational institutions has support provided when they return to their remained very limited to date. classrooms to implement what they have learned. There is limited evidence that The NEA calls for major reforms in all these training programs are demand-driven aspects of the Thai education system. or fully responsive to teachersû needs. Nor is Among them are three priority areas: (a) there evidence that the current approach learning reform, (b) teacher reform and (c) to teacher professional development leads education quality assurance. In the area of to improved teaching practice and learning reform, a more targeted core cur- improved student learning. The on-going riculum responding to capabilities and inter- decentralization process has also hindered ests of different groups of learners has been teacher development activities as it remains developed. The new basic education cur- unclear whether the MOE or ESAs should be riculum entered into implementation coun- playing a leading role in this arena. try-wide in 2001. A primary goal is for Moreover, delays in the development of teaching to be more learner-oriented and professional standards and enactment of 2 Formal education consists of two levels: basic education and higher education. Basic education covers the 12 years of education prior to higher education. Higher education is further divided into two levels, lower than degree and degree levels. Formal education includes schools under the jurisdiction of the government, including institutions that provide schooling for students with disabilities, as well as welfare-based schools that address the needs of stu- dents who are socially disadvantaged. Other types of formal education include schools for the ecclesiastic, special- ized education, vocational and special vocational consisting of sports and fine art schools. Non-formal education aims to reach those outside the formal school system, including early childhood and adult education. Pre-schools, lit- eracy programs and certain adult vocational training fit under the non-formal education category. Finally, informal education promotes self-learning and often operates out of community-based locations such as learning centers, libraries and museums. 16 Acts related to decentralized personnel Considerable structural change has been administration have had an unfavorable introduced in recent years. The agencies effect on teacher reform as a whole. mainly responsible for educational provision, namely the MOE, the Ministry of University The quality of education provision must be Affairs (MUA) and the Office of National monitored through both internal and exter- Education Commission (ONEC), were re- nal quality assurance mechanisms. The NEA organized into a single MOE. One hundred requires that internal evaluations for second- and seventy five ESAs have been estab- ary education be conducted annually by lished to handle education management at each individual institution on the basis of a decentralized level. ESAs have different MOEûs standards. These evaluations must capacities for absorbing service delivery be made available to the public. External functions transferred to them due to varia- evaluations are carried out by an independ- tions in coverage area, number of qualified ent agency - the Office for National Education personnel and endowed resources. Standards and Quality Assessment Sustained technical assistance will be need- (ONESQA). For the most part, it appears ed to build local-level institutional capacity that the internal and external quality assur- for efficient service provision. ance process to date has been largely pro- cedural. There is limited demand for infor- The 1997 Constitution ensures the right of mation on school quality and the formative local administration organizations to partici- evaluation process has not translated yet pate in the provision of education to into a reflective exercise to improve school improve outcomes through increased com- development planning or foster account- munity participation. The decentralization ability to education sector stakeholders initiative requires Local Government regarding the quality of service delivery. Organizations (LGOs) to assume greater responsibility in school management. LGOs Most recently, the National Education Plan began taking on some functions in late 2002-2016 has advanced a vision of educa- 2004, including the monitoring of child tion that embraces human-centered devel- development centers, developing appropri- opment and a holistic scheme integrating ate activities for pre-primary schools, over- education, religion, art and culture. It is seeing sub-district libraries and village read- hoped that the National Education Plan ing centers, and providing school milk and çwill (1) lead to a knowledge-based socie- lunches. However, the transfer of second- ty; (2) promote continuous learning and (3) ary school management to LGOs has been involve all segments of society in designing more complex. A study conducted by the and decision-making concerning public Office of Inspection and Evaluation found activitiesé (Bhangananda 2003). The goals that around 78 percent of LGOs were ready outlined in the National Education Plan to assume transferred functions in primary reflect an ongoing debate in Thailand education, but only 11 percent of these about the balance between educational LGOs were ready to take responsibility for development for the sake of promoting secondary schools. Further evaluation is economic competitiveness and for preserv- needed to assess the readiness of school ing çcultural self-relianceé (Witte 2000). This administration capacities before this transfer debate has intensified during the post-Asian occurs at the secondary level. Currently, Crisis years, as Thailand attempts to navi- the Cabinet has agreed to slow down the gate the tides and pressures of economic transfer process and requested that the globalization. MOE work closely with related stakeholders ThailandSocialMonitor:Improving SecondaryEducation 17 to ensure a smooth transition. In addition, cation, such as Education for All and the teachers have raised concerns about trans- Fast Track Initiative at the primary level. ferring administrative responsibility for sec- While there tend to be advocates that push ondary schools to LGOs, fearing that this for the expansion of primary and tertiary could lead to political interference in edu- institutions at the country level, secondary cational issues such as teacher job security, schooling tends to be neglected. Finally, curriculum and classroom pedagogy as well reaching political consensus for secondary as possible recruitment of teachers for elec- expansion and reform has been more diffi- toral purposes. cult than for primary or tertiary education, making secondary school policy choices In addition to recent reforms, the RTG cur- more ambiguous, risky and complex rently allocates more than 20 percent of its (Moreno 2005). Yet there has been increas- total budget to the education sector, ing recognition of the key role that second- reflecting its strong commitment to educa- ary education can play as the bridging tional development. The emphasis on the point between primary schooling, higher importance of education comes at a cru- education and the labor market. Below, cial time for Thailand. Emerging from the the link between primary expansion and Asian Crisis, the country has revived and secondary schooling in Thailand is outlined. projected itself towards being a competitive The importance of secondary education in nation in the global marketplace. Four connecting young people to tertiary educa- national priorities have been laid down, all tion and the labor market is also discussed. of which require better quality of education and knowledge management: (a) increas- During the 1980s, the RTG promoted univer- ing competitiveness; (b) reducing poverty sal primary education and reduced the and inequality; (c) developing social capital adult illiteracy rate through the heavy and (d) managing natural resources. expansion of primary education. In this peri- Furthermore, as economic growth picked up od, the government built at least one pri- speed after the crisis, exports have grown mary school with extension classes up to from USD50,000 million in 1998 to almost lower secondary levels for every two villages USD111,000 million in 2005. Foreign Direct throughout the country (Suwansathit 2002). Investment (FDI) also rose from USD6,900 mil- In addition, the change in the format of the lion to USD9,800 million over the same peri- education system from 4:3:3:2 to 6:3:3 od. The trend in economic growth and FDI ensured that children stayed longer in has placed pressure on local firms to be schools. Through this effort, universal pri- more competitive, as well as stimulate tech- mary education, measured in terms of the nological progress, and hence heightened Gross Enrollment Rate (GER), was successful- the demand for a better skilled and edu- ly achieved (104 percent in 2003). cated workforce. Household survey data largely confirm offi- cial statistics, showing a rising trend towards BRIDGINGBASIC universal primary completion. School EDUCATION,TERTIARYEDUCATION ParticipationRates(SPR) - thatisthepro- ANDTHELABORMARKET portion of children 6-11 years of age that enroll in school regardless of their schooling At a global level, limited investments in sec- level - have also demonstrated impressive ondary education have been an outcome outcomes. In 2003, the transition rate from of several factors. There have been no primary to lower secondary level was 93 comprehensive initiatives for secondary edu- percent, the retention rate at the primary 18 level was 90 percent and the literacy rate ical location or sex. Especially across was 96 percent (Thailand Ministry of income groups, recent trends between 1994 Education 2004). and 2002 show that the primary SPR for the poorest quintile has risen continuously over Such achievements at the primary educa- time, reflecting the successful inclusion of tion level were experienced across the Thai poor households (see Figure 1.A). population, regardless of income, geograph- Figure 1.A: Primary School Participation Rate by Income Quintile, 1994-2002 Source: Household Socio-Economic Survey 1994-2002 An analysis of public expenditure in 2000 tion level, about 31 percent of public also supports this finding. A calculation expenditure was allocated to the poorest based on the number of students enrolled quintile while only 8 percent went to the in public school and fixed unit cost show richest quintile (see Table 1.1). that at the pre-primary and primary educa- Table 1.1: Incidence of Public Spending for Primary Education by Income Quintile, 2000 Expenditure Incidence Level of Income Public School Enrollment Enrollment Fixed Unit (thousands) Cost (% share) Q1 (Poorest) 2,118 29,755 31 Q2 1,739 24,425 25 Q3 1,463 20,549 21 Q4 1,050 14,754 15 Q5 (Richest) 548 7,704 8 Source: Household Socio-Economic Survey 2000 As the RTG intensified efforts to achieve uni- expansion of tertiary education, while sec- versal primary education in the 1980s, it also ondary enrollments stagnated through this focused on boosting access to tertiary edu- period. Main obstacles have been poverty cation. This resulted in almost universal and the high direct and opportunity cost of access to primary education and a rapid education (World Bank 1998). Such factors ThailandSocialMonitor:Improving SecondaryEducation 19 resulted in a sluggish secondary Gross expansion at the primary level, the focus Enrollment Rate (GER) of around 30 percent shifted towards tertiary education, overlook- with almost half of all children finishing pri- ing the expansion of secondary education. mary education dropping out of the formal Consequently only a small number of sec- education system. ondary graduates constituted the new recruits entering the skilled labor force and In a recent World Bank study titled çClosing the middle income bracket. In other words, the Gap in Education and Technology, ç de the distribution of educational attainment in Ferranti et al. (2003) find that most OECD Thailand was çsqueezed from the middleé, countries and many East Asian countries fol- where the base got thinner and the top lowed an educational transition trajectory wider, while the middle remained relatively that resembles a pyramid: primary educa- unaffected. The distribution looked more tion was universalized first, followed by an like an anvil than a pyramid, with the major- expansion of the secondary system and ity of the population having only primary finally broader access to universities. education or less, but more individuals with Thailand, instead, took a different path. In tertiary education than secondary educa- the 1980s, after a decade of successful tion only (see Figure 1.B). Figure 1.B: Education Transition Patterns Source: De Ferranti et al. 2003 The achievement of universal primary edu- ondary students benefited from these sub- cation eventually led to increasing pressure sidy programs. However, poverty and the on the Government to expand compulsory cost of schooling have continued to be a education to cover the lower secondary constraint for expanding secondary school- level. Direct subsidies were introduced in ing and reaching universal coverage. 1987 covering free textbooks and uniforms, low cost dormitories, health/nutrition pro- Country comparisons show that Thailandûs grams and school fee exemptions. secondary GER was stagnant until around Approximately 50 percent of all rural sec- 1990 and then picked up dramatically. In 20 contrast, other Asian countries started with grow at a more modest pace during the higher secondary GERs and continued to same period (see Figure 1.C). Figure 1.C: Comparison of Secondary Gross Enrollment Rate Trends, 1970-2000 Source: World Bank 2005 Low levels of investment at the secondary ta quintile reveals large differences in sec- level through the 1980s had led to a rela- ondary school enrollment between the tively low-skilled workforce in Thailand. In poorest and the richest population groups. 1997, 70 percent of the total labor force These differences have remained quite sub- had only received elementary education or stantial over time. The gap in SPR between less, while 17 percent had obtained second- the wealthiest and the poorest population ary education and 8 percent had a univer- quintiles in 1994 was 24 percent (85% versus sity degree. But investments in secondary 61%) and stood at 17 percent in 2002 (93% education in the 1990s began to pay off. versus 76%). While it is clear that second- By 2004, the labor force with more than pri- ary schooling opportunities for the poorest mary education reached close to 40 per- children still have much room for improve- cent. Secondary GERs had jumped from 68 ment, Thailand had accomplished an percent in 1994 to 77 percent in 2002. impressive rate of expansion nonetheless But disaggregating by household per capi- (see Figure 1.D). ThailandSocialMonitor:Improving SecondaryEducation 21 Figure 1.D. Secondary Participation Trends by Income Quintile, 1994-2002 Source: Household Socio-Economic Survey 1994-2002 The incidence of public expenditure for sec- try, higher in urban than in rural areas and ondary education shows that spending is lower in those districts where ethnic groups distributed almost equally across quintiles account for a high percentage of the pop- (see Table 1.2). Such spending patterns ulation. The challenge facing the govern- partially explain why the gap in secondary ment is how to continue to expand basic achievement between the richest and the education to disadvantaged groups while poorest quintile has remained relatively sta- maintaining universal access to and good ble over time. Enrollment rates are higher quality of primary education. in Bangkok than in other parts of the coun- Table 1.2: Incidence of Public Spending for Secondary Education by Income Quintile, 2000 Expenditure Incidence Level of Income Public School Enrollment Enrollment Fixed Unit (thousands) Cost (% share) Q1 (Poorest) 843 11,829 19 Q2 987 13,846 23 Q3 967 13,575 22 Q4 885 12,412 20 Q5 (Richest) 646 9,064 15 Source: Household Socio-Economic Survey 2000 22 Mandatory education under the 1999 NEA vance. Anticipating the needs of the new only includes up to lower secondary school- knowledge-based economy will prove key. ing. Students who want to continue their The çnew general skillsé needed for such an education beyond the lower level have an economy go beyond reading and writing option between three years of upper sec- and mathematics to include such things as ondary or three years of lower vocational the ability to work in a team, to approach (technical) education. Most students who new problems creatively, to know how to choose upper secondary education aim to use a computer and at least to understand go to university. Those who choose lower English, if not speak it fluently (World Bank vocational education tend to continue their 2001b). The Thai secondary education sys- education at the upper vocational level. tem will act as the bridging point to a Students with diploma or upper vocational changing labor market only to the extent education can continue to a university that graduates attain such skills and their degree by continuing with two more years talents and creativity are harnessed into the of schooling at a university. economy. Enrollment estimates at lower and upper AKEYTOOLFOR secondary levels indicate that growth trends CLOSINGEQUALITYAND and between-group gaps are much larger EQUITYGAPS in the higher grades, probably as a result that in upper secondary education there Secondary education is a key tool for alle- has been greater room for improvement. viating poverty in Thailand. A recent report For example, the average SPR for lower sec- on Poverty and Public Policy (World Bank ondary (12-14 years age-group) increased 2001a) finds that in 1998-99, individuals with from 88 percent in 1994 to 94 percent in upper secondary education and vocational 2002, while at the upper secondary level and technical qualifications improved their (15-17 years age-group), it increased from standard of living. In contrast, those with 57 percent in 1994 to 77 percent in 2002. secondary or lower levels of schooling suf- The absolute (and relative) gains were larg- fered real income declines. Education also er in the poorest quintile: from 31% to 56% had a powerful effect on reducing rural at the upper secondary level. poverty. The risk of poverty declined by 66 to 74 percent when the highest educated In an effort to continue to expand second- adult in the household had primary or sec- ary education, Thailand faces two critical ondary education, as compared to no edu- challenges: developing an effective strategy cation. Finally, the less-educated popula- for further broadening access to upper sec- tion was at greater risk of increased pover- ondary education and finding an appropri- ty during times of crises. Between 1996 and ate balance between the academic and 1999, poverty incidence increased from 21 vocational tracks. Thailandûs ability to to 24 percent for households headed by an address these issues will greatly impact its illiterate person and from 12 to 19 percent ability to open up the pipeline from second- among those headed by a primary-educat- ary to tertiary education as well as to the ed person. In contrast, the rate did not labor market. change appreciably among households headed by persons with vocational and Furthermore, Thailandûs reform program postsecondary education. focuses not only on access, but also enhancement of quality and market rele- ThailandSocialMonitor:Improving SecondaryEducation 23 Secondary education is an important vehi- tional attainment of household head con- cle for bringing about broader income tribute to lingering inequality in income dis- equality and social equity in Thailand, par- tribution (see Table 1.3). Additionally, differ- ticularly since the country has one of the ences in education attainment alone highest Gini coefficients in the region (0.51 account for at least one-fifth of total over- in 2002, contrasted with the regional aver- all inequality and for an even larger frac- age of 0.38). Furthermore, Thailand was tion for within-region inequality. In particu- estimated to have the fifth worst income lar, inequality in access to secondary edu- distribution among developing countries in cation may adversely affect the extent of the 1990s (Phongpaichit and Sarntisart 2000). inequality in income distribution Income inequality in Thailand worsened (Phongpaichit and Sarntisart 2000). steadily from the 1960s to 1992, improved Evidence also suggests that as secondary marginally over 1992-98, and then lost all education is expanded, and as supply of lit- those gains over one year in the Asian eco- eracy and other skills is distributed among nomic crisis. Evidence from Thailand, as more youth, income inequality begins to well as other countries such as Turkey, Chile diminish (UNESCO 2003). and Russia, shows that differences in educa- Table 1.3: Gini Coefficient by Country and Share due to Differences in Educational Attainment of Household Head Country Gini Coefficient % of Total Income Inequality due to Education Thailand (1999) 0.53 19% Turkey (1994) 0.45 22% Chile (1999) 0.51 26% Russia (1995) 0.47 5% Source: World Bank 2000 and 2001a Apart from private rates of return, which are Prichett 1997). Additional years of schooling realized through higher earnings of individu- also serve as an effective prevention als, an increase in educational attainment scheme against HIV/AIDS, an epidemic that contributes to higher social returns to invest- is increasingly of concern in Thailand and ment in education, particularly with respect the region (World Bank 2002). Finally, chil- to health benefits. Social returns, for dren residing in households headed by an instance, may come in the form of a bet- educated individual have an increased ter educated mother who heads a family chance of continuing with additional years that is more health conscious, better nour- of schooling. ished and has the prospect of realizing the importance of education for the next gen- Secondary education attainment is also a eration. Evidence from 45 demographic contributing factor