Primary Education in Remote Indonesia: Survey Results from West Kalimantan and East Nusa Tenggara December 2019 © 2019 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—World Bank. 2019. Primary Education in Remote Indonesia: Survey Results from West Kalimantan and East Nusa Tenggara.© World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Primary Education in Remote Indonesia: Survey Results from West Kalimantan and East Nusa Tenggara SOCIAL DEVELOPMENT UNIT THE WORLD BANK – INDONESIA DECEMBER 2019 TABLE OF CONTENTS Table of Contents Acknowledgments ix Abbreviations xi Executive Summary xiii 01 Introduction 1 02 Scope of the Study: Locations, Instruments and Sample 5 Selection of Study Areas 5 Survey Instruments 7 Study Participants and Respondents 9 Description of Study Villages 9 03 Schooling Context 13 School Characteristics 13 School Availability 13 Student Distribution 13 School Facilities 14 School Budget 15 Language and Curriculum Utilized in Sample Schools 16 Teaching and Learning Time in Academic Year 2015/16 17 School Supervision and Meetings in Academic Year 2015/16 18 Teachers and Principals 19 Characteristics of Principals and Teachers in Study Schools 19 Living Conditions 22 Activities at School and Outside School 23 Incentives and Motivations of Principals and Teachers 29 Salaries and Allowances 29 Perceptions, Challenges, and Satisfaction of Principals and Teachers 32 iii. 04 Parent and Community Involvement in Education 37 Parents 37 Parent Background 37 Child Support and Activities at Home 38 Parents’ Expectations for Their Children 38 School Committees 41 Committee Background and Establishment 41 Committee Management 41 Involvement of Parents and Committees at School 43 Parental and Committee Involvement at School 43 Parent Satisfaction 44 Committee Satisfaction 45 47 05 Teacher Absence Definitions and Statistics 47 Definition and Measurement 47 Class Absence, or Classes Observed without Teachers 47 Teacher Absence from School 48 Teacher Absence from Teaching 50 Determinants of Teacher Absence 50 06 Student Absence and Learning Outcomes 55 Student Absence 55 Outcome Levels 55 Determinants of Student Learning 59 61 07 Conclusion Annex A. Primary School Availability, Student Test Scores, and Student Absence 65 Annex B. Student Competency Classifications 83 Reference 85 iv. TABLE of CONTENTS Table of Figures, Maps and Tables Figures Figure 1 School Meetings with External Stakeholders, 2015/16 18 Figure 2 Topics Covered During Internal Meetings 19 Figure 3 Principal and Teacher Education Levels 21 Figure 4 Principal and Teacher Longevity at Current School 22 Figure 5 Median Monthly Total Income 27 Figure 6 Median Monthly Base Salary 27 Figure 7 Median Monthly Total Allowances 28 Figure 8 Parent Satisfaction with Education Quality and Child Learning Outcomes 46 Figure 9 School Committee Satisfaction with Education and School Quality 46 Map Map 1 Participating Districts and Number of Schools in Each District 5 Tables Table 1 Characteristics of the Study Districts Compared with Provincial and National Averages 6 Table 2 Student Learning Assessment: Sample Description 7 Table 3 Study Participants and Respondents 8 Table 4 Characteristics of Study Villages 10 Table 5 Distance and Travel Time From School to Key Institutions 11 Table 6 Distance and Travel Time from The Village Hall to Administrative and Financial Institutions 65 v. Table 7 Village Leadership Characteristics 12 Table 8 Availability of Education Facilities in the Villages 14 Table 9 Characteristics of Student Population in the Sample Schools 15 Table 10 Gender Distribution of Students, by Grade 65 Table 11 Availability of Key School Facilities 66 Table 12 Funding Sources for Sample Schools, Academic Year 2015/16 16 Table 13 Fees Charged to Parents, Academic Year 2015/16 17 Table 14 Instructional Language, Curriculum, and Teaching Load, Academic Year 2015/16 64 Table 15 Instructional Time, Academic Year 2015/16 65 Table 16 School Supervision and Meetings, Academic Year 2015/16 18 Table 17 Principal and Teacher Demographics 20 Table 18 Teacher Certification Status 21 Table 19 Principal and Teacher Work Experience 22 Table 20 Principals’ Characteristics 23 Table 21 Teachers’ Characteristics 24 Table 22 Principals’ Living Conditions: Median Distance, Travel Time, and Transportation Cost from Home o School 67 Table 23 Teachers’ Living Conditions: Median Distance, Travel Time, and Transportation Cost From Home to School 68 Table 24 Teachers’ Activities at School, Academic Year 2015/16 25 Table 25 Principals’ Additional Activities: Trainings, Other Jobs, and Involvement in Local Organizations 26 Table 26 Teachers’ Additional Activities: Trainings, Other Jobs, and Involvement