to non-market public and health surveys across countries show benefits. In general, better educated citi- that children of mothers with secondary zens tend to participate more in public schooling have a mortality rate that is 36 affairs in the form of increased voting rates percent lower than those whose mothers and staying abreast of current events in pol- only have primary schooling (Filmer and itics. In addition, better educated citizens 24 tend to have less association with crime. ondary education across different groups of One study shows a decreasing likelihood of population is explored, as well as the obsta- youths engaging in criminal activities when cles behind universal access to secondary they attend school and work (Witte and education. Second, the quality of second- Tauchen 1994). Increased years of school- ary education is analyzed on the basis of ing, usually associated with higher earnings, comparative data from international assess- can also reduce the reliance on welfare ments. Third, this issue reviews the efficien- and public assistance programs. cy of the secondary education system, both in terms of resource allocation and internal The NEA paved the way for a new stage in efficiency. Finally, some general policy rec- the enhancement of Thailandûs education ommendations are provided. system. The most recent National Education Plan has the potential to further advance The report is organized as follows. Chapter the system with strategies currently in place 2 reviews the current state of the Thai sec- through 2016. In particular, efforts to bal- ondary education system as a whole and ance economic competitiveness and across different groups of the population. human-centered development mark a dis- Additionally, comparisons are offered across tinct phase in the thinking that frames regions. Based on these results, an analysis Thailandûs future educational advancement. of the demand side of secondary educa- Although recent trends in secondary educa- tion is presented, including effects of house- tion expansion show improving results, much hold decision-making in sending children to remains to be done to promote access and school. Chapter 3 looks into the issue of quality as well as the efficiency of the sec- the quality of secondary education in ondary education system. Thailand and suggests possible avenues for improving it. Chapter 4 investigates options In this regard, the current issue of the to enhance efficiency in the utilization of Thailand Social Monitor looks into the chal- financial resources. Finally, Chapter 5 sug- lenges lying ahead for secondary education gestes policy recommendation, taking into in Thailand. First, the interplay of demand consideration the goals of the RTG and the and supply is investigated, and in particular, distinct historical evolution of its education how their interaction affects the secondary system. education system. Equity in access to sec- ThailandSocialMonitor:Improving SecondaryEducation 25 II.ACCESSTOANDEQUITYIN SECONDARYEDUCATIONIN THAILAND 26 The RTG intensified its efforts to expand lines of gender, the rural-urban divide, access to secondary education in recent regional and income groups. years, particularly in rural areas. Its commit- ment is mirrored in the 1997 Constitution The analyses in this chapter are largely where universal access to 12 years of edu- based on two data sources: (1) administra- cation is guaranteed for all Thai children. tive data collected by the MOE and (2) a Greater commitment is reflected in the 1999 large-scale, national survey of representative NEA which extended compulsory education households, called the Household Socio- 4 from 6 to 9 years. To translate these com- Economic Survey or SES (National Statistics mitments into action, the RTG has set a tar- Office 1994-2002). While the MOE adminis- get to achieve universal lower secondary trative data are considered to be census- education by 2006 and universal upper sec- based, errors due to inconsistency and data 3 ondary education by 2015. manipulation are known to exist. The SES has been recognized as a credible source Currently, lower secondary education is at of data. However, its reliability is also limit- reach for most children. Compared to ed due to a small sample size - 45,000 enrollment rates in 1994, access to second- households for most of the analyses in this ary schooling has notably expanded for all chapter, but even smaller when focusing on socioeconomic groups. This is a result of a the secondary school age population concerted effort from the RTG to redress (between 12 to 17 years). A snapshot of inequities in education participation. data from these two sources is not always Despite important gains, much work remains consistent, but they reflect similar trends to be done. While 98.6 percent of children over time. were estimated to complete primary school in 2002, only 88 percent transferred to lower ACCESSTO secondary and 69 percent continued to SECONDARY upper secondary. Poor and rural children EDUCATION are at the greatest disadvantage. One of the challenges for the RTG lies in develop- There has been a consistent gradual rise in ing appropriate policies and strategies to the average number of years of education- enroll and keep disadvantaged children in al attainment for the Thai population aged secondary school in order to continue to 15 and over. For the population aged promote equitable development and between 15 and 21, in particular, the aver- growth opportunities. Thus, this section age years of schooling is nearly 10 years. offers an in-depth look at secondary educa- This suggests that most children have at tional attainment, disaggregated along the least completed lower secondary educa- 3 In September 2000, the RTG, along with other governments, signed the Millennium Declaration, pledging a commit- ment towards achieving the MDGs, where achieving universal primary education is one of the eight goals. The first Thailand MDGs Report, launched in June 2004, assessed the current Thai education system and showed that Thailand has already achieved universal primary education in terms of gross enrollment ratio. 4 The SES is conducted every two years by the National Statistics Office (NSO). It contains information on household income, household consumption patterns, changes in assets and liabilities, ownership of durable goods and housing characteristics. It was first conducted in 1957 with intervals of five years until 1988, after which point the survey has been undertaken every two years. ThailandSocialMonitor:Improving SecondaryEducation 27 Table 2.1: Average Years of Educational Attainment, 1999-2003 Age 1999 2000 2001 2002 2003 15 and over 7.2 7.4 7.6 7.8 7.1 15-21 9.5 9.6 9.7 9.8 9.4 15-59 7.8 7.7 7.8 7.9 7.7 60 and over 3.6 3.7 3.8 3.9 3.5 Source: Office of Education Council 2004a Educational attainment profiles for ages 16 least grade 6, indicating an impressive pri- to 19 show a notable improvement over mary school completion rate. Moreover, 80 time. Based on the SES, the number of percent of the population aged between youth between 16 and 19 years-old who 16 and 19 had completed lower secondary completed primary education (grade 6) education in 2002, which is a significant and lower secondary education (grade 9) increase compared with 50 percent in 1994. rose continuously since 1994 (see Figure However, a closer review shows that most 2.A). By 2002, about 95 percent of the 16 of the gains in educational attainment took to 19 year-old population completed at place before 1998. Figure 2.A: Thailand Educational Attainment Profiles for Ages 16 to 19, 1994-2002 Source: Household Socio-Economic Survey 1994, 2002 Overall, secondary education enrollment has difference between gross and net enrollment improved after a period of stagnation prior to rates). However, both GER and NER have the 1990s. Even through the financial crisis in risen at a decreasing rate in recent years. the late 1990s, enrollment growth remained The challenge of maintaining a constant relatively constant. GER rose from 68 to 77 growth rate over time is common among percent between 1994 and 2002, while Net countries as they approach universal access Enrollment Rate (NER) also rose from 68 to 74 to schooling, known as the çceiling effect.é percent (see Box 2.1 for a discussion on the 28 Box 2.1: Educational Enrollment - GER or NER? There are several methods of measuring educational enrollment and participation. Traditional aggre- gated measures of educational enrollment include gross and net enrollment rates. These two indica- tors measure the number of children enrolled in a specific education level as a percentage of the school age population corresponding to the same education level. GER measures the number of all children enrolled, regardless of age, as a percentage of the population of children in the specified education level. NER includes only those children in the official age group for a particular education level, again measured as a percentage of the population of children in the specified education level. In the case of Thailand, the official age group for secondary education is 12 to 17 years. GER is widely used to show the general level of participation in a given education level. Unlike NER, it reflects the extent of over-aged or under-aged enrollment. A secondary GER higher than 100 per- cent indicates the presence of children who are either over-aged or under-aged in the system, thus exceeding the number of children in the official secondary level age group. A high repetition rate might produce a GER higher than 100 percent, resulting in shortage of space for children whose age corresponds to the official secondary age group. A comparison between GER and NER can, in this sense, be useful for analyzing internal efficiency of a system through the construction of elaborated indicators based on cohort analysis. Achieving 100 percent NER is perhaps unrealistic, given that it would require every child to enter school at exactly the official age, proceed through the education level with zero repetition or disruptions, thus resulting in a 100 percent on-time completion rate. This is a highly improbable scenario, particularly in countries with constraints on on-time enrollment. Source: Monitoring Educational Performance in the Caribbean (Di Gropello 2003) Disparities between gross and net enrollment children attend school at the grade corre- rates in Thailand are minimal, suggesting lit- sponding to their age, the vast majority of tle age mismatch. Unlike countries such as children are within one year of their expect- Brazil, where the ratio of gross to net enroll- ed official grade level (see Table 2.2). This ment rate is high (i.e. 1.63 for primary and is good news for Thailand. Often, students 1.24 for lower secondary), Thailandûs ratio at who are significantly over-aged run a high- the secondary level is around 1.1 (Larach er risk of dropping out of school for reasons 2001). The gap between secondary NER including the rise in the opportunity cost of and GER reflects the extent of over- and education to foregone income from labor under-age students in the education system. as a child grows older. For most grades, while less than one third of Table 2.2: Age Distribution by Grade, 2002 Age 12 13 14 15 16 17 18 19 Grade 7 12 36.46% 55.02% 6.81% 1.23% 0.47% Grade 8 13 3.92% 33.48% 54.78% 6.59% 1.23% Grade 9 14 1.10% 3.46% 34.59% 53.84% 7.02% Grade 10 15 29.81% 59.22% 8.74% 1.65% 0.59% Grade 11 16 3.77% 27.77% 58.73% 8.08% 1.64% Grade 12 17 1.60% 3.79% 28.33% 57.49% 8.80% Source: Ministry of Education 2002 ThailandSocialMonitor:Improving SecondaryEducation 29 Education participation for teen-age chil- Repetition is a minor problem. MOE data dren has expanded, but there is still room from 2002 show that while there was a slight for improvement. The SPR of the cohort repetition rate (1 to 2 percent) at the pri- aged 12 to 17 years increased from 75 to mary level, there was none for secondary 86 percent between 1994 and 2002. This education. Dropout is relatively a source of reveals an improvement in general access greater concern. An analysis of school to and permanence within the education dropouts in 1999, which sampled 1,157 system since SPR measures the school par- schools and included grades 1 to 12, found ticipation level of children in a particular that only 2 percent of students dropped out age group, regardless of the grade attend- of school. However, examining data that ed. Given that there were about 5.8 mil- follow three cohorts from grades 1 through lion children in the 12 to 17 year-old cohort 12, it is clear that students primarily drop out in 2002, this leaves about 800,000 children of the education system when they transi- excluded from the education system. The tion from one level to another (depicted by challenge for the RTG lies in reaching the dips in cohort enrollment between grades 6 last 14 percent of the 12 to 17 year-old and 7, as well as 9 and 10 in Figure 2.B). cohort and successfully bringing them into Thus once students start grade 1, they tend the formal education system. To achieve to stay on through grade 6. From grades universal secondary education, it will be cru- 6 to 7, some students drop out of the sys- cial to formulate appropriate strategies for tem while those that continue tend to stay reaching out-of-school youth and target on through grade 9. interventions to address the obstacles they face in attending secondary school. Figure 2.B: Transition of Three Cohorts from Grades 1 to 12, 2001-2003 Source: Ministry of Education 2001, 2002, 2003 30 The vast majority of secondary school stu- a different equity challenge - keeping boys dents are enrolled in traditional general in school - has arisen. academic programs. A snapshot review of secondary students in 2004 using MOE data Girls outperform boys in secondary school shows that out of approximately 5.8 million participation and completion. NER for girls youth between 12 to 17 years-old, around in 1994 was 63 percent contrasted with 61 74 percent are enrolled in the formal edu- percent for boys, while in 2002, NER for girls cation system. About 84 percent of all stu- was 72 percent and 64 percent for boys. dents attend a school administered by the Thus, while enrollment for both girls and MOE. Among those who attend schools boys is on the rise, girls outperform their facilitated by the MOE, approximately 75 male counterparts and this gap has grown percent of secondary students go to institu- over time (see Figure 2.C). In addition, the tions under the Office of Basic Education grade 9 completion rate for females has Commission (OBEC). For upper secondary overtaken that of males between 1994 and level, approximately 63 percent of students 2002. While in 1994 about half of the go to an academic track while 37 percent cohort for both females and males had go to a vocational track. Less than 1 per- completed grade 9, the completion rate cent of students attend welfare schools that was 84 percent for females and 76 percent are provided for those in need of financial for males by 2002. assistance and less than 1 percent of sec- ondary students attend schools for the dis- abled or with special programs. The private sector plays a small role in general second- ary education, accounting for 11 percent of student enrollments in lower secondary and 20 percent in upper secondary education. The highest proportion of private enrollments is found in the vocational education track at the upper secondary level, accounting for approximately 38 percent students enrolled in 2004. EQUITYINSECONDARY EDUCATION Equity in access to secondary education has improved. Participation rates in second- ary education demonstrate that access across gender, the rural-urban divide, regions and socioeconomic status groups has expanded, benefiting all groups in some measure. However, while the income and urban/rural gaps have grown smaller, on average the gender gap has grown signifi- cantly larger and to the benefit of girls. In contrast to many other developing nations, ThailandSocialMonitor:Improving SecondaryEducation 31 Figure 2.C: School Participation Rates, Gross Enrollment Rates and Net Enrollment Rates by Sex and Gains, 1994-2002 Source: Household Socio-Economic Survey 1994, 2002 Coverage has been expanded to rural access to secondary education for rural areas. With respect to the rural-urban children, following massive expansion of divide, net enrollment and school participa- schools to remote locations in the early tion rates in non-municipal, or rural, areas 1980s. Similarly, the urban-rural gap for have pulled alongside municipal areas, and grade completion has decreased from 25 in some cases even exceeded rates of their percent in 1994 to only 8 percent in 2002 urban counterparts. The gradual rise in (see Figure 2.D). both SPR and NER reflects improvement in Figure 2.D: Grade Completion by Urban/Rural Location, 1994 and 2002 Source: Household Socio-Economic Survey 1994, 2002 32 The Northeast, the poorest region in maintain that position through 2002 (see Thailand, is catching up with other regions. Figure 2.E). Most of the growth took place In the 1980s, there was great concern between 1994 and 1996. Like many other about lagging enrollment rates of the most regions, the Northeast experienced a dip in impoverished regions in Thailand, particular- NER in more recent years. More striking is ly the Northeast. However, SES data show the NER trend in Bangkok, where the NER that by 1994, NER for the Northeast was fell between 1994 and 1998, while rates in higher than other regions and continued to other regions continued to climb. Figure 2.E: Secondary Net Enrollment Rates by Region, 1994-2002 Source: Household Socio-Economic Survey 1994, 2002 Disparities exist across provinces, especially provincial per capita income, school partic- among the richest and the poorest ipation rates are relatively consistent among provinces. Based on data collected by the the bottom four quintiles (see Table 2.3). MOE, most provinces achieved secondary However, the provinces in the wealthiest GERs around 60-70 percent by 2002, reflect- quintile achieved higher rates than those in ing notable improvements but also extensive the bottom quintiles, across all age groups. room for further expansion. Grouping The difference among provinces is especial- provinces into five quintiles according to ly striking for younger children. ThailandSocialMonitor:Improving SecondaryEducation 33 Table 2.3: School Participation Rate by Income Quintile and Provincial per Capita Income, 2002 Quintile 12-14 Years Old (%) 15-17 Years Old (%) 12-17 Years Old (%) Q1 (Poorest) 51.73 36.47 44.03 Q2 51.84 35.40 43.58 Q3 55.49 35.81 45.59 Q4 55.04 36.69 45.74 Q5 (Richest) 66.36 41.21 53.91 Source: Estimation with data from Ministry of Education and Ministry of Interior Figure 2.F shows Kaplan-Meier survival curves grades 7 to 9. By 2002, the gap between for all children in the 6 to 15 year old age- the richest and the poorest quintilesû com- group in 1994 and 2002, both overall and pletion rates had narrowed, but persisted disaggregated by wealth quintile. Overall despite improvements in absolute levels for survival probabilities through grade 6 were both quintiles. While the richest quintileûs already high in 1994, and are more than 98 survival rate was almost 100 percent for percent for 2002. In both years, the transi- grade 9, it was only slightly above 80 per- tion between primary and secondary is cent for the poorest group. These survival where attainment drops off. Nevertheless, estimates confirm that overall the problem by 2002 overall grade 7 completion was of retention is more pronounced in the tran- over 90 percent. This average masks signif- sition between primary and lower secondary icant differences by quintile though. In and only minor across each grade within 1994, the completion rate gap between the these levels. Again, however, there is some poorest quintile and other groups was less dropout among children from the poorest than 5 percent through to grade 4, and group within both the primary and second- then widened progressively: 6 to 10 percent ary cycles, even among this most recent for grades 5 and 6, 20 to 50 percent for cohort of children. Figure 2.F: Survival Curve Estimates for Ages 6 to 15 in Thailand, 1994 and 2002 Source: Household Socio-Economic Survey 1994, 2002 34 Urbanicity is relatively a minor factor in International comparisons show that explaining inequities of survival and comple- Thailand made dramatic improvements over tion. Previously shown in Figure 2.D, the the past two decades. Data from the urban-rural gap on grade survival curves is World Development Indicators (World Bank smaller than that of household wealth (of 2003) confirm that Thailand started out in the poorest and the richest) depicted in fig- the late 1980s with one of the lowest sec- ure 2.F. Survival probabilities to grade 6 of ondary gross enrollment rates in the region, urban and rural children were largely similar but subsequently picked up pace in later in 1994 and more so in 2002. Urban-rural decades to position itself in the top tier due differences grow larger across into the lower to a concerted Government effort to secondary grades, albeit less pronounced expand access (see Figure 2.G). now than a decade ago. Figure 2.G: Trends in Secondary Gross Enrollment Rates across Asian Countries, 1980-2003 Note: Data for Thailand and Philippines in 2000 are from World Development Indicators 2003 Source: World Development Indicators Database, World Bank 2006 ThailandSocialMonitor:Improving SecondaryEducation 35 GERs for the upper secondary level have and India in this realm (OECD 2005b). also experienced a notable increase. The Furthermore, Thailandûs share of secondary share of upper secondary school graduates school graduates was largely at par or larg- to the population has steadily increased, er than countries with a higher GDP per reaching 59 percent in 2003. Thailand per- capita, such as the Philippines, Argentina or forms comparatively better than other Mexico (see Figure 2.H). regional neighbors such as Indonesia, China Figure 2.H: Percentage of Upper Secondary Graduates to the Population, 2003 Source: OECD 2005a REACHINGOUTTOTHEEXCLUDED Thailandûs expansion of secondary educa- ly decision-making regarding