in Local Organizations 27 Table 27 Evaluation of Teachers by Principal Academic Year 2015/16 28 Table 28 Principals’ Salary Delivery Mode 29 Table 29 Teachers’ Salary Delivery Mode 29 Table 30 Average Amount of Principals’ Allowances 68 Table 31 Average Amount of Teachers’ Allowances 69 Table 32 Principals’ and Teachers’ Allowance Delivery 70 Table 33 Principals’ Opinions of Teachers and Students 32 Table 34 Challenges Experienced by Teachers 33 vi. TABLE of CONTENTS Table 35 Teachers’ Reported Satisfaction with Education Stakeholders 72 Table 36 Teachers’ Reported Satisfaction with Their Salary (Percentage of Teachers) 73 Table 37 Parents’ Background Information (% Parents) 39 Table 38 Child’s Education and Parent Involvement 40 Table 39 Child Participation in Paid, Unpaid, and Household Works 41 Table 40 Parents’ Expectations of Child’s Education 41 Table 41 School Committee Characteristics 74 Table 42 School Committee Respondent Characteristics, Selection Process, and Funding Sources 43 Table 43 Parents’ Involvement in School, Academic Year 2015/16 45 Table 44 School Committee Activities, Academic Year 2015/16 75 Table 45 Teacher Presence and Students’ Activities in Classrooms 48 Table 46 Teacher Absence from School 49 Table 47 Teacher Absence from Teaching 51 Table 48 OLS Regressions of Teacher Absence on Selected Teacher and School Characteristics 52 Table 49 Student Absence and Reasons, by Grade 76 Table 50 Student Test Scores: Descriptive Statistics 56 Table 51 Student Test Scores in Indonesian and Math, by Parent Education 53 Table 52 Classification of Student Competency in Indonesian, by Grade 75 Table 53 Classification of Student Competency in Math, by Grade 78 Table 54 OLS Regressions of Student Test Scores on Selected Student and School Characteristics 60 Table 55 Primary School Availability at The Subdistrict Level 79 Table 56 Mean Student Test Scores, by Grade and Gender 80 Table 57 Student Absence, by Grade and Gender 81 Table 58 OLS Regressions of School, Principal, and Teacher Characteristics on Reported Teacher Shortage at School 82 vii. viii. Acknowledgments The Indonesia KIAT Guru: Improving Teacher Performance and Accountability analytical study was led by Dewi Susanti (Senior Social Development Specialist) with a core team that included Christopher Bjork (Professor of Educational Studies, Vassar College); Arya Gaduh (Associate Professor of Economics, University of Arkansas); Jan Priebe (Research Fellow, German Institute of Global and Area Studies); and Menno Pradhan (Professor at the Amsterdam Institute for International Development - Free University and University of Amsterdam). Adama Bah (Development Economist, KIT Royal Tropical Institute), Jan Priebe and Dewi Susanti wrote the report. Team research analysts were Rajius Idzalika (Monitoring and Evaluation Analyst, Former Consultant, World Bank); Kurniawati (Data management Analyst, TNP2K); Sharon Kanthy Lumbanraja (Research and Knowledge Management Analyst, TNP2K); and Indah Ayu Prameswari (Research and Knowledge Management Associate, TNP2K).  Data collection and cleaning was led by Dedy Junaedi (Survey Team Leader, Consultant, World Bank); Lulus Kusbudiharjo (Survey Assistant Team Leader, Consultant, World Bank); Anas Sutisna (Survey Assistant Team Leader, Consultant, World Bank); and Mulyana (Data Programmer, Consultant, World Bank), with a great contribution of data collection conducted by Anggitya Eki Adista, Agustian, Aulia Subur Prasetyo Aji, Sulthani Al Aziz, Riska Ayu Andriyani, Meritaningrum Anggraeni, Fitri Nur Annisa, Rizwar Ansyari, Kartika Yoga Asmara, Doni Aswandi, Yohannes Adio B, Budiyanto, Fiske Kristina Chandrawati, Sartika Dewi, Santi Dwiningsih, M Makhrus Effendi, Nurul Endrastuti, Mochamad Faizin, Niza Ferlina, Paulus Fernandez, Astarina Fiona, Muhammad Firdaus, Fitra, Ikhwanul, Hendri Gunawan, Yuyu Gustiana, Rois Habibi, Vembri Harjanti, Syarif Hidayat, M Ikhsanudin, Farikha Fathul Imami, Nurul Isnaini, Wulan Kusuma Jati, Hikmat Catur Jayusman, Panggung Dwi Kuncoro, Mugi Lestari, Ade Liska, Budi Marwanto, Amrinsiana Merry, Siti Munawaroh, Syirojan Muniron, Ria Arbiati Ningtyas, Irvan Noer, Nur Aji Nugroho, Nurbaiti, Sigit Sawung Pamuji, Andriyani Prabawati, Aninda Pratiwi, Jayus Priyana, Nunik Pudyastini, Lilik Hadi Purwanto, Deny Puspitasari, Dani Ramdani, Ramdhony, Rano, A Rifa’i, Deni Riyanto, Ade Rizky, Nur Rochim, Anton Rohmadi, Sabiruddin, Rahmat Saiful, Dedy Samsiar, Itmamul Wafa Samudra, Teguh Santoso, Idha Ayu Setyawati, Setyorini, Isti Sofia, Subadri, Sukiyanto, Ika Sundari, Fanser Syahtriawan, Nanang Tanjung, Taufan, Taufiqqurohman, M Yahya Yogo Utomo, Ariani Widiastuti, Rini Widiastuti, Wiwit Widiyani, Arief Setio Widodo, Ika Widyaningsih, Dewi Widyastuti, Tyani Aji Windu, Retno Suci Wulandari, Tri Widadi Wulandari, Rohmad Yasin Y, Rangga Sukma Yana, Hanifan Yudhistira, Yuliawati, Zezen Zaenudin and Muhammad Zulfan. Caroline Tupamahu (Project Team Leader, BaKTI); Setiawan Cahyo Nugroho (Technical Coordinator, KIAT Guru, TNP2K); and Tri Yuni Rinawati (Operations Coordinator, KIAT Guru, TNP2K) coordinated inputs and supported the survey implementation on behalf of TNP2K and BaKTI. Members of the extended team included Gregorius Kelik Endarso (Operations Analyst, World Bank); Yulia Herawati (Social Development Specialist, World Bank); Lily Hoo (Former Senior Social Development Specialist, World Bank); Audrey Sacks (Senior Social Development Specialist, World Bank); and the Indonesia Education cluster team, under the leaderships of Tazeen Fasih (Lead Economist) and Noah Bunce Yarrow (Senior Social Education Specialist). Megha Kapoor (Knowledge Management officer, World Bank); Chatarina Ayu Widiarti (Program Analyst, World Bank); and Fazlania Zain (Communication and Operations Consultant, World Bank) provided overall support to the team. Audrey Sacks and Andrew B. Ragatz (Senior Education Specialist, World Bank) reviewed the final draft and provided constructive feedbacks for the final version. Publishing process by Dinda Putri Hapsari (Knowledge Management Consultant, World Bank), and Design and Layout by Yohanes Cahyanto Aji. ix. The World Bank is thankful to the Indonesian Ministry of Education and Culture (MoEC) and the National Team for Acceleration of Poverty Reduction under the office of the Vice President of Indonesia (TNP2K) for their advice and support which was critical to the success of this initiative. The Bank is especially thankful to Bapak Bambang Widianto (Executive Secretary, TNP2K); Bapak Sumarna Surapranata (Former Director General for Teachers and Education Personnel, MoEC); Bapak Supriano (Director General for Teachers and Education Personnel, MoEC); Bapak E. Nurzaman (Secretary of the Directorate General for Teachers and Education Personnel 2015-2018, MoEC); Ibu Dian Wahyuni (Head of Legal and Organization Bureau, MoEC); Bapak Praptono (Director for Development of Basic Education Teachers, MoEC); Bapak Temu Ismail (Section Head of Legal, Governance and Employment, MoEC); and Ibu Budi Kusumawati (Sub-Section Head of Legal, Governance and Employment, MoEC) for their insightful guidance and support. The Government of Australia’s Department of Foreign Affairs and Trade generously financed this study and provided guidance and continuous support to the Team. In addition, the Task Team is grateful for the guidance extended by the Country Management Unit including Rodrigo Chaves (Country Director); Rolande Simone Pryce (Operations Manager); and Camilla Holmemo (Program Leader). Kevin Tomlinson (Former Social Development Program Manager), and Nina Bhatt (Social Development Practice Manager) provided oversight of the Task Team. x. Abbreviations ACDP Analytical and Capacity Development Partnership ASER Annual Status of Education Report KIAT Guru Kinerja dan Akuntabilitas Guru (Teacher Performance and Accountability) MoEC Ministry of Education and Culture NTT Nusa Tenggara Timur (East Nusa Tenggara) OECD Organization for Economic Co-operation and Development OLS Ordinary Least Squares PISA Programme for International Student Assessment PLN Perusahaan Listrik Negara (Indonesian State-Owned Electricity Company) PNS Pegawai Negeri Sipil (Civil Servants) IDR Indonesian Rupiah SLA Student Learning Assessment TAS Teacher Absence Survey TIMSS Trends in International Mathematics and Science Study UNICEF United Nations Children’s Emergency Fund xi. xii. Executive Summary Competitiveness in an increasingly globalized world requires a highly skilled and educated workforce. The Government of Indonesia recognizes National and district that a highly educated and skilled workforce is critical to reducing inequality government budgets and poverty. To ensure schools are given adequate attention, the 2003 Law for education 20 on National Education System mandates that 20 percent of national and district government budgets is for education. This target was achieved in 2009 and has continued thereafter. Indonesia has made considerable progress in achieving universal enrollment at the primary and secondary school levels. The 20 achieved in 2009 % Government’s attention to education through its policies as well as the two decades favorable economic growth has enabled gross enrollment at the Gross enrollment primary school levels at about 100 percent, with gross enrollment at the at the secondary school secondary school levels increasing from 55 to more than 86 percent.1 increasing