the demand tion is a success story, but many children for secondary education and the level of remain excluded from sharing in the bene- investment in human capital. On the other fits of school participation. For Thailand to hand, there are supply side variables that achieve its commitment of universal second- affect access to secondary education. ary education by 2015, it is necessary to Understanding the interplay between these tease out the supply- and demand-side variables will allow the RTG to develop constraints to education faced by the appropriate strategies for better targeting remainder 14 percent of the school age out-of-school children, broadening coverage population outside the formal education sys- even further and achieving universal partic- tem. On the one hand, there are several ipation objectives. socioeconomic factors that determine fami- 36 Box 2.2: The Path to Universal Secondary Education in Korea The Republic of Korea achieved nearly universal primary and secondary education in just four decades, following the end of the Korean War in 1953. Educational expansion in South Korea was accompa- nied by a declining Gini coefficient, indicating that equality gaps were narrowed during the same time period. Korean students are also among the top performers in both mathematics and science in OECD countries, as illustrated by recent PISA and TIMSS results. According to a case study on Koreaûs path to universal secondary education in çExpanding Opportunities and Building Competencies for Young People: A New Agenda for Secondary Educationé (World Bank 2005), several factors played a key role. First, Korea included strategies for a strength- ened and broadened education system in its national development plan as early as the late 1950s. Education was identified as a top priority area in the 1960s and a focus on secondary education was established in the 1970s, followed by tertiary level in the 1980s. Second, Korea included equality and equity considerations in its educational expansion strategies. In 1968, the government abolished entrance examination for middle schools and introduced a lottery system for student placement, which was intended to democratize access at this level. The High School Equalization Policy (HSEP) was passed in 1974, which was intended to equalize schools inputs such as operating expenditures, student intake, class size and education facilities. As a result, there is no discernible quality difference across public schools or between private and public institutions. Third, government expenditure on educa- tion has increased steadily since the 1950s. Education expenditure as a percentage of GDP increased from 2.9 percent in 1970 to nearly 5 percent in 2003. Fourth, private school participation has played a significant role in sustaining expansion. While providers of secondary education were greatly sup- ported by government tax incentives, fees, family contributions and foreign aid at first, government revenues have been reinvested in education following the introduction of school-leveling policies. MOE statistics primarily report on children cational system. But this policy has not who are already in school. These data also been clearly articulated and children with provide some insights on studentsû decision disabilities have traditionally not been making regarding whether to continue or emphasized as a priority target group to abandon school. The Thai education man- reach Education for All targets. Thus, they agement information system, however, has have remained largely excluded from edu- limited information on out-of-school children. cation participation. According to national Statistics (or acceptable estimates) of these statistics, in 2004 there were only 175,000 children from various organizations vary children with disabilities enrolled in pre-pri- widely. mary through upper secondary school. The total population between 3 and 17 years Children with disabilities have been a group old in 2004 was 13,774,909. Thus, the share largely neglected from efforts to universalize of children with disabilities enrolled in school basic education. The MOE manages 41 was approximately 1.27 percent. Estimates specialized basic education schools for chil- from other middle-income countries suggest dren with disabilities throughout Thailand that the share of children with disabilities that serve approximately 13,000 students. In tend to oscillate between 4 and 5 percent, addition, the MOE has generally espoused suggesting that there are likely a few hun- an inclusive policy of mainstreaming chil- dred thousand children with disabilities in dren with disabilities within the regular edu- Thailand outside the school system. ThailandSocialMonitor:Improving SecondaryEducation 37 Table 2.4: Number of Disabled and Special Students by Type and Gender, Academic Year 2004 Type Male Female Total Seeing-impaired 5,898 5,445 11,343 Hearing-impaired 3,578 3,302 6,880 Mentally impaired 15,302 14,125 29,427 Physically impaired 8,343 7,701 16,044 Students having difficulties in learning 39,478 36,442 75,920 Speaking-impaired 5,994 5,532 11,526 Students with autism 1,965 1,814 3,779 Students with behavioral/ emotional problems 5,001 4,617 9,618 Students with more than one characteristic of disability 5,518 5,094 10,612 Total 91,077 84,072 175,149 Source: Office of Education Council 2005 Another group that has remained at the Education Council (OEC) covering 250 chil- margins of inclusive education policies is dren of alien workers in Samut Sakhon comprised of children of non-Thai citizens province shows that the provision of free currently living in Thailand. According to education has resulted in a 46 percent MOEûs and Ministry of Interior (MOI)ûs regu- decline in the number of school dropouts lations, non-Thai children have a right to and child labor abuse cases. Yet nearly receive basic education with financial sup- half of the schools which were providing port from the RTG. While this policy is in free education for alien children had not effect, it has been irregularly implemented. received state subsidies. Demand-side constraints keep a large share of non-Thai children out of school. As noted earlier, the pattern of school par- Language of instruction has also been a ticipation in Thailand reveals that the major- problem as alien children may not be fluent ity of school abandonment is occurring in Thai. Efforts to reach immigrant out-of- between cycles. Figure 2.D suggests that school youth have been small in scope and transitions between cycles tend to be sharp- mostly led by specialized non-governmental er in rural areas in particular, although these organizations. Furthermore, schools that do have diminished between 1994 and 2002. enroll non-Thai children oftentimes do not Possible explanations for this phenomenon claim their entitled governmental per capi- may relate to basic supply constraints in ta funding for these students due to igno- which students simply do not have access rance of existing policies or fail to receive to the next education cycle and are forced their entitled allocation. At present, there to drop out. Between-cycle abandonment are approximately 45,000 non-citizens in Thai may also be due to low perceptions of the schools receiving budgetary per capita enti- benefits of the subsequent education cycle. tlements. A study of the Office of Low real or perceived returns to education, 38 high private costs, or high opportunity costs Figure 2.I). Financial constraints tend to may cause students not to enter higher lessen as students progress to higher grades; education cycles. Lack of relevance or poor while relocation grows in relative signifi- school quality can also play an important cance into the upper secondary level. role in a household decision to keep their Nevertheless, relocation also shows a dimin- children in school. ishing impact when moving along the high- er grades. The çotheré category in this sur- MOE data indicate that while financial con- vey probably captures lack of knowledge straints are often identified as the main about the reasons for dropout due to an obstacle keeping students out of schools, inability to track students outside the formal another important factor is student reloca- educational system. tion (without transfer to a new school) (see Figure 2.I: Reasons Cited for Dropping Out, Grades 7-12, 2004 Source: Ministry of Education 2006 A multivariate analysis model was construct- tion weigh more heavily into participation ed to determine how demand side indica- decisions. In 1994, there was a probability tors affect school participation rates of chil- of 12 percent that children would dropout dren ages 12 to 17 years old. The effect of school if they were one year older. This of age and household composition on probability declined to just 5 percent in schooling decisions have waned. In gener- 2002. Likewise, it is common that children al, as children grow older, they stand a from larger households have less of a higher chance of dropping out of school as chance of attending secondary school, the direct and opportunity costs to educa- most likely because limited resources have ThailandSocialMonitor:Improving SecondaryEducation 39 to be allocated among a greater number participation respectively in 1994. While of children. Recent household data show effects have decreased over the years (and that the effect is presently marginal and has stand at 5.5, 8.0, 8.8 and 9.9 respectively for declined over time. 2002) they remain statistically significant. An alternative specification that replaces quin- Children with educated parents tend to tile dummy variables with the (natural) log achieve higher secondary school participa- of per capita household expenditure sug- tion rates. The educational attainment of gests that a 10 percent increase in month- adult males and females are both positive- ly real per capita expenditure is associated ly associated with childrenûs participation in with almost a 1.5 percentage point increase secondary school. The higher the educa- in childûs school participation in 1994. This tional attainment of the adult in a house- magnitude is about 1 percentage point for hold, the likelier the children in that partic- 2002 data. ular household will participate in school. The effect is even stronger with females. Direct costs to education act as a barrier The marginal effect of adult educational to access to secondary schooling. The tran- attainment has increased over time sition from one level of education to anoth- between 1994 (2.5 percent) and 2002 (3.5 er is heavily influenced by financial consid- percent). erations. According to the Children and Youth Survey (CYS), regularly conducted by The relationship between school enrollment the NSO, inability to pay is overwhelmingly and household expenditures has remained the prime reason for children who are com- strong. Relative to children from the poor- pleting an educational cycle not to est quintile, children from the second, the progress to the next educational level (see third, the forth and the fifth (the richest) Figure 2.J). Other minor reasons include quintiles have 7.3, 11.9, 17.0 and 18.3 per- having to work, distance from school and centage points higher probabilities of school have sufficient knowledge for work. Figure 2.J: Reasons for not Continuing Lower and Upper Secondary Education, 2002 Source: Children and Youth Survey 2002 40 The gap between the rich and the poor for spending per student at the secondary level household expenditures in secondary edu- was equal to about 2,300 Baht in 2002, but cation is widening. Average household again this estimate masks wide variations spending on education was 3,449 Baht per across income quintiles (see Table 2.5). year in 2002 with vast differences across Households in the poorest quintile spent quintiles - from 840 Baht in the poorest quin- about 860 Baht per secondary students as tile to over 7,870 Baht in the richest quintile compared to over 6,800 Baht in the richest - and across regions - from about 1,750 Baht quintile - close to eight times as much. This in the Northeast to over 9,585 Baht in gap appears to have widened over time. Bangkok. Poor households allocate on the In 1994 household per secondary student order of 1.5 percent of total expenditures to spending in the richest quintile was only education whereas richer households allo- about 4 times as much as that in the poor- cate 2 or 3 percent. The SES does not col- est quintile; in 1999 it was 6 times as much lect information about spending on educa- and by 2002 it had grown to 8 times as tion at the student level, but estimates can much. Moreover, while spending in second- be derived indirectly by calculating the ary education in real terms by the poorest average increase in household expenditures quintile between 1994 and 2002 experi- associated with an additional child enrolled enced a downward trend, spending by the at each level of education. The average other four quintiles increased. Table 2.5: Private Expenditure Estimates on Education by Income Quintile (Real Baht), 1994-2002 Source: Household Socio-Economic Survey 1994-2002 ThailandSocialMonitor:Improving SecondaryEducation 41 çFreeé education for all does not truly meet Figure 2.K). Meals and transportation costs the needs of poor households. While the combined represented about 65 percent of RTG committed to provide 12 years of free total education-related expenditures. A education for all under the 1997 recent study from the National Human Constitution, tuition covers only a small part Rights Commission also indicates that stu- of total expenditure that households must dents often had to pay çhidden feesé to uti- bear in association with education. Data lize libraries, have access to computers and from the CYS show that in 2002, school sit for examinations. Other students report tuition and textbooks represented only 19 to being charged for school landscaping or 25 percent of the total cost of sending a after-school tuition classes (Thai News child to lower or upper secondary level (see Agency 2006). Figure 2.K: Average Annual Expenditure on Education per Person by Education Levels, 2002 Note: Average annual expenditure for upper secondary level includes both aca- demic and vocational streams. Source: Children and Youth Survey 2002 Opportunity costs may hinder households perceive the impact of income foregone, from sending children to school. A large which is often affected by conditions in the number of rural children, especially in poor labor market. For the poorest households, households, financially contribute to their this indirect cost may be substantial, with families by working inside or outside the seasonal variations relating to the demand home. By sending these children to school, for agricultural and other labor. In Thailand, households forego these financial benefits. this is apparently a problem for only a very Opportunity costs relate to how households small fraction of children at the lower sec- 42 ondary level. According to the SES, children precluding school attendance at a higher begin to become involved in productive rate in the upper secondary level. In 2002, work at around 15 years of age (see Figure 17 percent of 16-year olds were engaged 2.L). Data from the CYS confirms that only in productive work and 14 percent of chil- a small minority of out-of-school youth cite dren in Grade 9 cited work as a reason to having to work as a reason not to enter stop their education at the lower secondary lower secondary education. As expected, level. work begins to compete with schooling by Figure 2.L: Share of Children and Youth at School or Work by Age, 2002 Source: di Gropello 2006 ThailandSocialMonitor:Improving SecondaryEducation 43 But in order to achieve universal secondary INTERVENTIONS TO STIMULATE education, it is crucial to ensure not only EQUITABLE SECONDARY SCHOOL that there is strong demand but also suffi- EXPANSION cient supply of educational services. In the early 1990s, a school infrastructure develop- So what accounts for the extraordinary ment program was pushed forward by the growth in secondary school enrollments in Department of General Education (DGE). the 1990s? As noted immediately above, There were on average 18 public second- the expansion of the schooling network, ary schools in every province. In addition, either through a new infrastructure program the Office of National Primary Education or the addition of lower secondary grades Commission (ONPEC) expanded existing to existing primary schools, is credited in rural primary schools under its jurisdiction to dramatically opening new educational cover lower secondary education. opportunities for children throughout Expanding schools into the rural areas Thailand. Studies conducted in rural areas enabled students to attend schools by provide evidence that parents saw second- reducing transportation costs for households. ary education as a means to break the With such heavy expansion, the enrollment cycle of poverty for their children or open gap between children in the urban and up greater job opportunities in the expand- rural areas was dramatically reduced. But ing labor market. Bringing schools closer to while secondary schools seats have grown the point of demand reduced the costs over time, a simple analysis shows that there associated with schooling and minimized may still be room for expansion. Comparing concerns about their children falling into the number of students in the official sec- bad company. The use of excess physical ondary level age group between 12-17 facilities and teachers in primary schools, years old and the number of secondary resulting from the declining primary school- classrooms available by province, it is clear age population, facilitated the transition to that more classrooms may be needed. The lower secondary education. The primary student to classroom average ratio ranges schools prioritized for expansion had to be from 27:1 in Pattani to 42:1 in Nontaburi and successful in teaching vocational subjects Samutprakarn. For many provinces, if all where students could earn income while students in the respective age group decid- studying, staffed by an adequate number ed to attend secondary school, there would of teachers and located in areas with low not be enough classrooms to hold them. continuation rates due to poverty (Jones 2003). The number of primary schools sup- plying secondary education increased from 119 in 1990 to 2,688 in 1992, 4,081 in 1994 and 6,281 in 1997. Despite this very dramat- ic expansion, as the rough estimations already provided suggest, Thailand may still face a supply side shortage to offer ade- quate opportunities for all children. Financial incentives were introduced to reduce the cost burden on families. While the Government has expanded its school network and committed to provide free tuition for grades one to twelve, expendi- 44 tures for school-related costs such as uni- government-initiated programs, the student forms, textbooks and transportation are loan program, established in 1996, receives large and significant. In order to address more than 185 billion Baht and is the largest demand-side constraints, the RTG launched education-related intervention. More than 2 a comprehensive set of programs to assist million students have received loans to poor children in response to the evidence attend upper secondary and tertiary schools that lack of financial support is a significant to date. constraint to secondary school participation. The main features of these programs are described in Box 2.3 below. Among the Box 2.3: Summary of Government-Initiated Interventions Policy Intervention Target Population Target Level Total Budget School Lunch Program Children with nutri- P r e - p r i m a r y Around 4,000 Objective: The fund, established in 1992, aims to tional problems, in and primary million Baht in ensure sufficient nutrition for pre-primary and pri- Rachaprachanukrao, levels FY 2005 mary students across countries, including pre-pri- in border police mary children in mosque and temple schools. schools, from hill Each student receives 6 Baht per day for 200 days tribes and the in each school year, totaling 1,200 Baht per per- poor, which son per school year. This program has also been accounts for 30 replicated independently with community contribu- percent of all chil- tions across the nation. dren. Responsible agencies: Ministry of Education Current situation: In 2005, more than 2 million stu- dents received free school lunches Scholarships for Poor Children from Essay Writing Poor students All levels Objective: Scholarships for poor children, using Governmentûs revenue from lottery tickets sales. Scholarships are granted to children who have submitted an essay describing hardships they face. Scholarships are awarded after verifying atten- dance with schools and teachers. Responsible agencies: Ministry of Education (Office of Basic Education Commission, Office of Vocational Education Commission, Commission on Higher Education) tracks students and monitors project outcomes. Current situation: Currently, around 250,000 chil- dren have received a scholarship with 150,000 applications being processed. In total, the pro- gram is currently using about 67 percent of its allo- cated resources. ThailandSocialMonitor:Improving SecondaryEducation 45 Box 2.3: Summary of Government-Initiated Interventions Policy Intervention Target Population Target Level Total Budget Scholarships for Poor Students whose parents are Poor children of All levels dedicated to social activities government offi- cials and perma- Objective: The project aims to help poor students nent employees of whose parents have devoted their time and efforts every ministry, vol- to government and social activities. unteers on special taskforces (public Responsibility agencies: Ministry of Education, health, anti-drugs), Office of Lottery Service junior police offi- cials Current situation: The project started in 2004. There were 66,560 scholarships granted. Bicycle Lending Project Poor students who Primary and Around 500 Objective: Beginning in 2003, students who live in have to travel long secondary lev- million Baht remote areas at least 3 kilometers away from distances to school els has been allo- schools and whose family annual income is less cated. than 300,000 Baht can borrow a bicycle to travel to schools. The borrowed bicycles are returned to the project upon graduation. Responsible agencies: Ministry of Education and Ministry of Industry. Current situation: 426,734,545 Baht have been allo- cated and 75,900 bicycles have been provided on loan for students Student Loan Scheme Objective: Established in 1996, the fund aims to provide financial assistance to children from poor Children from poor Upper second- 26,045 million families who wish to continue to upper secondary households whose ary and terti- Baht in FY 2004 or tertiary level on both general and vocational family total annual ary levels tracks. This also includes learners in non-formal income is less than education programs who wish to further their stud- 150,000 Baht (for ies beyond lower secondary. The loan provides recipients before 55,400 Baht per year for upper secondary and 1998 their annual 100,000 Baht for bachelor degree studies. household income should be no more Responsible agencies: Ministry of Education is than 300,000 Baht) responsible for loan recipients. Ministry of Finance is responsible for allocation of loans while Krung Thai Public Company Limited is responsible for debt repayment from students after graduation. Current situation: In 2004, around 880,000 students were recipients, using around 26 billion Baht of stu- dent loan funds. 