Paradoxically, despite success in education enrollment, Indonesian students have low learning outcomes, particularly in rural and remote areas of the country. Findings show that years of education and enrollment figures do not correlate with the quality of education provided. In other words, “schooling ain’t learning” (Pritchett 2013; World Bank 2018a). In all international 55 to % assessments (such as the PISA, TIMSS, and PIRLS), Indonesian students rank mo r e t h a n bottom among all countries assessed (Hanushek and Woessmann 2007; OECD 2017; World Bank 2017). Over the past 20 years, Indonesian student learning outcomes have tended to remain flat (OECD 2017; Beatty et al. 2018). In addition, studies show that primary and secondary schools located in rural and remote areas have substantially lower learning outcomes compared with 86 % their urban counterparts (Stern and Nordstrum 2014; BPS 2017; Beatty et al. 2018). Gaps remain between rural and urban education outcomes which Teacher absenteeism have both supply and demand side dimensions. From a supply-side s ur ve y i n Indo n e s i a fou n d a n a t i o n a l a ve r a g e of perspective, teacher absenteeism is a key challenge in remote parts of the country. A first teacher absenteeism survey in Indonesia found a national average of 15 percent teachers being absent from school, with those working in remote areas having a 24 percent absenteeism rate (Chaudhury, et al. 2006). A more recent survey found that while teacher absenteeism in general has reduced over time, rural teacher absence rates remain high at 20 15 (2006) % percent, compared to urban teacher absenteeism at six percent (ACDP 2016). Likewise, teacher absenteeism rate in more remote locations like Papua and T e a che r a bs e nt e e ism West Papua was even higher, with a rate of 37 percent (UNICEF 2012). In in remote areas addition, two-thirds of schools in remote areas do not have teachers yet two- 24 % thirds of urban schools have too many (World Bank 2013b). Because rural poverty rates remain high, this fact constrains rural families from enrolling their children in school (BPS 2016; World Bank 2016). Parental views on education also shape enrollment. Early schooling does not yield immediate (2006) 1 Gross enrollment is a percentage of the population who were at school regardless of age, compared to the number of school-age population for the particular school level. World Bank World Development Indicator Database, https://data.worldbank.org/indicator/ SE.SEC.ENRR?locations=ID xiii. dividends. It often means there is one or several less persons to help earn K I A T G u ru rural incomes. Thus, with education not being valued in the same manner, b a s e l i n e s urvey rural enrollment fares worse. whic h is summ ari zed i n thi s Since the early 2000s, the Government of Indonesia has shifted its rep ort, was c on ducted i n education policies towards a focus on quality and equity. Quality 270 improvements have included providing schools with more education resources (through the School Operational Support); improving teacher qualification; enhancing community participation through school committees; and benchmarking student performance using international assessments. remote primary schools In addition, the Government of Indonesia now provides cash transfers to d uring 2016 - 2017 students from impoverished backgrounds to attend schools; and sets a policy of rotating teachers systematically to ensure there is equity of distribution. Landak Sintang Starting in 2016, the World Bank has supported the Government of Indonesia to improve teacher performance and community participation in education through KIAT Guru. The KIAT Guru (Improving Teacher Performance and Accountability) has been piloted in five districts across Indonesia, namely, Ketapang, Landak, and Sintang (in West Kalimantan province) Ketapang Manggarai Manggarai and West Manggarai and East Manggarai (in East Nusa Tenggara province). Barat Timur Prior to piloting the KIAT Guru, a baseline survey, which is summarized in this report, was conducted in 270 remote primary schools between 2016-2017 with multiple local stakeholders including school staff (principals and teachers); school committees; village heads; and parents. The survey instruments were designed to achieve better understanding of the challenges of primary education service delivery in remote areas. This report presents a detailed description of the six main findings of the survey. Connectivity challenge 1. Schools and villages face shortages in key areas which hampers good educational outcomes. There are connectivity challenges: some households and schools are on average 149 km 149 km or five hours away from district