46 Box 2.3: Summary of Government-Initiated Interventions Policy Intervention Target Population Target Level Total Budget One District One Scholarship Program Poor students Tertiary level Allocation of whose family Objective: The project aims to address issues of both in-country 1 billion Baht income is less than equal access to quality education for Thai stu- and overseas on an annual 100,000 Baht per dents. Scholarship applicants are poor students basis. year whose family income does not exceed 100,000 Baht per year. In addition, these students are required to have a minimum of 3.00 GPA, pass the general test (on mathematics, science, social sci- ence), pass an English writing test and interview. Responsible agencies: Office of Lottery Service and Ministry of Education. Current situation: In 2005, the first year of the pro- gram, 921 students have received scholarships, out of which 740 have studied abroad and 181 have attended universities in country. The second batch of 2006 is in process. Source: Office of National Economic and Social Development Board (NESDB) and Ministry of Education Government interventions have proliferated, any assistance and an additional 600,000 but their impact could be improved. In poor youth were out of school (Ziderman 2000, about a quarter of students at the 2003). Furthermore, more recent evidence upper-secondary level (general and voca- suggests that loans do not seem to reach tional) received student loans (including 40 its intended beneficiaries: students who percent of students enrolled in private voca- would be unable to attend schooling in the tional upper secondary). Yet the loan pro- absence of a loan. Interview data shows gram itself did not produce the expected that only 30 percent of borrowers would boost in overall upper secondary education drop out if the loan was terminated. enrollments. An in-depth study of the According to a recent study by Student Loan Scheme shows that over the Chulalongkorn Universityûs education faculty, period 1996-2000, upper-secondary educa- 58 percent of secondary school recipients tion evidenced a declining overall share in did not need financial assistance to stay in terms of number of borrowers and total size school, while 21 percent of rejected appli- of loan disbursements (from 63 to 47 per- cants at all educational levels were actual- cent and 46 to 21 percent respectively). In ly in a worse off financial position than 1999, the loan scheme covered approxi- scholarship recipients (Bangkok Post, 2006). mately 453,000 children or 33 percent of In addition, there is a problem with slow poor students enrolled - that is, two thirds of repayment and low recovery rates. eligible poor students enrolled in upper sec- Comparing with other Asian countries, ondary schools that year did not receive Thailand has the lowest repayment rate (21 ThailandSocialMonitor:Improving SecondaryEducation 47 percent), contrasted with 55 percent in dents. At present, Non-Formal Education Korea and 79 percent in China. Moreover, (NFE) programs are prohibited to serve chil- the recovery rate, which includes default dren under the age of 16, unless they are and administrative costs, is at 8 percent referred by an ESA. The NEA encourages considered to be very low, compared with schools to organize flexible classes for out- 53 percent in China (Ziderman 2004). To of-school youth children and youth, but this ensure that the student loan program is an practice is not very widespread. In 2002, effective intervention, the RTG needs to over 4 million children were enrolled in NFE urgently improve its outreach and targeting institutions at all educational levels. Most of to the poorest students. these students attended schools under the jurisdiction of the MOE; hence they are Scholarships may act as a stronger incentive included in counts of participation and for poor families to send their children to gross and net enrollment rates collected by secondary school, while cutting down relat- the MOE as well as the SES. Unfortunately, ed administrative costs of the scheme. data on specific age groups are not avail- Indeed, the Government has announced an able at the moment. It is thus difficult to intention to replace loans for upper second- estimate how many children aged between ary students by grants and use an Income 12 and 17 enroll in alternative education Contingent Loan (ICL) for tertiary education. programs. Data from ONEC indicate that The ICL is scheduled to start this academic approximately 1.8 out of 4 million were at year 2006. Student loans for upper second- the secondary education level. Alternative ary students are expected to be gradually education pathways provide opportunities phased out and replaced by student grants for those who cannot attend schools during for grade 10 students. regular hours, and such students tend to come notably from disadvantaged families. There are no prominent schemes for assist- The Vocational Education Certificate Course ing poor students at the lower secondary involves a community-based curriculum over level. Official assistance programs targeting a three-year period. An evaluation con- lower secondary students in need are rela- ducted in 2000 showed that only 30,000 stu- tively small and unfocused. For instance, dents per year opted for this option (Jones the bicycle lending program, which does 2003). Increasing the role of credible alter- not specify an education level, tends to native education programs may provide overwhelmingly support students at the pri- enhanced opportunities for children from mary level. Lower secondary students poor households and further strengthen the account for 25 percent of recipients in this equity dimension of the educational system. scheme and upper secondary students for only 5 percent. Thus, while many of the Promoting private sector involvement may programs do target poor households, some alleviate some of the educational expendi- interventions could be better directed or ture demands on the Government. The specifically earmarked to at-risk lower sec- expansion of public schools places addition- ondary school students in order to redress al burden on government resources, espe- the notable drop in enrollments in the tran- cially when the government is committed to sition beyond primary schooling. providing 12 years of free education for all students. Promoting the expansion of pri- Increasing the role of alternative education vate schools, especially in urban areas, service delivery programs may provide fur- could potentially alleviate pressures on gov- ther opportunities for rural and poor stu- ernment expenditure on education, allowing 48 the government to redirect additional the student population in 1992 to 85 per- resources to rural areas or better target out- cent by 2002. In other words, the private of-school children. At the lower secondary sector share in secondary education has school level, over 90 percent of students either remained largely stagnant or dimin- attend public schools (see Figure 2.M). ished over the past decade, suggesting that Public sector provision of upper secondary there is room for improvement. education grew from around 72 percent of Figure 2.M: Ratio of Number of Students Attending Public to Private Schools at Secondary Level, 1992-2002 Source: Ministry of Education 2002 Educational reforms have stimulated private Thailand could draw lessons from several partnerships. Currently various schemes financing mechanisms implemented in other have been put in place to promote private countries. Programs such as the Programa , partnerships in the education system. For Nacional de Educacion, Salud y , example, any individual or organization Alimentacion (PROGRESA) in Mexico or which establishes a school or institution is Bolsa Escola in Brazil have proven to be permitted to deduct 30 percent of the prof- effective mechanisms to stimulate demand its from the operation on a tax free basis. for education and encouraging households Additionally, incentives such as tax rebates to send children to school. They are or exemptions are provided for contributions expected to produce higher school atten- from non-profit organizations. A Revolving dance rates while lowering school dropout Fund for Developing Private Higher rates. The most comprehensive programs, Education Institutions was launched in 1999 such as PROGRESA in Mexico, resulted in a to provide loans to private sector agents. A reduction of child labor, increased educa- closer analysis of the impact of these meas- tional attainment and improved health and ures would be desirable to assess their via- nutrition of the most impoverished. This par- bility for expanding educational opportuni- ticular program targets the poorest popula- ties. tion in rural areas and provides monetary assistance to each child under 18 years old that enroll in school between grades 3 and ThailandSocialMonitor:Improving SecondaryEducation 49 9. Each year the grant amounts increase to the family income. More than 97 per- as the student progresses to the next level. cent of eligible families choose to partici- This grant compensates for a householdûs pate in the program. foregone income as their children attend school instead of working and contributing Box 2.4: Demand Side Financing Mechanisms Demand side financing programs raise family income and reduce the cost of attending school. They have also proven to be cost effective. Such programs can be targeted to specific populations iden- tified to be in greatest need, such as households below a certain income level or living farther away from school. Incentive schemes are proving to be very effective in increasing access to secondary education. This is true in high enrollment countries such as Mexico as well as countries where enroll- ment rates, particularly for girls, have historically been very low, as is the case in Bangladesh or Cambodia. A variety of demand side financing mechanisms include : A stipend is a cash payment that a public agency makes to a family to offset schooling relat- ed costs for a child. By granting a poor student a tuition-earmarked stipend, for example, the local education authority reduces parentsû direct cost of sending this child to school. For reducing the direct costs of schooling, the benefits of schooling will outweigh costs and the childûs family will elect to send the child to school. Stipends act to increase demand for schooling by lowering the direct and/or indirect costs of education and making it more likely that the utility maximizing option for a family will be to send a child to school. Stipends can be earmarked for core expenses such as books, tuition and transport, and incidental expenses such as materials, game fees and cloths. These can also be known as scholarships or conditional cash transfers. A targeted voucher is a cash payment given directly to students/families, typically by a public entity, to be used at a school that the student selects. The primary reason voucher programs are typically considered is to increase the range of choices available to students, thereby benefiting both students and school systems. By making more schooling options available to students, it is possible that families will be able to choose an alternative that better meets their needs. Targeted bursaries are cash payments that may go directly to schools, municipalities or provinces and are earmarked for specific purposes, such as improving the curriculum or increasing school access for minority, indigenous or poor children. They are not given to students or their families but are made available to financial officers or the bursars at school or relevant administrative offices. When target- ed bursaries are used to reduce or eliminate school fees, their effect is essentially the same as a stipend. Community grants are given to a community of students in a lump sum but are tied to attend- ing a community created institution. The term voucher-like is sometimes used because the amount of money is related to the number of students and the approach has an element of choice. Parents choose to send their children to a community school, thereby making the community school eligible for cash payments. Payments may cover some expenses or the full cost of schooling. The grants can be used, for instance, to address gender equity issues by giving scholarships in lump sum to schools that girls attend. Source: A Review of Demand Side Financing Initiatives in Education (Patrinos 2002) 50 Special groups of children also require addi- administrators with some basic tools to sup- tional resources. Although children who port children with disabilities and strive have a physical or intellectual disability are towards an inclusive school system. These small in number, they often require support have been important first steps in the right as well as additional resources and tailored direction, but a more comprehensive and services.5 Schools have been incentivized clearly articulated policy to mainstream chil- and supported to attract and serve children dren with disabilities will be required to with disabilities by providing increased per achieve Education for All targets. Other capita expenditures in addition to alloca- barriers to educational access include inac- tions for the general student body under the cessible school infrastructure to accommo- NEA. Additionally, mainstreaming strategies date children with disabilities and negative for inclusive education within the existing social perceptions about disability that have curriculum and teacher training programs tended to keep these children at home or have provide teachers, principals and in segregated institutions. ThailandSocialMonitor:Improving SecondaryEducation 51 III.QUALITYOFSECONDARY EDUCATIONINTHAILAND 52 Accurately capturing and evaluating the International assessments are attempting to various facets of educational quality is diffi- ameliorate this weakness, at least partially, cult. UNESCO has defined education qual- by testing skills and practical applications of ity around four principal concepts, (i) learn- content rather than content alone. One of ing to know; (ii) learning to do; (iii) learning the principal international assessments, the to live together; and (iv) learning to be OECD Programme for International Student (UNESCO, 2005). Education quality encom- Assessment (PISA), does this by measuring passes multiple areas of learning, ranging content çliteracyé, a concept that encom- from content knowledge of both external passes how 15 year-old students apply and local or indigenous subjects, skills to knowledge and skills; how they identify, apply what is learned in the larger society solve, and interpret problems; and how they and labor market, qualities to build more analyze, reason and communicate. The cohesive, peaceful, and equitable societies, Trends in International Mathematics and and opportunities to develop personally. Science Study (TIMSS) is another internation- Unfortunately, many of these areas of edu- al assessment. TIMSS is a curriculum-based cation quality are difficult to assess, meas- test for mathematics and science adminis- ure, or compare. Because of this, most tered to eighth-grade students (typically 14 research is confined to the more conven- to 15 years of age). Together with PISA, tional area of content knowledge; and as these tests have proved to be valuable and UNESCO states, çIt could be judged unfor- reliable instruments for measuring education tunate that the quantitative aspects of edu- quality comparatively across countries and cation have become the main focus of explore the reasons that affect student per- attention in recent years for policy makers.é formance (see Box 3.1). Thailand has par- ticipated in TIMSS assessments in 1995 and 1999 and in PISA assessments in 2000 and 2003. Box 3.1: Measuring Quality of Education across Countries In the late 1950s, the International Association for the Evaluation of Educational Achievement (IEA) was formed. It initiated what would become a major set of studies aiming to measure cognitive achieve- ment at various levels of education in several countries and to identify the main causes of differences in outcomes. Twelve countries joined its first mathematics study. By 2000, some 50 countries were par- ticipating in surveys covering mathematics and science (now called the Trends in International Mathematics and Science Study or TIMSS), reading (the Progress in International Reading Literacy Study or PIRLS) and other subjects. Strongly influenced by the IEA experience, several other such studies, usu- ally regionally focused, have since been established. They include the Programme for International Student Assessment (PISA), set up by the OECD in 1998 and now covering 59 mainly industrialized and middle income countries; the Southern and Eastern African Consortium for Monitoring Educational Quality (SACMEQ), which since its first survey in Zimbabwe in 1991 has expanded to 15 African countries; the Latin American Laboratory for the Assessment of Quality in Education (LLECE), which began in 1997 and covers sixteen countries; and the survey in French-speaking Africa known as the Programme dûAnalyse des Systemes Educatifs de la CONFEMEN (PASEC). At present, both UNICEF and the World Bank are sponsoring separate East Asian regional training programs with selected countries to strengthen nation- al capacity to regularly monitor and assess student achievement. ThailandSocialMonitor:Improving SecondaryEducation 53 The RTG has also recognized the impor- QUALITYOFTHAIEDUCATION: tance of educational quality in realizing the LESSONSFROM potential of all young children and main- INTERNATIONALDATA taining long-term economic competitiveness. Thus, it has placed improving the quality of This section uses international assessments to education as one of its top priorities. analyze education quality in Thailand com- Thailand has adopted several national pared to other countries in East Asia and mechanisms for monitoring student learning Latin America. An important conclusion and assessing progress in educational stands out from benchmarking the perform- achievement. First, every Thai student must ance of Thai students internationally: take a school-based mid-year examination Thailand has higher scores than other coun- at the end of the first semester and a final tries at similar income levels, suggesting that examination at the end of the school year it has been generally successful at providing in order to pass on to the next grade. educational services of certain quality equi- Second, standardized national tests at the tably. end of primary, lower secondary and upper secondary levels were introduced in the The bars in Figure 3.A show the proportion year 2000 by the Office of Education of students in participating East Asian and Assessment and Testing Services, under the Latin American countries divided into PISAûs Department of Curriculum and Instruction six proficiency levels for the PISA 2003 exam Development, and were carried out until in mathematics. The line running through the 2003. These tests were comparable in figure shows the mean test score for each measuring student performance within and country. The figure groups countries accord- across provinces. They were known as the ing to the World Bankûs income groupings. General Aptitude Test (GAT) and the Figure 3.B shows the same information for Scholastic Aptitude Test (SAT). The GAT the PISA 2000 exam in reading literacy. In covered Thai, mathematics, science and this case, student scores were divided into English in grade 6; grade 9 students had to five rather than six proficiency levels. sit for an additional social studies exam; while all 12 graders had to take a Thai, In both years and in both subjects a clear English and social studies exam and addi- trend of lower income countries scoring tional subjects such as general science, below higher income countries is present. chemistry, biology, physics and mathematics Thus, Thailand trails significantly behind its depending on studentsû study programs. Asian counterparts: Japan, Hong Kong, SAR The SAT, only applicable to Grade 12 stu- and Korea. This is not surprising. The aver- dents, covered 4 skills of language, mathe- age 2003 PISA mathematics test score for matical calculation, analytical thinking and lower middle income countries (381) is more scholastic aptitude. At present, the National than 150 points below that of upper income Institute of Education Testing Services (NIETS), countries (542). This indicates a real quality founded in 2004, is responsible for the eval- gap between wealthier countries, such as uation and testing of Thai education at all Japan and South Korea, and developing levels, including secondary education. In countries, like Thailand and Indonesia. The academic year 2006, NIETS introduced the same pattern is true for 2000 PISA Reading Ordinary National Educational Test (O-NET) Literacy, where the average test score for and the Advanced National Educational lower middle countries (385) is 139 points Test (A-NET) at grade 12. These are below that of upper income countries (524). required tests for university admission. 54 Figure 3.A: PISA 2003 Test Score Results in Mathematics Literacy by Income Group Source: di Gropello 2006 Figure 3.B: PISA 2000 Test Score Results in Reading Literacy by Income Group Source: di Gropello 2006 However, when analyzing the performance of above the other countries in its income group- Thailand against other lower middle income ing in the 2003 mathematics test. While 20 countries, such as Indonesia or Brazil, we percent of Thai students scored at or above observe that Thailand performs better than its proficiency level 3, notably fewer Brazilian, peers both in terms of mean test scores as Indonesian or Mexican students reached that well as proficiency level distribution. In fact, benchmark. A similar phenomenon can be Thailand performed more than 50 points observed in the 2000 reading literacy exam. ThailandSocialMonitor:Improving SecondaryEducation 55 This substantial congruence in the results of This point is further buttressed when regress- both tests over time suggests that Thailand ing mean PISA test scores against log GDP has been more successful at producing bet- per capita. The trend line in Figure 3.C ter student outcomes - at least as is meas- shows the expected performance of coun- ured by the skills and contents of the PISA tries at a given income level. Thailand exam - than other countries at similar clearly performs above the trend line; that income levels. is, student achievement is higher than would be predicted by its GDP level. Figure 3.C: Trend Line of PISA Test Scores against Log GDP per Capita Source: di Gropello 2006 The results of the 1999 TIMSS study show a national income levels, Thailandûs perform- very similar pattern to the PISA results. ance could be rated as adequate (see Thailand ranked 27th in mathematics and Figure 3.D and Figure 3.E). In both mathe- 24th in science among 38 participating matics and science, Thailand ranks above countries (statistically equal to the interna- the trend line for its per capita Gross tional average), among the bottom tier of National Income (GNI) level. participating countries. Yet, relative to Figure 3.D: Trend Line of TIMSS Mathematics Scores against 2003 GNI per Capita Source: Richter 2006 56 Figure 3.E: Trend Line of TIMSS Science Scores against 2003 GNI per Capita Source: Richter 2006 Not only Thai students on average perform expected, there are statistically significant well relative to their peers in other countries differences in test scores across wealth quin- at similar income levels, but the distribution tiles in all countries. Individuals from wealth- of knowledge across Thailand is fairly equi- ier households perform better than individu- tably distributed. Figure 3.F graphs the per- als from poorer households. But in the case formance in mathematics of individuals from of Thailand, these differences are less different wealth quintiles in PISA 2000, but sharply marked than in other countries illus- the basic results hold for other subjects and trated, such as Argentina, Brazil and Chile. years. The wealth variable is based on an Wealth accounts for a very modest share of index of several household asset-ownership the total variation in mathematics scores 6 and housing conditions variables. As overall. Figure 3.F: Average Mathematics Performance by Wealth, 2000 Note: Q1 refers to the poorest quintile, while Q5 refers to the richest quintile. Source: di Gropello 2006 6 These include the availability of a dishwasher, a room of their own, educational software and a link to the Internet and the number of cellular phones, television sets, computers, motorcars and bathrooms at home. This type of variables has been used in many other studies as a good proxy for household welfare in the absence of consumption information (see Filmer and Pritchett, 1999). ThailandSocialMonitor:Improving SecondaryEducation 57 Differences across schools can make a dif- Thailand presents the least share of ference in how much students learn and between-school variation in mathematics for how they perform. In an equitable educa- the countries selected. Much of this tional system, between-school differences between-school variation can be explained are minimized; that is, all students have an by differences on average socioeconomic equal opportunity to learn regardless of background of student population among what school they attend. Figure 3.G shows schools. But the sorting of students across that most of the variation in student per- schools by wealth is less pronounced in formance in Thailand is due to within- rather Thailand than in most other comparison than between-school differences. In fact, countries. Figure 3.G: Between- and Within-School Variation in Mathematics Scores, 2000 Source: di Gropello 2006 This finding provides an additional explana- by a correlation coefficient of 0.6). tion for Thailandûs significant gains in educa- tional attainment across individuals from all On the other hand, while PISA and TIMSS socioeconomic backgrounds. The growth in results suggest that Thailandûs performance student participation can be explained not is acceptable given its income level and only strictly in terms of an increase in access knowledge is rather fairly distributed, it also to schooling but also to greater equity in underscores that there are problems with access to quality schooling, as low quality education quality that demand urgent schooling provides little incentive to remain attention. First, it is clear that very few chil- in school. Comparing the secondary com- dren score in the top proficiency levels. For pletion profiles with test scores as measured PISA, less than 10 percent of students through the PISA assessment for the five scored beyond levels 4 in mathematics or countries for which we have both types of reading. This is in stark contrast to all three information (Indonesia, Mexico, Argentina, participating East Asian upper income coun- Thailand and Brazil), there is confirmation tries, where roughly 50 percent of students that higher secondary completion is gener- in mathematics and 40 percent in reading ally related to higher quality (as illustrated scored above this level. The same holds 58 true for TIMSS. Only 16 percent of students extensive background data to appraise the in Thailand reached the upper quarter pro- relationship between various educational ficiency benchmark in the mathematics test, inputs and outcomes. We now turn our while in Singapore, Chinese Taipei, Korea, attention to the contributions of teacher, Hong Kong, SAR and Japan at least 64 per- school and household characteristics to stu- cent of students performed at this level. dent learning and performance. Furthermore, a very large share of students Teacher Characteristics is performing below acceptable proficiency levels. Thailand had roughly 40 percent of Teacher quality is considered one of the students performing at or below the PISA most important contributing factors to level one in literacy and over 50 percent of improving student achievement (Rice 2003; students performing at or below the PISA Rivkin, Hanushek and Kain 2005). In 2005, level one in mathematics. This contrasts about 87 percent of secondary school greatly with the upper income countries teachers under OBEC had earned at least where only around 10 percent of students a bachelorûs degree. Only a small minority score at or below level one. In summary, of teachers in lower and upper secondary a vast proportion of students are function- schools possessed a Masterûs degree or ing at or below the most basic level of lan- above - about 4 and 11 percent respective- guage, mathematics and science ability. ly. The Bangkok Metropolitan Region (BMR) and Northern provinces tended to have a greater concentration of teachers with a FACTORSAFFECTINGQUALITYOF Masterûs degree or higher (see Table 3.1), EDUCATIONINTHAILAND while teachers from other regions tended to have comparatively less education. The dis- What do we know about school inputs that tribution of teachers with higher education affect educational outcomes? In the sec- degrees is also more heavily biased towards tions below, selected educational input vari- schools that cater children of higher income ables are discussed in order to better under- brackets, while schools that serve lower stand which factors positively affect educa- income populations have teaching staff with tional outcomes in the Thai context. fewer years of professional training. International assessments have collected Table 3.1: Percent of Teachers with Masterûs Degree or Higher, 2002 Region Lower Secondary Upper Secondary BMR 5.9 18.6 North 5.4 15.5 Northeast 2.6 8.4 Central 3.3 9.7 South 3.3 7.3 Whole Kingdom 3.8 11.0 Source: Ministry of Education ThailandSocialMonitor:Improving SecondaryEducation 59 What is the evidence regarding Thai teach- students were taught by instructors who felt ersû contributions to student learning? less than adequately prepared in science. About 80 percent of students interviewed by Only 13 and 18 percent of students were the PISA 2003 test reported that their teach- taught by instructors who felt confident in ers showed an interest in every studentûs their abilities in science and mathematics learning, gave extra help when students respectively. Higher levels of teacher con- needed it, helped students with their learn- fidence were associated with superior stu- ing, continued teaching until students under- dent scores. stand and gave students the opportunity to express opinions (PISA 2004). These are sub- Traditionally, Thai secondary classrooms jective perceptions of teacher performance, have relied on a front-style lecture format, rather than an accurate description of rather than on child-centered pedagogical teachersû attitudes and classroom practices. methodologies or active learning environ- Nonetheless, they portray an overall positive ments. On average, Thai students spend general classroom environment in which stu- about one fifth of class time each month in dent learning can be adequately nurtured. lecture-based classes and another one-fifth on teacher-guided student practice. Such On the other hand, on the basis of data distribution of class time is similar to aver- from questionnaires completed by school ages reported by students from other coun- principals, an index of teacher adequacy tries participating in TIMSS. The distinction was constructed reflecting perceptions between Thailand and others lies in the very about the extent to which teacher supply limited opportunities for Thai students to hindered student learning. Thailand was apply knowledge to new situations or exer- amongst the countries where principals cise creative thinking during the school day. reported shortage or inadequacy of teach- In response to a question which asked to ers. For instance, 37 percent of students identify activities frequently undertaken in had principals who believed instruction in the classroom, 91 percent of students mathematics was hampered by teacher reported that teachers actually showed inadequacy while 24 percent reported lack them how to do their mathematics work. of experienced teachers (PISA 2004). Moreover, poor studentsû perception of More strikingly, only 19 percent of Thai stu- teacher-student relations - that is, students dents reported working independently on who disagree that most teachers are inter- class mathematics projects, in contrast to ested in their well being, treat them fairly or Hong Kong, SAR (67 percent) or Korea (46 provide help when needed - had on aver- percent). The international average was 36 age a strong and significant negative cor- percent. More than 90 percent of Thai stu- relation on mathematics performance (PISA dents practice computational skills in most 2004). mathematics lessons; yet, only 54 percent of students report that their teachers explain TIMSS constructed an index based on the reasoning behind an idea and a mere teachersû own perceptions of their ability to 12 percent report the use of tables, charts teach various topics (e.g. properties of geo- or graphs to represent and analyze relation- metric figures, solving linear equations, ships. While 93 percent of students report earthûs features, chemical reactivity, etc.). that mathematics teachers use the black- In 1999, 55 percent of students were taught board for presentations, only 33 percent of by instructors who felt less than adequately students get called to the front of the class, prepared in mathematics and 58 percent of compared to an international average of 60 60 percent. A similar trend can be observed cational practices do not come about from in science instruction. one off workshops, but are the product of sustained capacity development efforts. An On the other hand, science instruction ongoing pilot program has sought to train appears to adequately emphasize scientific leader teachers (facilitators) and subject investigation and practical application. A teachers through innovative distance train- majority of Thai students report conducting ing models. Leader teachers and educa- experiments in science classes. Over half of tional supervisors also provide continuous students report that they spend more than support through roving teams. They observe 25 percent of class time in practical demon- classrooms, make recommendations and strations always or pretty often, in contrast provide advice with teaching planning. This to an international average of 38 percent is a promising model. Research evidence and largely at par with Singapore, Japan, suggests that among alternative quality Malaysia and New Zealand. improvement interventions at the school level, teacher training investments can have Overall, these findings from TIMSS and PISA high payoff in terms of student retention, indicate that there is a need for targeted promotion and, particularly, student learning teacher support and skills upgrading policies (see Rowe 2003 for example). to complement existing shortfalls and imbal- ances in teacher performance. The hetero- School Characteristics geneous distribution of confidence in teach- ing capacity suggests limited opportunities The average student teacher ratio in sec- for professional development. Enhancing ondary education is 28:1, ranging from as teacher professional development could low as 12 to 1 in Phetchabun to as high as potentially translate into significant improve- 37 to 1 in Chiang Rai. Although student ments in student flows and learning. teacher ratios generally stand below 30:1, class sizes in Thailand are notably larger. There is an emerging recognition within the The average secondary school class size for MOE that lasting changes in teacher edu- Thailand is 43 students (see Figure 3.H). Figure 3.H: Student Teacher Ratio and Class Size in Secondary Schools by Region, 2002 Source: Ministry of Education ThailandSocialMonitor:Improving SecondaryEducation 61 The relationship between class size and stu- large classes performed just as well or slight- dent achievement is difficult to extricate. ly better than their counterparts in smaller Although TIMSS and PISA data concur that classes. PISA results corroborate these find- large class sizes do not appear to have a ings. No negative effects of adding stu- negative impact on student test scores in dents could be discerned at any relevant Thailand, there are many other factors influ- class size in the PISA literacy or mathemat- encing higher scores achieved among stu- ics exams (PISA 2003). In Thailand, large dents from larger classes. Class size in East classes are generally better equipped and Asia tends to be larger than the internation- in highly competitive schools. In those al average, ranging from 36 in Japan to as schools, students tend to come from higher high as 50 in the Philippines. In Thailand, socio-economic backgrounds, therefore, about 75 percent of eight grade students receiving better support for education at tested in TIMSS enrolled in mathematics home. The number of schools with class classes with 36 or more pupils, while less size larger than 40 students, as specified by than 5 percent enrolled in a class with 1 to the MOE, is only about 300 secondary 20 students. Interestingly, students from larg- schools (from approximately 10,000 schools er classes achieved better scores in both in total) throughout Thailand. Most schools mathematics and science (see Figure 3.I). with small class size tend to be located in This was also applicable to other Asian rural areas where students are from less priv- countries, including Japan, South Korea, ileged backgrounds and cannot afford to and Hong Kong, SAR where students in attend large schools in cities. Figure 3.I: Cross-National Comparison of Average Mathematics Achievement and Class Size, 1999 Source: TIMSS 1999 62 PISA 2000 collected extensive data on prin- more important obstacle to student cipalsû perceptions about the adequacy of achievement than school physical infrastruc- school resources (PISA 2003). Based on their ture. Figure 3.J shows that, according to responses, a distinction can be drawn Thai school administrators, the quality of between the impact of physical infrastruc- school physical infrastructure is rated around ture (condition of buildings, the instructional the OECD country mean, while school edu- space, and heating, cooling and lighting cational resources are rated more than two systems) and educational resources (com- thirds of a standard deviation below the puters, library material, multi-media OECD mean (-0.82). The marginal effect resources, science laboratory equipment of an increase by one unit in the index of and facilities for the fine arts) to student the quality of education resources is associ- learning. School principals perceived the ated with an increase by about 11 points in quality of educational resources to be a reading literacy scores. Figure 3.J: Perceived Adequacy of Physical Infrastructure and Educational Resources Source: PISA 2003, p. 195 A key area of concern is that Thailand is in The impact of school resources for instruction the lowest decile among all PISA participating on student achievement in mathematics and countries with regards to the difference in the science is complex, but overall TIMSS data quality of school resources between the top also suggest that they play an important role and bottom quarter of schools as character- in advancing student learning. TIMSS created ized by a proxy of school socioeconomic an index of availability of school resources for background (PISA 2005). Moreover, the distri- mathematics and science instruction. This bution of school resources is inequitably distri- index includes general instructional items - buted among schools. There is a statistically such as basic materials, budget for supplies, significant performance difference between instructional space-as well as more specific schools in the top and bottom quarters of this mathematics and science instructional items - educational resources index. such as computers, library materials and audio-visual resources. Over half of the stu- ThailandSocialMonitor:Improving SecondaryEducation 63 dents in Thailand report that shortages Thailand and translate into higher learning affect instructional capacity some or a lot, outcomes. compared to 18 to 20 percent internation- ally. Less than 40 percent of eighth grade Household and Individual students had access to a calculator in Characteristics mathematics class, compared to an interna- tional average of 73 percent. Not surpris- It is widely acknowledged that higher fami- ingly, 85 percent of students have never ly income and household resources have a used a computer in class. Countries that positive effect on student achievement. tended to report shortages in the availabili- Some of the obstacles that disadvantaged ty of instructional materials were significant- students face include poorly educated par- ly below the international average in math- ents, greater demand to supplement family ematics or science achievement. income through labor, peers with low aca- demic performance and lesser expectations A study focused on the relationship about the future. between schooling inputs and student learn- ing performance in Thai public schools also Parents with more education tend to place points in the same direction. Annual aver- greater emphasis and provide more support age public expenditure in 2001-2002 for sec- for their childrenûs education. According to ondary education ranged between 15,000 TIMSS, eighth grade students with parents and 16,000 Baht per student. Per student who are well-educated tend to score bet- expenditure varies across regions, with the ter in mathematics and science tests. This Northeast receiving the lowest share. A pro- was the case across Asia, including duction function analysis using data from Thailand. PISA 2003 data also support these the Bureau of Budget and the MOE sug- findings. Students whose parents have more gests that an increase in non-salary recur- education score higher across subjects - rent expenditures per student at the provin- reading literacy, mathematics and science. cial level results in a higher mean on nation- For example, students whose mothers al test scores (Punyasavatsut et al. 2005). obtained an upper secondary degree On average, for secondary education, an increased their raw test scores for mathe- increase in per pupil expenditure by 10,000 matics (32 points), reading (28 points) and Baht raises the mean of national test scores science (28 points), compared with those by 4.8 points. whose mothers only have primary or lower secondary education (see Figure 3.K). Thus, the evidence gathered by PISA, TIMSS and other research persuasively suggests that there is a shortage of resources for learning in Thai schools and this is generally perceived as a constraint to higher student achievement. Greater investments in basic school resources in order to provide a min- imum set of materials for effective use by teachers to support instructional content could well buttress student performance in 64 Figure 3.K: Mean Scores by Level of Mothersû Education Source: PISA 2003 Yet, as noted earlier, the differences among along