capitals; study areas have only 29 percent connection to the electricity; limited internet access (17 percent); and long distances to financial institutions, which affects or five hours away from salary retrieval for teachers (52 km or 2.3 hours on average). district capitals Infrastructure barriers can adversely affect placements of younger and highly motivated teachers in rural and remote areas. In addition, transaction costs associated with remoteness also reduce 29 have connection to % how government prioritizes funding for schools. The available resources within pilot schools may suggest adequacy: 91 percent have toilets which are reasonably gender balanced (50 percent are for females); 54 percent of schools have a library; and 39 percent the electricity have sufficient textbooks. Observations indicate that gaps in these areas can be met by attention to how funds are allocated rather 17 % than simply a focus on allocation amount. limited internet acces 2 The exchange rate was approximately IDR 14,000 to US$1 at the date of publication. xiv. 2. The specificities of school characteristics in terms of class sizes, teacher composition, and school management is not Teacher composition matched with the needs at hand. While class ratios compare 40 % well with national averages (20 students per class in remote areas compared with 23 at the national level); the uniqueness of remote area teachers is that they often substitute for absent teachers; and must teach multiple grades (in 25 percent of the schools) despite teachers who are part lacking the training to do so. Survey results show that permanent of the civil service teachers who are part of the civil service stands at 40 percent, with 42.5 remaining teaching gaps met by short term contract teachers who constitute the majority of the teaching force (42.5 percent contracted by schools and 15.8 percent contracted by districts or provinces). % Compared to permanent teachers, contract teachers have lower contracted by schools qualifications, much lower salary, and therefore more likely to have second jobs. Not having tertiary education degrees was fairly 15.8 % widespread: 34 percent of teachers and 18 percent of principals only have high school degrees. Likewise, although Indonesian should be the main language of instruction in primary schools, the study reveals that this is often not the case. For instance, in East Nusa Tenggara, contracted by districts nearly a third of schools utilize another language for teaching. While or provinces this may improve learning comprehensions, it imposes a challenge when students take national exams in Indonesian. 3. Teacher incomes fluctuate substantially within schools, which may affect motivation. Differential teacher income stems from Teacher absence whether they are civil servants and have certification status. While u n a n n o u n c e d v is it s to certified civil servants have median monthly income levels of IDR s a mp le s c h o o ls in d ic a te d tha t 8.4 million (US$600), noncertified civil servants earn roughly IDR 4.6 25 % million (US$329) monthly. The lowest income earners are non-civil servants, with a median monthly income of IDR 0.55 million (US$40).2 4. Teacher absence from school and teaching is a serious issue. Unannounced visits to sample schools indicated that 25 percent of classrooms did not classrooms did not have a teacher, and 17 percent of teachers did have a teacher not come to school on a given day. Our analysis indicate that teacher absence is associated with being a male, with civil servant status, 17 % and experiencing less supervision by the school principal. In other words, female and contracted teachers were teaching more often. Our analysis also shows, however, that evaluated teachers tend to have better presence in school. Hence, ensuring that teachers are teachers did not monitored and supervised could reduce teacher tardiness. come to school on 5. Student learning outcomes were low. Most students tested were performing two grade levels below their current grade and had not a given day mastered basic standards of their former grade level. For example, a grade four student demonstrated the competency of a grade two student. Our analysis associates low student learning outcomes with low parental education; less time dedicated to their child’s schooling; and far less engagement with school committees and teachers. 