socioeconomic lines. Low-performing socioeconomic groups are not as stark in students are not concentrated among lower Thailand as in other countries. For instance, income quintiles necessarily. Rather, given the variance in student performance on the Thailandûs low levels of academic achieve- PISA mathematics test was more than 15 per ment in absolute terms, policies to raise the cent below the OECD average variance (PISA overall performance level of the general stu- 2003). Although children from wealthier dent population are imperative, such as households perform better than children from enhancing teacher preparedness and prac- poorer households, the distribution of knowl- tice, providing adequate instructional materi- edge is fairly equitable across income groups. als, promoting performance incentives for school staff, encouraging greater accountabil- Along these lines, parental education is not ity for results or strengthening a formative an influential factor shaping student aspira- evaluative culture. tions for tertiary education in Thailand. Students tended to have high expectations Early childhood care and development pro- independently from their parentsû educational grams are often cited in research literature as background. Although 30 percent of Thai stu- an important contributor to readiness to learn dents had a parent who did not finish primary at primary school entry as well as to later school and another 40 percent had a parent academic success (see for example Evans who did not complete upper secondary 2000). Evidence from PISA provides further school, 55 percent of students expressed an support to this body of work in the Thai con- intention to graduate from university (TIMSS text. Thai students who attended pre-school 1999). for at least one year showed a modest advantage in PISA test scores in mathematics, Thus, it appears that efforts to improve student even when controlling for socioeconomic performance need not be primarily targeted background factors (PISA 2004). ThailandSocialMonitor:Improving SecondaryEducation 65 QUALITYASSURANCE in 2001. As of March 2005, nearly 98 per- MECHANISMSINTHAIEDUCATION cent of primary and secondary education schools had been evaluated. The assess- ments monitor student academic perform- Since the NEA was promulgated, the quali- ance, teacher performance, administratorsû ty of the Thai education system is officially vision and leadership as well as school lab- evaluated both internally and externally. oratories and equipment. Internally, schools are expected to conduct self-assessments of their institutional quality. The NEA authorizes ONESQA to submit cor- This process is meant to be woven into the rective measures and actions for schools administrative apparatus of each institution. that are performing poorly in order to Schools are expected to prepare annual improve their function. If an institution con- reports which are then submitted to govern- tinues to perform poorly, a report is submit- ment agencies and made available to the ted to the Basic Education Commission for public. further action. But anecdotal data suggests that the review process, frequency and External assessments are meant to comple- type of the advice provided for school ment and follow up on school self-assess- improvement by external independent eval- ments. The NEA established ONESQA as the uators is insufficient to generate actual agency responsible for overseeing quality changes in schooling practices. There is assessment studies of every educational room for ONESQAûs overall supervisory func- institution at least once every five years (see tion to be further strengthened in order to Box 3.2). These results are shared with re- translate the monitoring and evaluation levant agencies and made available to the process into tangible improvements in edu- public, with the first assessments conducted cational quality. 66 Box 3.2: The Office for National Education Standards and Quality Assessment ONESQA is the public organization established by the 1999 NEA to serve external quality assessment institutions along with support to internal quality assessments conducted by the MOE. ONESQA con- sists of an Executive Committee, the Committee for Development of an Assessment System for Basic Education, and the Committee for Development of an Assessment System for Higher Education. ONESQA espouses the following operational principles: i. Emphasis on quality and efficiency in functioning ii. Promotion of educational institutions for their development to the highest potentiality iii. Establishing an information system for linking with concerned agencies iv. The assessment is aimed at quality enhancement, not passing judgment for either reward or pun- ishment v. External assessment must be transparent and supported by concrete evidence vi. Private, professional or academic organizations will be encouraged to participate in the training of external assessors vii. Selection of external assessors, based on the criteria of competency for accreditation and registra tion, as well as periodic assessment of these assessors viii. Networking with other parties concerned in order to create an awareness of the collective respon sibility for educational quality ONESQA also seeks to promote school self-evaluations through a variety of mechanisms, such as a dis- tance education via satellite and trainings for directors and deputy directors of education area offices. ONESQAûs budget for FY04 was approximately USD16 million. Source: Excerpted from http://www.onesqa.or.th (ONESQA 2006) The MOE has also spearheaded selected and administration are shifting and being research initiatives to further review and redefined. Sustained coordination and strengthen Thailandûs educational quality cooperation between government agencies assurance mechanisms. A project initiated and educational institutions is vital for by the ONEC, now the OEC, strives to bet- improving quality assurance and enhance- ter understand how government agencies ment mechanisms in Thailand. and schools can work together to improve student outcomes. In-depth data was col- There has also been increasing recognition lected from 250 schools in 2001 and 2002. that the school governance structure can One of the most notable findings is that play an important role in school improve- education activities across different ministries ment, in particular the role that school com- need to be better harmonized. While munity leaders can exert (Gamage and recent reform efforts have attempted to Sooksomchitra 2004). A pilot program streamline the administrative and opera- examining different types of school boards tional work of government agencies, schools to explore ways to effectively improve the still consider the system confusing and less efficiency of local government and its role than adequate. This situation has been fur- in supporting education development is ther compounded by the ongoing educa- ongoing. Training modules for school tion decentralization process, where the boards and administrators have been devel- roles and responsibilities for service provision oped and are being evaluated for possible ThailandSocialMonitor:Improving SecondaryEducation 67 implementation nationwide. learning are and timely identify shortcom- ings in education service delivery so that Thailandûs newly established framework for interventions can be put in place to support educational quality assurance is promising, the instructional function of schools, while but at present it is still in an incipient devel- parents must play a more active role in opment stage. Teachers and principals school decision-making so that as partners need better information about student per- in the schooling process they can better formance so that they can adequately articulate their needs or aspirations as well respond to the educational needs of stu- as hold schools accountable for the quality dents, policymakers need to understand of educational services received. what the conditions that positively influence 68 ThailandSocialMonitor:Improving SecondaryEducation 69 IV.EFFICIENCYOFTHE THAIEDUCATIONSYSTEM 70 This chapter outlines the basic characteris- PUBLICSPENDINGONEDUCATION tics of educational spending in Thailand, including both public and private sectors, The RTG spends more than one fifth of its with a focus on secondary education. total budget on education, but education Secondary education financing in Thailand allocation has shown a declining trend in is confronted by the dual challenges of recent years. Over the past decade, the expanding access while improving quality. Thai government allocated more than 20 It is crucial not only to secure sufficient percent of its total budget to education, resources to finance the system, but also to accounting for 4 to 5 percent of the coun- ensure that those resources are apportioned tryûs GDP. This level of allocation was main- in the most productive manner. Outlined tained even through the economic crisis in below is an overview of salient issues on the late 1990s. The budget allocation for secondary education financing, followed by education as a share of GDP and total selected recommendations for possible budget peaked in 2000 but has experi- improvements in allocative efficiency. enced a slight declining trend in recent years (see Table 4.1). On the other hand, the share of basic education in the overall education budget has been steadily main- tained, with a slight upward trend. Table 4.1: Education Budget, 2000-2005 Fiscal Year 2000 2001 2002 2003 2004 2005 Education 220,621 221,603 222,990 235,092 251,233 262,938 budget (million Baht) As % of total 25.7% 24.4% 21.8% 23.5% 24.4% 21.9% government budget As % of GDP 4.5% 4.3% 4.1% 4.0% 3.9% 3.7% Basic education 67.2% 68.1% 68.1% 69.3% 71.5% 70.2% budget as % of total education budget Basic education 3.0% 2.8% 2.8% 2.8% 2.8% 2.6% budget as % of GDP Source: Bureau of Budget, Office of Education Council and World Bank calculations ThailandSocialMonitor:Improving SecondaryEducation 71 More than 80 percent of the education More than two thirds of the education budget is allocated to current expenditure. budget is allocated to basic education, with In 2003, education had the largest average pre-primary and primary levels receiving the per capita current spending rate of all sec- largest proportion. Education expenditure is tors. Per capita current spending was at divided into six categories: pre-primary and 2,425 Baht, four times more than the sector primary; secondary; tertiary; services not with the second largest current spending defined by level; educational support; and rate, public order and safety affairs (NESDB miscellaneous. Almost 50 percent of edu- 2003). Sixty percent of current expenditure cation resources are allocated to pre-pri- is used to pay for salaries for permanent mary and primary education levels, with and temporary staff, while the rest pays for another one quarter spent on secondary non-personnel items such as remuneration education and 15 percent on higher edu- and other services. Education budget cation (see Figure 4.A). Resources allocat- trends show that increasing resources have ed to educational support have increased been allocated toward current expenditure recently, rising from 6 billion Baht in 1996 to in recent years. Although capital expendi- more than 28 billion Baht in 2003. tures are not made on a regular basis, its share has been declining over time, from 25 to 6 percent between 1997 and 2002. Figure 4.A: Share of Education Budget by Spending Category, 1997-2004 Source: Ministry of Education Although basic education (primary and sec- hand, tertiary education commands close to ondary levels) received almost 70 percent of 15 percent of the budget while only 9 per- the education budget, it accounts for 90 per- cent of the total student population is enrolled cent of all students enrolled in the Thai edu- at this level. cation system (see Table 4.2). On the other 72 Table 4.2: Education Budget Allocation and Student Enrollment by Spending Category, 2002 Level Total (million Baht) Percentage of budget Percent of enrollment (GER) Pre-Primary 98,228 41.7% 62% and Primary Secondary 64,770 27.5% 29% Tertiary 33,348 14.2% 9% Services not 3,377 1.4% defined by level Education 28,868 12.3% support Miscellaneous 6,854 2.9% Total 235,444 100% 100% Source: Office of Education Council 2004b Thailand allocated approximately 28 percent ed by very low shares of private resources of its total education budget and 1.13 per- coming to secondary education compared to cent of GDP to secondary education in 2003, other countries. For example, while the falling behind what countries with strong sec- Philippines only invested 0.69 percent of GDP ondary education sectors typically spend. On in secondary education, the share of private average, OECD countries spent 2.1 percent of resources was 84 percent or 0.58 percent of GDP in secondary education. Lower-middle GDP. In Thailand, private sector contributions income countries allocated on average 40 amounted to 0.06 percent of GDP, an equiv- percent of their total education resources and alent of 5 percent of the public sector financ- 1.86 percent of GDP to the secondary level. ing (see Table 4.3). This is a level too low to The limited resourcing of secondary education be efficient. through public financing is further compound- ThailandSocialMonitor:Improving SecondaryEducation 73 Table 4.3: Total Secondary Education Expenditure as Percent of GDP by Source of Funding, 2003 Share of Private Income Level Public Resources Private Resources Resources in all Domestic (% of GDP) (% of GDP) Resources Argentina Upper-middle 1.58 0.38 24% Chile Upper-middle 1.49 0.70 47% Colombia Lower-middle 1.53 1.00 65% Hong Kong, SAR High 1.50 0.75 50% Indonesia Lower-middle 0.48 0.28 58% Mexico Upper-middle 1.52 0.40 26% Philippines Lower-middle 0.69 0.58 84% Korea High 1.83 0.60 32% Thailand Lower-middle 1.13 0.06 5% Source: UNESCO 2005b Analysis of unit costs shows that per student Local governments are reliant on subsidies public spending on secondary education is from the central government to finance lowest. Per student spending usually rises education. The RTG has encouraged the as students matriculate from primary to sec- decentralization of educational manage- ondary to tertiary levels. In East Asia and ment in order to improve local participation Pacific, on average, putting a student and ownership. Decentralization also through secondary education during the includes increasing the share of local gov- school year 2002-2003 was 74 percent more ernment resources spent on education. expensive than primary education. These However, most local agencies still rely on countries spent USD1,711 on each primary subsidies from the central government to level student and USD2,409 on each sec- finance education. In general, resources ondary student (in purchasing power parity from local government only comprise 20 to converted terms). Among OECD countries, 30 percent of education spending. each primary school student cost USD4,818 Additionally, recent data from the MOE and each secondary school student cost 39 show that subsidies from the central govern- percent more at USD6,688. Yet in Thailand, ment have been increasing in recent years, the unit cost for secondary education is the opposite of what should have been lower than the unit cost for primary educa- observed (Punyasavatsut et al. 2005). tion. In 2000, the unit cost for pre-primary and primary was 13,770 Baht, for secondary 8,564 Baht and for tertiary 32,336 Baht. Similar trends persist in 2002 data: per stu- dent spending in primary was estimated at 13,226 Baht and 10,011 Baht in secondary. 74 Figure 4.B: Sources of Revenue for Local Governments, 1997-2004 Source: Bureau of Budget Budget allocation to local levels needs to student appears to advantage the poor at be more equitable. Although guidelines for the primary level, this is not the case at the how to allocate the national education secondary level (Punyasavatsut et al. 2005). budget at the provincial level exist, they are not strictly followed. In practice, provincial Bangkok and its vicinity accounts for the budget allocations are calculated based on largest share of total educational expendi- how much each province received the pre- ture. Unlike any other region, however, vious year. As a result, provinces that start- educational expenditure is concentrated at ed off receiving larger shares of the nation- the tertiary level in Bangkok, totaling more al education budget continue to receive than pre-primary, primary, secondary, voca- proportionately more each year, regardless tional and others combined. The Northeast of the number of schools, students or teach- region spent the most on pre-primary and ers. Additionally, findings from a study fund- primary levels and only about a tenth of ed by the Asian Development Bank point that amount on tertiary education. out that budget allocation at the second- However, a breakdown of data by per stu- ary level favors wealthier provinces and that dent spending paints a different picture (see this trend has remained constant over time Table 4.4). Because of the large population (Cresswell 1999). On the other hand, the of school-aged children in the Northeast government tends to abide by the funding region, per student cost is 2,705 Baht com- formula for the allocation of resources to pared with 3,953 Baht for Bangkok and its ESAs, which takes into account factors such vicinity. At the secondary level, while per as enrollment, number and type of schools, capita educational expenditure in Bangkok student teacher ratio and number of class- and its vicinity is as high as 764 Baht per rooms. Yet when ESA budget allocations head, the Northeast received only 575 Baht are disaggregated by Gross Provincial per head, the lowest share across regions. Product (GPP), while average spending per ThailandSocialMonitor:Improving SecondaryEducation 75 Table 4.4: Per Capita Educational Expenditure by Region (Baht), 2002* Region BMR Central Northeast North South National National Average Average (w/o BMR) Pre-Primary 706 1,595 1,737 1,718 1,724 1,536 1,699 and Primary Secondary 764 699 575 623 760 663 643 Vocational 251 248 142 214 260 208 200 Tertiary 1,960 151 180 442 349 537 257 Others 271 96 71 110 87 118 87 3,953 Total 2,791 2,705 3,107 3,180 3,062 2,886 * Divided by entire cohort population, not the corresponding school-age groups. Source: Comptroller Generalûs Department While education receives the largest share 4.C). However, analysis of data from the of the national budget across sectors, SES and the Comptroller Generalûs whether those resources are equally and Department (CGD) shows secondary educa- efficiently distributed among different tion expenditure as slightly favoring the income groups is debatable. We have esti- wealthy (see Figure 4.D). Spending for ter- mated expenditure shares by per capita tiary education is clearly regressive, regard- income quintiles. The results, disaggregated less of which data are examined. The by levels of education, show that the poor- wealthiest 20 percent of the population est 40 percent of the population receives 56 receives 53 percent of total spending. Thus, percent of total spending in pre-primary although Thailandûs educational expenditure and primary levels, reflecting a pro-poor pattern is proportionately distributed across allocation of resources. The quintile distribu- income quintiles as a whole, there are some tion for secondary level is also distributed notable discrepancies in ensuring an equi- relatively equally and slightly skewed table distribution, particularly at the tertiary towards giving more to the poor (see Figure level. Figure 4.C: Incidence of Public Expenditure across Income Quintiles by Education Level, 2002 Source: Household Socio-Economic Survey 2002 76 Figure 4.D: Distribution of Public Spending on Education by Level, 2000 and 2002 Note: Estimates by geographically disaggregated unit-cost. Source: Household Socio-Economic Survey (2000-2002) and CGD (2000-2002) HOUSEHOLDEDUCATION resources comprise about 95 percent of the EXPENDITURE domestic education budget for secondary education, with only a tiny share provided by local governments, and 5 percent origi- The Government is not the only source of nate from private resources (UNESCO educational financing in Thailand. 2005b). This section focuses on the current Households, the private sector, non-profits state of household spending on education, and international organizations can all con- with a particular focus on differences tribute significantly to improving the Thai between wealthy and poor families. secondary education system (for examples, see Box 4.1). As already noted, public Box 4.1: Private Participation in Education: Examples from Korea, Africa and Chile In the Republic of Korea, the government introduced subsidies and tax exemptions in the 1970s. Such subsidies were based on the difference between the schoolûs own budget and a standard budget for a public school of the same enrollment size and type. Private school institutions responded to the subsidies and tax exemptions by scaling up their capacity to reach more students. Consequently, as of 2000, private sector enrollment had reached 20 percent for middle schools and 55 percent for high schools. In Africa, fees have been maintained at the secondary level, even in countries where free primary education is provided, such as Malawi, Uganda, Tanzania and Kenya. Such fees at the secondary level, however, are controlled and increases beyond stated levels must be cleared with the Ministry of Education. The cost burden to poor families, as well as to those with female children or living in hard to reach areas, are sometimes subsidized through bursaries or scholarships. Chilean schools were previously not allowed to charge fees if they were a public institution or a pri- vate school that received government subsidies. This policy was changed in the mid-1990s and now private schools are able to charge fees within certain parameters and continue to receive subsidies. The Chilean system ensures that schools charging fees exempt a proportion of parents from fee pay- ment and to use some of the revenue to set up internal cross-subsidies from fee-paying parents to non-fee-paying parents. Source: Expanding Opportunities and Building Competencies for Young People: A New Agenda for Secondary Education (World Bank 2005) ThailandSocialMonitor:Improving SecondaryEducation 77 Average household spending on education those in other regions. Education as a in 2002 was 3,449 Baht per year, but this share of household expenditure also varies expenditure varies dramatically across across regions. While on average house- regions. It ranges from 1,753 Baht in the holds in Bangkok spend between 2.3 to 3.9 Northeast to over 9,585 Baht in Bangkok percent on education, the share of house- (see Table 4.5). A trend analysis over time hold educational expenditure in the shows households in the Northeast consis- Northeast ranges between 1.3 to 1.9 per- tently spending less on education than cent. Table 4.5: Household Expenditure on Education by Region, 1994-2002 1994 1996 1998 1999 2000 2002 a. Average Household Expenditure on Education (per year, real Baht*) Expenditure Quintile Bangkok 6,278 6,564 9,666 10,553 10,428 9,585 Central 2,111 2,436 3,291 3,654 2,979 3,280 North 1,702 2,218 2,742 3,958 2,655 2,922 Northeast 1,429 1,579 1,887 1,933 1,665 1,753 South 2,300 2,411 2,815 3,043 2,948 2,882 Total 2,452 2,738 3,426 3,912 3,394 3,449 b. Share of Total Household Expenditure (%) Expenditure Quintile Bangkok 2.3 2.3 3.6 3.9 3.7 3.2 Central 1.4 1.5 1.8 1.9 1.7 1.6 North 1.4 1.7 1.9 2.5 2.1 2.0 Northeast 1.3 1.5 1.7 1.9 1.7 1.6 South 1.6 1.7 2.0 2.2 2.1 1.8 Total 1.5 1.7 2.0 2.3 2.1 1.9 * Prices are deflated by regional and yearly CPIs (Base region = Bangkok, Base year = 2002). Yearly CPIs: 1994 = 75.0, 1996 = 84.1, 1998 = 96.0, 1999 = 96.3, 2000 = 97.8, 2002 = 100 Source: Household Socio-Economic Survey 1994-2002 78 Poorer households spend less on education. cate around 1.5 percent of total household Average household spending by per capita expenditures to education, compared with income quintile shows that the poorest quin- 2 or 3 percent in wealthier households. The tile spends only 840 Baht on education financial burden on poor households is com- while the wealthiest quintile spend over pounded during times of economic crises. 7,870 Baht (see Table 4.6). The difference A World Bank study examining the impact in spending across income quintiles is of the Asian Crisis on educational spending reflected not just in the absolute levels of in Thailand finds that an increase in per spending on education but also as a share capita income of one baht raised educa- of total household expenditures. Given that tion expenditure by 0.12 baht. Larger poor households spend most of their income households spent less on education per on food and other basic needs, fewer capita than smaller households (World Bank resources are allocated for education. 1999). Hence on average, poor households allo- Table 4.6: Household Expenditure on Education by per Capita Income Quintile, 1994-2002 1994 1996 1998 1999 2000 2002 a. Average Household Expenditure on Education (per year, real Baht*) Expenditure Quintile Expenditure Quintile Poorest 777 934 1,023 1,044 1,048 840 Second 1,145 1,314 1,432 1,538 1,415 1,322 Third 1,611 1,939 1,928 2,108 1,930 1,822 Fourth 2,445 2,916 3,455 3,325 3,024 3,214 Richest 4,981 5,268 7,237 8,940 7,471 7,878 Thailand 2,452 2,738 3,426 3,912 3,394 3,449 b. Share of Total Household Expenditure (%) Expenditure Quintile 1.8 1.3 Poorest 1.3 1.5 1.6 1.7 1.8 1.5 Second 1.4 1.5 1.6 1.8 4.8 1.5 Third 1.5 1.6 1.7 1.9 2.0 2.0 Fourth 1.6 1.8 2.2 2.2 2.7 2.7 Richest 1.8 1.9 2.7 3.3 2.1 1.9 Thailand 1.5 1.7 2.0 2.3 * Prices are deflated by regional and yearly CPIs (Base region = Bangkok, Base year = 2002). Yearly CPIs: 1994 = 75.0, 1996 = 84.1, 1998 = 96.0, 1999 = 96.3, 2000 = 97.8, 2002 = 100 Source: Household Socio-Economic Survey 1994-2002 ThailandSocialMonitor:Improving SecondaryEducation 79 As noted already in Chapter 2, wealthy cate a primary school child, while they households are allocating more resources spent 2,350 Baht for a secondary school stu- towards each secondary school student dent. The premium paid by households for and the spending gap between the rich secondary education does not make up for and the poor has widened over time. the relative decline in public financing, Results from an estimation of per student even if primary and secondary education spending from the SES show that real spend- had similar per student costs. In summary, ing per student has increased between 1994 as noted in an earlier section, Thailand lags and 2002 at all levels of education, but behind in secondary education resourcing notably at the primary and tertiary levels by international standards. (see Table 2.4). This jump is not observed at the secondary level. Furthermore, there Despite relatively low levels of financial is a wide variation in household secondary resourcing, Chapter 3 documents that Thai expenditure across per capita income quin- students perform on average above pre- tiles. In 2002, households in the poorest dicted levels when taking into consideration quintiles spent about 860 Baht per second- Thailandûs economic developmental stage. ary school student, compared with over Moreover, the distribution of knowledge 6,800 Baht in the wealthiest quintile, approx- across Thailand is fairly equitably distributed. imately 8 times as much. This gap appears These successes on student outcomes would to have widened over time. In 1994, house- indicate that funding for secondary educa- hold expenditure per secondary school stu- tion has been put to work in a relatively effi- dent in the top quintile was about 4.5 times cient manner. as much as that in the poorest quintile, while in 1999 it was 6 times as much. Yet, it is also apparent by Thailandûs per- formance in international assessments that RAISINGEFFICIENCYOF student achievement levels are low. A vast SECONDARYEDUCATION proportion of students are functioning at or FINANCING below the most basic level of language, mathematics and science ability. Poor aca- Examining the efficiency of the Thai educa- demic performance and low educational tional system may be a key factor in quality have increasingly raised concerns improving student outcomes. On the one regarding Thailandûs competitiveness relative hand, Thailand allocates a significant share to more developed East Asian countries. of its national income to the education sec- Secondary education is an important tool tor. On the other hand, the secondary for growth, economic development and level only captures less than a third of the social stability. This raises the question of overall public budget, while the private sec- allocative efficiency in the Thai educational tor is a relatively minor player in secondary system. education financing. It has been argued that lower per student expenditures for sec- Is there empirical evidence that financial ondary education, in contrast to primary resources act as a constraint for the education, are a reflection that school improvement of student outcomes? expenses are significantly subsidized by pri- Although results from international assess- vate contributions in the form of tuition fees ments demonstrate that the impact of or parental support. Evidence from the SES socio-economic background on student suggests that this is unlikely. In 2002, fami- performance in Thailand is significantly lower lies spent on average 1,700 Baht to edu- than the OECD average, results from PISA 80 also indicate that çamong the more lowering the overall cost of education by advanced group of students, home back- minimizing the number of teachers in order ground makes a greater difference to stu- to maximize teacher utility and raising dent performance in mathematics. In other achievement through the provision of more words, the greater the socio-economic teachers who can then give individualized advantage, the greater the advantage it attention to pupils. A study on the efficien- has in terms of student performanceé cy of expenditure in education provision by (OECD 2004, p. 182). Households from the comparing PISA 2003 results with teachers richest quintile more than double private per student and time spent at school sug- expenditures on education than households gest that Thailand is performing significantly in the fourth quintile. According to the SES, below the production function frontier for while the fourth quintile spent on average education. In other words, inefficiencies in 2,960 Baht in private education expendi- Thailand are relatively high compared to tures, the richest quintile spent 6,890 Baht. other countries. These findings hold even International experience points to the fact when correcting for GDP per capita and that quality improvements are costlier at the parental educational attainment under the secondary level than at the primary level. assumption that a wealthier and more cul- Furthermore, achieving universal secondary tivated environment are catalysts for better education will depend on making concert- student performance (Afonso and St. Aubyn ed efforts to redirect resources to enrolling 2005). poor children currently excluded. The Thai government has sought in recent Given Thailandûs already sizable investments years to merge small rural schools together, in education, is there scope for a more effi- wherever possible, in order to maximize effi- cient allocation of resources? Evidence ciency. Closer scrutiny of other utilization from the benefit incidence analysis of pub- ratios, such as the average number of stu- lic spending on education provides clear dents per class and number of classes evidence that tertiary education investments taught by teachers may also shed light on have disproportionately favored wealthier ways of increasing internal efficiency (see sectors of the population. Thus, there may Table 4.7). On the other hand, teacher be scope for further investigation about the redeployment efforts must also take into appropriate distribution of the education account that service delivery points must budget across levels in order to improve the remain within a reasonable distance from efficiency of the Thai system. childrenûs home to remain accessible. Furthermore, any teacher redeployment Another important indicator that should be effort must also consider the plea of remote examined within the context of raising effi- rural and other disadvantaged areas that ciency of schools is the student teacher experience teacher shortages and rely pri- ratio. There is a delicate balance between marily on volunteer teachers for staffing. ThailandSocialMonitor:Improving SecondaryEducation 81 Table 4.7: Utilization Ratios by Education Level, 2002 Lower Secondary Upper Secondary - General Upper Secondary -Vocational Number of 39 35 37 students per class Number of 1.63 1.59 1.48 classes taught by teachers Student teacher ratio 24 22 25 Source: Office of Basic Education Commission Conventionally, an education system is con- education. This means that for every stu- sidered internally efficient if it generates dent graduating from primary school, the maximum output at minimal cost (Hossain government spent approximately 80,000 1996). Outputs are often measured by indi- Baht over 6 years. As for secondary level, cators such as transition rates or cohort sur- public spending per student was 10,011 vival rates. In Thailand, the transition rate is Baht in 2002, thus the total cost of produc- high from primary to lower secondary level, ing one high school graduate is over 60,000 with an average of 90 percent of students Baht. In total, this means that the govern- from grade 6 continuing to grade 7. The ment spends nearly 140,000 Baht for 12 rate declines for students graduating from years of schooling. With only 40 percent of grade 9 and entering into grade 10 to students graduating from secondary school, around 80 percent. The majority of lower however, per pupil expenditure for every secondary graduates continue to the gen- secondary school graduate comes closer to eral, rather than vocational, track. 350,000 Baht. However, a closer examination by following the same cohort throughout the entire basic The NEA ensures that free basic education education system shows that only 40 per- will be available to all students through gov- cent of students entering grade 1 make it ernment subsidies, whether a student to grade 12. The RTG must continue to attends public or private schools. The fund- monitor closely transition rates, particularly ing mechanism for the provision of free at the junction between lower and upper basic education is driven by the number of secondary level. students at each school. Under the NEA, schools receive per pupil subsidies based on Retaining a larger share of students in the expected number of students each year, in education system will lower per student principle with differential formulas for differ- expenditure for every secondary school ent çtypesé of students to adjust for the fact graduate. Annual public spending per stu- that poor or other disadvantaged children dent in 2002 was 13,226 Baht for primary may require additional resources to achieve 82 a similar level of educational output. Thus, steps in the right direction, the current edu- student counts largely drive the resource cational system lacks a coherent framework allocation process. In addition, both private of checks and balances that fosters greater and public schools receive additional funds accountability relations among school and for expenditures such as equipment, build- local actors as well as central administrators. ings and special programs through this fund- The reforms proposed under the NEA are ing mechanism. Since allocation of the likely to remain incomplete until stronger education budget is dependent on the accountability measures are adequately number of expected students, it is hoped built in. that schools will become more competitive in an attempt to attract more students. The NEA articulates a vision for free basic School autonomy will be further encour- education during 9 years of compulsory aged through a proposed block grant sys- education. In addition, it proposes ambi- tem so that operational budgets, including tious targets on education service provision, both salary and non-salary costs, are direct- including a student teacher ratio adjusted ly transferred to schools. Schools then have to 25:1, a new teacher compensation struc- full authority to determine how to allocate ture, increased and better integrated use of funds. Block grants will continue to be cal- information technology, and additional culated based on the number of students, funds to encourage more children to enroll giving an incentive for schools to improve in the system. These are worthy goals. But quality and attract more students. these commitments require substantial fund- ing upfront, either from the public or the pri- However, as noted in Chapter 3, what has vate sectors. Current spending is insufficient not been fully established along with this in fulfilling these commitments. A study esti- funding mechanism is a functional account- mating the costs associated with free basic ability system to act as another avenue for education, based on several different sce- the promotion of educational quality and narios, shows that an additional 188 to 229 efficiency enhancement. Maximizing the billion Baht would be required (Thailand impact of schooling inputs can entail Development Research Institute 2000). empowering principals and administrators to Changes to the teacher compensation manage for results and making them scheme, increased use of information tech- accountable for these results. Adequate nology and the cost to attract more chil- information allows local stakeholders to dren into the education system would need reflect on existing practices, support inter- an additional 95 to 199 billion Baht. Raising ventions to bring about efficiency changes, the cost effectiveness and performance of monitor performance and demand results. the education sector will be imperative to The school external assessments conducted enhance outcomes and contain costs. under the auspices of the ONESQA are in Without serious considerations to efficiency its infancy. The feedback loops to dissemi- measures, many of the commitments prom- nate information and raise community ulgated under the NEA could remain elu- awareness on school management issues sive. are at present largely notional. The pilots on school governance and community boards are also a promising development to improve oversight on school management practices that can lead to efficiency gains, but operate on a small scale basis. Albeit ThailandSocialMonitor:Improving SecondaryEducation 83 V.ADDRESSINGTHENEXT GENERATIONOFCHALLENGES 84 The 1999 NEA addresses critical issues and The Education Provision Policy for challenges faced by Thailand with regards Disadvantaged Children lays down a vision to secondary education. Subsequently, the to bring into the formal education system National Education Plan 2002-2016 translates traditionally disenfranchised groups, such as it into a plan, guiding education-related children with disabilities and ethnic minori- agencies towards reforming Thai education- ties, through specifically-targeted programs. al administration and service delivery. This However, a clear blueprint to translate these chapter provides an overview of current ideals into a step-by-step implementation Government initiatives and policy recom- plan is missing. Key agencies responsible mendations to embrace MOEûs vision for for assuming a leadership role in such pro- greater equitable access, better quality and grams should be identified and resources more efficiency in its secondary education needed for implementation appropriately system. allocated. INCREASINGACCESSANDEQUITY Existing education guidelines regarding chil- dren living in Thailand without Thai citizen- Governmentûs initial policies to expand sec- ship need to translate into on-the-ground ondary education focused on increasing the actions. Although non-Thai children are available infrastructure to accommodate included as one of the 16 disadvantaged classrooms beyond grade 6. The use of groups who can potentially benefit from excess physical facilities and teachers in pri- existing education provision policies, actual mary schools, resulting from the declining practice is complex and unable to reach primary school-age population, helped pro- the majority of children outside the formal mote a nearly universal transition rate for system. Currently, it is reported that approx- lower secondary education. imately 45,000 non-Thai children are in school and benefiting from per capita In recent years, secondary enrollment budget allocations from the RTG. However, expansion efforts aptly shifted focus from the out-of-school population is estimated to supply- to demand-side constraints. The be considerably higher. MOE introduced a variety of schemes to tackle the cost barriers associated with The role of alternative education service schooling participation so as to realize the modalities can be strengthened. Existing commitment to çfreeé basic education pro- programs, although encouraged by the vision inscribed in the NEA. The introduction NEA, are small and do not seem to satisfy of a variety of financial incentive schemes the potential demand. Flexible education sought to reduce the cost burden on fami- arrangements can play a key role in provid- lies. As described in Chapter 2, the effec- ing opportunities to disadvantaged children tiveness of these schemes could be in accessing secondary education. Greater enhanced through better targeting. flexibility in terms of learning sites, class However, a dearth of demand-side interven- schedules and curriculum can provide a tions remains at the lower-secondary level. more suitable environment to fit the needs In order to reach current excluded popula- of children who cannot participate in tradi- tions, the Government will need to embark tional school settings, such as rural migrant on a concerted effort to address cost bar- workers. In order to ensure that alternative riers faced by the poorest children at risk of education enrollments contribute to reach- dropping out. ing formal universal secondary education goals, systematic data should be collected. ThailandSocialMonitor:Improving SecondaryEducation 85 At present, there is no available information The so-called çfive new school designsé pro- of students between 12-17 years old attend- gram, overseen by the Bureau for ing non-traditional secondary programs. Innovative Development in Education, has begun in late 2003 in a small number of Financial disparities between provinces and pilot schools nationwide on a voluntary income groups should be minimized. basis. Autonomous schools decentralize Resource allocation formulae should factor authority in academic, financial, personnel in conditions and different level of needs and general administrative activities. A par- across provinces and income groups. Policy ticipatory approach among involved parties action should be determined based on is promoted through school committees. actual data, collected at operational levels Buddhist Way Schools seek to apply teach- (i.e. school or ESA) rather than from small ings of çmorality, meditative concentration sample groups. Resources could, therefore, and wisdomé in learners as well as in school be better targeted to those who are really management. Strategic Plans for Gifted in need. Children emphasize science, music, sports and Thai performing arts for high-performing More generally, in order to include the children. Bilingual schools administer English excluded and increase school participation, Programs and Mini English Programs through efficient data collection and analysis is a a language immersion curriculum. Schools prerequisite. Data and management infor- can charge additional fees from their stan- mation systems should be able to timely dard rates. The Information and estimate children outside the formal educa- Communication Technology (ICT) Schools tion system, enabling the design of suitable integrate computers to teaching and learn- programs and providing strong evidence for ing, distance education and university- policy decision making. For instance, school linkages. although the gender gap in secondary edu- cation participation has been increasing, to In addition, the