2 The exchange rate was approximately IDR 14,000 to US$1 at the date of publication. xv. Parents’ satisfaction 6. SParents’ satisfaction with the quality of education and learning outcomes is in contrast with the study findings. During interviews, parents report that they actively support their children’s studies at home, including knowing the subjects that their children do not master. As a sign of active parent involvement in their children’s education, more than four-fifths of parents in the sample went to their child’s school, and more than four-fifths of the school committees held separate meetings with principals and parents during academic year 2015/16. In general, the parents and school committees reported that they were satisfied the parents and with the quality of education and learning outcomes. This latter finding school committees reported that they is somewhat surprising, considering high teacher absence and weak were satisfied student learning outcomes in the study areas. The result indicates with the quality of education and that parents have either very moderate expectations of the quality of learning outcomes education that is delivered at school, or are not fully informed with the service standard that should be delivered by teachers. Based on survey findings, we identify six policy recommendations for improving learning outcomes in remotely located schools. Although Indonesia’s favorable economic growth can contribute to better education, the past has shown that economic growth itself is no guarantee of improvements in student learning outcomes. To provide children in remote areas with better education quality, it appears crucial for the Government of Indonesia to have a comprehensive set of policies. Improvements in infrastructure—better roads, telecommunication, and electricity—for remote areas will contribute to making them more accessible. Relatedly, renovations of school facilities, including housing provision for teachers, need to be prioritized to improve working conditions for teachers appointed to these areas. Teacher absenteeism is a serious problem that needs to be addressed, as it directly affects whether students learn in school. Improvements in infrastructure will likely reduce teacher absenteeism, as supervision of schools becomes easier, while commuting time to financial, health, and other institutions becomes faster. The Government of Indonesia may consider making electronic payments for teacher salaries and allowances, to reduce the need for these teachers to travel. In addition, various ways to improve teacher accountability need to be tested for effectiveness, along with means to sanction underperforming teachers. Improving student learning outcomes needs to start by making the results digestible to education stakeholders, tracking development over time, and ideally benchmarking of school-level outcomes against district- or national-level results. Collaborations among teachers and parents to support student learning will likely generate higher aspirations and career outlooks among students. Raising parents’ knowledge and awareness of high teacher absenteeism and low learning outcomes will likely increase their demands for improved education quality. This may entail setting up service standards expected from teachers and known to education stakeholders. In addition, increasing awareness of parents’ role and participation in supporting their children’s learning is crucial for their increased engagement. xvi. Mechanisms to make teacher allowances more effective in improving teacher performance and student learning outcomes need to be identified and tested for implementation by the government. For example, the amounts of allowances paid need to be made conditional based on teacher presence, teacher performance, and/or a fraction of student learning outcomes. Qualifications and skills of teachers in remote areas need to be upgraded. In the short run, more capacity development trainings should be delivered, by prioritizing teachers in remote areas or requiring a set percentage of training participants for teachers in remote areas. In the long run, more, better qualified, and younger teachers should be distributed to remote areas with an agreeable time limit, which will increase their credit points toward more quickly becoming certified or qualified to become civil servants. Improvements in infrastructure will also enable teachers to consider admission to higher education, join capacity development trainings, or take distance learning courses. Through KIAT Guru, the Government of Indonesia, with technical assistance from the World Bank, has started addressing some of these education service delivery challenges in remote areas. Recognizing that teacher absenteeism is a serious impediment to education service delivery, KIAT Guru raises stakeholders’ awareness on this issue, along with the fact that student learning outcomes in the 270 schools was, on average, two grades below the national curriculum standard. KIAT Guru