MOE has launched a sepa- the benefit of girls, there is a lack of clear rate initiative, çOne District One Lab Schoolé understanding about the reasons that are to break away with the perception that driving this phenomenon and virtually no most çqualityé schools are in Bangkok and policy discussion as to how to redress this sit- other large cities. The Lab Schools aim at uation. establishing centers of educational quality in non-urban districts so as to reduce social IMPROVINGQUALITY disparities. Their methods will include learn- er-centered approaches as well as self- The heart of Thailandûs education reform is learning skills through the use of ICT. to improve the human capital of Thai chil- dren by providing equal access to quality All these pilot programs are commendable education. In reality, schools have varied and can provide important insights about degrees of readiness and resources to pro- various methodologies and approaches to vide such education. Studentsû learning improve educational service provision, if capacity also differs. Thus, the Government properly evaluated. But these programs are has emphasized the need both to ensure all relatively small and, while they could minimum quality standards for service provi- clearly benefit certain groups of students, sion for all, while establishing a çfast tracké there are concerns as to how they can be for those students that can cope with a implemented on a larger scale. There is more challenging program. also a question as to how these initiatives 86 link to other education reform efforts in assisted instruction as possible media for Thailand and a need to avoid duplicating capacity development and the creation of previous efforts. The bottom line is that, a master teacher network in order to although these meritorious initiatives trial exchange experiences with direct assistance innovative approaches to education quality from çmobile support teams,é Rajabhat enhancement, they do not add up to a Universities and ESA staff. cohesive plan for educational quality enhancement and lack an overarching These are innovative approaches to capac- framework for systemic and progressive ity development and the on-going pilots quality improvement. suggest promising results. Conventional teacher development approaches such as Evidence gathered from international assess- seminars or mass distance training have ments indicate that there is a shortage of proven to be less effective as they do not resources for learning in Thai schools and respond to specific needs of teachers. this is generally perceived as a constraint to Again, the challenge lies on the translation higher student achievement. Greater from vision to implementation of this strate- investments in instructional resources and gic plan, so that it reaches a scale with teaching aids may be warranted in order to tangible national results and goes beyond endow schools with a minimum set of mate- ad hoc small scale initiatives. rials. The development of Local Learning Resource Centers can make an important Internal and external quality assurance contribution to expand availability of instruc- mechanisms have been established. tional resources. However, the supply of However, the relevance and quality of such materials is not a sufficient condition to assessments themselves needs to be guar- make a difference in student learning. anteed. Internal or self-assessment is new Teaching aids need to be put to use effec- to Thai culture, so it will take some time for tively by teachers to support instructional schools to assimilate this process as an content delivery. organic part of school development plan- ning and turn into a meaningful reflective It is widely accepted that teacher quality is exercise. On going capacity building efforts among the most crucial factors contributing at school and ESA levels will contribute to to student learning achievement. The institutionalize this process. Office of the Teacher Civil Service Commission has developed a strategic plan The proposed approach for direct but con- and put in place several efforts to enhance structive external assessment also has the teacher quality including pre-service and in- potential to improve education quality. The service development. A major task has MOE has adopted a cautious approach to been to develop criteria and methods to manage the çstakesé of this assessment, in upgrade teacher standards and quality order to encourage school staff to through a çwhole school approach,é where approach it openly as an opportunity for capacity building for teachers and princi- self-improvement. Again, this type of form- pals is tackled together. The Education ative review is unfamiliar to Thai administra- Reform Program aims to provide in-depth tors and educators. Thus, it will take time understanding of education management, to de-mystify the process and assimilate it. curricular change, child-centered teaching On the one hand, recommendations made and classroom research. The MOE has by external evaluators must be tangible and espoused distance learning and computer achievable. And the quality of external ThailandSocialMonitor:Improving SecondaryEducation 87 independent evaluators itself needs to be and needs. monitored and evaluated for the process to be meaningful. On the other hand, school ASSURINGEFFICIENCY staff will need to demonstrate action to turn around ineffective practices in order to The government has adopted an ambitious make a difference in student performance. organizational reform program to enhance Mechanisms to provide systematic rewards the efficiency of the national education sys- for improvements in academic or institution- tem. The NEA specifies that the financing al outcomes could be weaved into the cur- system will be restructured by providing rent system. Performance-based incentives block grants to ESAs and schools on the could provide the necessary impetus to fuel basis of a standard capitation formula in administrative and instructional behavioral addition to other per capita top ups changes. according to poverty levels and other pro- visions for disadvantaged students. Enhancing accountability systems for school performance can also operate as a strong In principle, the decentralization process is incentive to improve educational quality. expected to produce administrative savings. The authority over curricula, personnel and Estimates prepared under the RTG-World finance will be decentralized to ESAs once Bank Country Development Partnership for they have met çreadinessé criteria. Governance suggest that if all the function- Significant citizen participation is expected al decentralization targets of the to take place in the management of ESAs. Decentralization Action Plan are realized, a Additionally, pilot programs examining differ- substantial share of central government ent types of school boards to explore ways apportionments to local boards could be to enhance the relevance and responsive- potentially reallocated from administrative to ness of education service delivery to local service delivery purposes. In practice, a sig- needs are under way. Training modules for nificant share of local expenditures is cen- school boards and administrators have been trally mandated, such as personnel expen- developed and are being evaluated for ditures. These central directives could in possible implementation nationwide. fact lead to over-staffing and over-spend- ing. Moreover, ESAs are not necessarily co- Step-by-step implementation plan for terminous with local government boundaries decentralization of education provision and or responsibilities generating additional management, however, needs to be rein- administrative demands. The current lack of forced. The transfer process has been slow clarity on specific duties to be conducted to date. Teachers have been unwilling to at different government levels also carry the renounce to the terms of employment as risk of overlapping functions leading to civil servants under the MOE for positions resource waste. Overall, there has been with local governments. Local bodies need reluctance in transferring functions, so cen- to build their capacity. Existing technical tral positions have not been eliminated; assistance and institutional strengthening while local capacity to take on new func- schemes have been ad hoc and uncoordi- tions and responsibilities has been slowly nated. As administrative and service deliv- developing. Thus, at this stage, it is unclear ery functions are being devolved, a strong to what extent decentralization will indeed accountability system must be actively nur- materialize into a more efficient governmen- tured in order to foster a service-oriented tal administration. culture that is responsive to local aspirations 88 Another possible avenue for raising efficien- The private sector plays a small role in gen- cy could entail a careful review of student eral secondary education, accounting for teacher ratios. The average student 11 percent of student enrollments in lower teacher ratio in secondary education is 28:1. secondary and 20 percent in upper second- Yet while student teacher ratio remains rel- ary education. Its overall share has either atively low, class sizes are relatively large. remained largely stagnant or diminished The average secondary school class size is over the past decade. In terms of financial 43 students. The Thai government has contributions, the private sector accounted sought in recent years to merge small rural for approximately 5 percent of overall schools together, wherever possible, in order domestic secondary education resources. to maximize efficiency. However, school Mobilizing private resources can be an consolidation efforts must be balanced by important source of secondary education assurances that service delivery points will financing and could free up public remain at a reasonable distance from chil- resources for improved targeting to disad- drenûs homes and that incentives are pro- vantaged populations or service delivery vided to staff remote rural and other disad- quality enhancements. While the vantaged areas that face severe personnel Government has sought to institutionalize shortages. A functional review of teacher formal channels to encourage funding from deployment and school staffing arrange- private firms, further promotion efforts are ments may also contribute to distribute needed to increase the flow of private staffing resources more equitably across funds into secondary education. schools and alleviate staffing constraints in establishments with large overcrowded Thailand has achieved remarkable improve- classrooms. ments in education secondary provision and participation. Much has been accom- Another possibility to maximize systemic effi- plished in the last decade. The RTG has ciency may entail reallocating resources now embarked on finding solutions to the from other educational levels, such as terti- next generation of challenges: consolidat- ary, to secondary education. The bottom ing equitable access, improving quality and line is that secondary education access and enhancing efficiency. An ambitious reform quality are not likely to experience notable program is gathering momentum. improvements without an infusion of addi- Dedicated efforts and a continued focus in tional resources. ESAs will also be responsi- redressing existing systemic inefficiencies can ble for raising additional funding, but the realize the potential to fulfill the goal of a level of local revenue generation is uncer- high quality universal secondary education tain and, according to current estimates, for all Thai children. likely to be low. ThailandSocialMonitor:Improving SecondaryEducation 89 90 REFERENCES Afonso, Antonio and Miguel St. Aubyn. 2005. Cross Country Efficiency of Secondary Education Provision: A Semi-Parametric Analysis with Nondiscretionary Inputs. Working Paper Series no. 494. Lisbon, Portugal: Institute for Economics and Business Administration (ISEG), Technical University of Lisbon. Amatyakul, Poonpit, Maliwan Thammasaeng and Prayat Punong-ong. 1995. Sectoral Survey on Special Education in Thailand: Education for Children with Disabilities. Salaya, Nakorn Prathom, Thailand: Mahidol University Press. Bhangananda, Kulvitra. 2003. çEducational Development toward a Knowledge-Based Economy.é In Human Resource Development Toward a Knowledge-Based Economy: The Case of Thailand, ed. Minoru Makishima and Somchai Suksiriserekul, 41-82. Tokyo: Institute of Developing Economies (Japan External Trade Organization). Bunnag, Sirikul. 2006. çFree Schooling Cuts Abuse.é Bangkok Post: General News. http://www.bangkok post.com (accessed 3 August, 2006). Cawthera, Andy. 2005. Computers in Secondary Schools in Developing Countries: Costs and Other Issues. Washington, D.C.: World Bank. Cresswell, Anthony M. 1999. Educational Finance in Thailand: A Review and Recommendations for Improving Allocative Efficiency. Bangkok: UNESCO-Bangkok for the Asian Development Bank Social Sector Program Loan. de Ferranti, David and others. 2003. Closing the Gap in Education and Technology. World Bank Latin American and Caribbean Studies. Washington, D.C.: World Bank. di Gropello, Emanuela. 2003. Monitoring Educational Performance in the Caribbean. Washington, D.C.: World Bank. di Gropello, Emanuela. 2006. Meeting the Challenges of Secondary Education in Latin America and East Asia: Improving Efficiency and Resource Mobilization. Washington, D.C.: World Bank. Evans, Judith. 2000. Early Childhood Counts: A Programming Guide on Early Childhood Care for Development. Washington, D.C.: World Bank Institute. Filmer, Deon and Lant Prichett. 1997. Child Mortality and Public Spending on Health: How Much Does Money Matter? Washington, D.C.: World Bank. Gamage, David and Pacharapimon Sooksomchitra. 2004. çDecentralisation and School- Based Management in Thailand.é International Review of Education 50: 289-305. ThailandSocialMonitor:Improving SecondaryEducation 91 Hossain, Shaikh I. 1996. Making an Equitable and Efficient Education: The Chinese Experience. Washington, D.C.: World Bank. Jones, Gavin. 2003. Strategies and Achievements in Expanding Lower Secondary Enrollments: Thailand and Indonesia. Asian MetaCentre Research Paper Series; No. 13. Singapore: National University of Singapore. Larach, Linda. 2001. Brazil Secondary Education Profile. A Summary of Secondary Education: Time to Move Forward. Secondary Education Series no. 24555. Washington D.C.: Human Development Network, World Bank. Mass Communications Organization of Thailand (MCOT). Thai News Agency (TNA). 2006. çFree Education Not Really Free.é http://etna.mcot.net (accessed 31 July 2006). Ministry of Education. 2004. çNational Report.é In Quality Education for all Young People: Challenges, Trends and Priorities: Forty-Seventh Session of the International Conference on Education, 8-11 September 2004. Geneva: International Bureau of Education, UNESCO. Moreno, Juan Manuel. 2005. çChanges in Secondary Education: From Weakest Link to Cornerstone.é In International Semimar on Growth Strategies for Secondary Education in Asia, 19-21 September 2005. Kuala Lumpur: World Bank. National Statistics Office (NSO). 1994-2002. Household Socio-Economic Survey. Bangkok: NSO. _______. 2002. Children and Youth Survey. Bangkok: NSO. çNew Criteria Announced for Granting Scholarships.é Bangkok Post. 7 May 2006, 6. Martin, Michael O. and others. 2000. TIMSS 1999 International Science Report. Hamburg (Germany): International Association for the Evaluation of Educational Achievement (IEA). Mullis, Ina V.S. and others. 2000. TIMSS 1999 International Mathematics Report. Hamburg (Germany): IEA. Office of Education Council (OEC). 2004a. Education in Thailand 2004. Bangkok: OEC. _______. 2004b. Thailand Education Statistics 2003. Bangkok: OEC. _______. 2005. Thailand Education Statistics 2004. Bangkok: OEC. Office of National Education Standards and Quality Assessment. 2006. Functions. http://www.onesqa.or. th (accessed 27 June 2006). Organization for EconomicCo-operation and Development (OECD). 2001. Knowledge and Skills for Life: First Results from PISA 2000. Paris: OECD. 92 _______. 2003. Literacy Skills for the World of Tomorrow: Further Results from PISA 2000. Paris: OECD and UNESCO. _______. 2004. Learning for Tomorrowûs World: First Results from PISA 2003. Paris: OECD. _______. 2005a. çEducation at a Glance 2005-Tables.é Education and Training: Statistics. http://www. oecd.org (accessed 28 June 2006). _______. 2005b. School Factors Related to Quality and Equity: Results from PISA 2000. Paris: OECD. Patrinos, Harry Anthony. 2002. A Review of Demand-Side Financing Initiatives in Education. Washington, D.C.: World Bank. Phongpaichit, Pasuk and Isra Sarntisart. 2000. çGlobalisation and Inequality: the Case of Thailand.é In Poverty and Income Inequality in Developing Countries: A Policy Dialogue on the Effects of Globalisation. Paris: OECD. Pongsapich, Amara and Peter Brimble. 1999. Assessing the Social Impacts of the Financial Crisis in Thailand. Manila: Asian Development Bank. Pritchett, Lant and Deon Filmer. 1999. çWhat Education Production Functions Really Show: A Positive Theory of Education Expenditures.é Economics of Education Review 18 (2):223- 239. Punyasavatsut, Chaiyuth and others. 2005. Technical Consultancy for the Country Development Partnership Program in Education-Component 1: School Finance Reform. Bangkok: Unpublished manuscript. Rice, Jennifer King. 2003. Teacher Quality: Understanding the Effectiveness of Teacher Attributes. Washington, D.C.: Economic Policy Institute. Richter, Kaspar. 2006. Thailandûs Growth Path: From Recovery to Prosperity. World Bank Policy Research Working Paper; no. 3912. Washington, D.C.: World Bank. Rivkin, Steven G., Eric A. Hanushek, and John F. Kain. 2005. çTeachers, Schools, and Academic Achievement.é Econometrica 73: 417-458. Rowe, Ken. 2003. çThe Importance of Teacher Quality as a Key Determinant of Studentsû Experiences and Outcomes of Schooling.é In Building Teacher Quality: What Does the Research Tell Us?, 15-23. Camberwell, Australia: Australian Council for Educational Research. Suwansathit, Savitri. 2002. çQuality and Equity in Education.é In The 8th International Conference on Education: Innovations in Secondary Education: Meeting the Needs of Adolescents and Youth in Asia and the Pacific, 26-29 November 2002. Bangkok: UNESCO Asia and Pacific Regional Bureau for Education. ThailandSocialMonitor:Improving SecondaryEducation 93 Thailand Development Research Institute (TDRI). 2000. Financing Basic Education under the New National Educational Act. Bangkok: TDRI. UNESCO. 2003. The 8th International Conference on Education: Innovations in Secondary Education: Meeting the Needs of Adolescents and Youth in Asia and the Pacific, 26-29 November 2002. Bangkok: Asia and Pacific Regional Bureau for Education, UNESCO. _______. 2005a. EFA Global Monitoring Report: The Quality Imperative. Paris: UNESCO. _______. 2005b. Global Education Outlook 2005. Montreal: UNESCO Institute for Statistics. Witte, Ann Dryden and Helen Tauchen. 1994. Work and Crime: An Exploration Using Panel Data. Cambridge, MA: National Bureau of Economic Research. Witte, Johanna. 2000. çEducation in Thailand after the Crisis: A Balancing Act between Globalization and National Self-Contemplation.é International Journal of Educational Development 20: 223-245. World Bank. 1998. Thailand: Education Achievements, Issues, and Policies. Washington, D.C.: East Asia and the Pacific Region, Education Sector Unit, World Bank. _______. 1999. Thailand Social Monitor: Coping with the Crisis in Education and Health. Bangkok: World Bank. ______. 2000. Turkey - Economic Reforms, Living Standards and Social Welfare Study. Washington, D.C.: Poverty Reducation and Economic Management Unit, Europe and Central Asia Region, World Bank. _______. 2001a. çPoverty and Public Policy.é in Thailand Social Monitor Report., Bangkok: World Bank. _______. 2001b. Thailand: Secondary Education for Employment, Volume 1. Bangkok: World Bank. _______. 2002. Education and HIV/AIDS: A Window of Hope. Washington, D.C.: World Bank. _______. 2003. World Development Indicators. Washington, D.C.: World Bank. _______. 2005. Expanding Opportunities and Building Competencies for Young People: A New Agenda for Secondary Education. Washington, D.C.: World Bank. _______. 2006. ç Summary Education Profile.é EDSTATS - Global Country Data. http://devda- ta.world bank.org/edstats/cd1.asp (accessed August 30, 2006). Ziderman, Adrian. 2003. Student Loans in Thailand: Are they Effective, Equitable, Sustainable?. Bangkok: UNESCO. 94 _______. 2004. Policy Options for Student Loan Schemes: Lessons from Five Asian Case Studies. Bangkok: UNESCO. The World Bank Office, Bangkok 30th Floor, Siam Tower 989 Rama I Road, Pathumwan, Bangkok 10330, Thailand Tel. (662) 686-8300 Fax. (662) 686-8301 Comments to: Luis Benveniste Ibenveniste@worldbank.org ThailandSocialMonitor:Improving SecondaryEducation 95 Ngandee Creation Co.,Ltd Tel. (6612) 318-7462 'Ë Ò °à"«÷ß¿Y¡`À--ߢÕß°"»÷°...",,ª­» ÿª "À--ºYâ`À" ¬ ´÷Ëߪ­°Õ¥â«¬`"ߪ­«--`»" å¢Õß °"ªÆ`Yª ·­·«,â¡'Ë°`¥¢÷È,,TMà«ß­¬­«" °"¡--¬¡»÷°..."¡'»--°¬¿"æ,,°"ªì«`'"ßæËÕ ¡à"¡"'È¢Õß°"ª­¡»÷°..."·­°" "ª Y৫"¡°â"«Àâ"¢Õß--°'¬ --Èß'Èæ"­°" Õÿ¥¡»÷°..." »÷°..." "¡"æ--"--°...­·­º`·ßß"'Ë¡' 'Ë Ú ªì°"« ¿"æªí®®ÿ--¢Õß­ §«"¡TM'ˬ«TM"´÷Ëßß"¡§«"¡âÕß°"¢Õß"¥ °"¡--¬¡»÷°..."¢Õ߬,,¿"æ«¡·­,,°ÿà¡ ·ßß" 'Ë®`ß·â«°"ßÿ,,°"¡--¬¡»÷°..."®­ ª­TM"°à"ßÊ Õ°®"°'Ȭ--ß¡'°"ª'¬'¬ °Á°'ˬ«ºª­,¬TMå¥â¡"°¡ËÕ°"¡--¬¡»÷°..." ­À«à"ß¿Y¡`¿"§Õ'°¥â«¬ ®"°º--æåÀà"'È®÷ߥâ¡' "Àâ"'˪ì--«TMËÕ¡,¬ß'Ë "§--­À«à"ß°"»÷°..." ° " «` § " ­ Àå ¥â " § « " ¡ â Õ ß ° " ¢ Õ ß ° " ¢--ÈæÈ" °"Õÿ¥¡»÷°..." °--"¥·ßß" ¡--¬¡»÷°..." TMຢÕß°"--¥ `,,®¢Õߧ--«Õ Õ¬à"ß°Á"¡°"¡--¬¡»÷°..."¬--ßÕ"®ªìÕÿª § ,,°" àßÿÀ"¢â",ß'¬ "§--,,°"ªÑÕß°--°"¢¬"¬,Õ°" "ß°" 'Ë Û æ`®""ª­¥ÁªíÀ"¥â"§ÿ¿"æ¢Õß »÷°..."'Ëà"'¬¡°-- °"¡--¬¡»÷°..."¢Õߪ­» °"¡--¬¡»÷°..."·­ Õ"ßÕ°'˪쪥â,, ¬,,ªí®®ÿ--°"--ßÕ¬Yà­À«à"ß"ß Õß·æàß °"ª--ªÿߧÿ¿"æ¢Õß°"`°""ß°"»÷°..." ­À«à"ß°"¡'»--°¬¿"æ,,°"ª--ªÿß,Õ°" ,,ª­»¬ "À--¬"«TM ÀÕ°"°"¬ªì¢âÕ®"°--¥ºY°æ-- 'Ë Ù ªì°" "«®"ßÕ°,,°" àß `¡ °--°"æ--"§«"¡°â"«Àâ""ß»...°`®·­°" ª­ ``¿"æ,,°",,TMâ--æ¬"°"ß°"ß` ·¢àߢ--¢Õߪ­» ¥--ß--È®÷ߪì,Õ°" Õ--¥',, 'Ë i ªì°",,Àâ¢âÕ Õ·­¥â",¬"¬,¥¬ °"««¡§«"¡ "Á®'Ë°`¥¢÷È,,TMà«ß­¬­«" æ`®""÷ߪÑ"À¡"¬¢Õß--"·­«`«--"°" ¡à"'È,,¥â"§«"¡°â"«Àâ"¢Õß°"¡--¬¡»÷°..." "ߪ­«--`»" å'Ë "§--¢Õß­°"»÷°..."¬ ¢Õߪ­»¬ Õ¥®æ`®""÷ß `Ëßâ""¬'Ë Õ¬YàÈÕßÀâ"