tests two mechanisms to improve teacher presence, teacher service performance, and student learning outcomes. A Social Accountability Mechanism (SAM) provides community members with an explicit role to monitor and evaluate teacher service performance and to ensure teacher accountability. There is also a Pay for Performance Mechanism (PPM), which links the payment of teacher remote area allowance (Tunjangan Khusus Guru, or TKG) with either teacher presence or teacher service quality. The two mechanisms are combined into three intervention groups i.e. (1) SAM; (2) SAM + PPM based on teacher presence; and (3) SAM + PPM based on a broad measure of the quality of teacher service performance. The 270 schools included in the survey were randomly assigned into the three intervention groups and compared to a control group. The World Bank conducted an evaluation of KIAT Guru and found positive impacts. A second round of survey in the 270 schools was conducted at the beginning of 2018, and the results were compared and analyzed with the first round of survey covered in this report. The impact evaluation found that all three KIAT Guru intervention groups performed statistically and significantly better than the control group where no KIAT Guru intervention was implemented. The SAM combined with the PPM based on teacher presence (“Group 2”) had the strongest positive effects on student learning outcomes in mathematics and Indonesian language (at 0.19 and 0.17 standard deviations respectively). It increased the presence of TKG-recipient teachers in classrooms and improved parental involvement in meeting with teachers and in supervising learning at home (Gaduh, et al, 2019). The IE is accompanied by a qualitative research conducted in nine case study schools, which findings reinforced recommendation for Group 2 as the most effective intervention (Bjork & Susanti, 2019). The impact evaluation, qualitative research, and process monitoring attributed the success of the interventions to four key elements: (a) increasing parental awareness of learning outcomes and their involvement in improving learning; (b) keeping teachers accountable through a few simple and objective performance evaluation indicators; (c) actively engaging external stakeholders in supporting, monitoring, and evaluating education service delivery; and (d) paying teacher allowance based on objective performance indicator. xvii. xviii. 01 INTRODUCTION Introduction The Government of Indonesia recognizes the importance of education in improving productivities and reducing poverty and inequality. With the government’s budget for education tripling in real terms since 2001, Indonesia has made considerable progress in achieving universal enrollment at the primary and junior secondary school levels. Law 20 on National Education System, which was passed in 2003, mandated 20 percent of national and district government budgets for education, and this has been met since 2009. In 2017, the gross enrollment rate of Indonesian children ages 13 to 15 exceeded 95 percent.3 Education policies in Indonesia have shifted focus toward quality and equity. Quality improvements have included providing schools with more control of education resources (through the School Operational Support)4, improving capacities and qualifications of teachers, enhancing community participation through school committees, and benchmarking student performance through participation in international assessments. The Government of Indonesia has also provided students from poor backgrounds with cash transfers and attempted to improve equity by redistributing teachers. A Major education policy reform has focused on improving teacher welfare, although teacher performance has remained stagnant. In 2005, Law 14 on Teachers and Lecturers increased the minimum teacher academic qualification to a bachelor’s degree, requiring all teachers to complete their certification process successfully by 2015.5 For teachers who complete the certification process6, the law introduced a certification allowance, in the amount up to 100 percent of their base salary.7 Those working in remote areas receive a remote area allowance, also up to 100 percent of their base salary. Half of the national education budget has been allocated for payment of close to three million teachers’ salaries and allowances, which in 2018 amounted to US$16.1 billion. However, recipients of the remote area allowance had a higher teacher absenteeism rate 3 Badan Pusat Statistik 2018. 4 Bantuan Operasional Sekolah (BOS) is a school-managed operational fund allocated by the Ministry of Education and Culture based on the number of registered students in the school. In 2018, BOS amounted to Rp 800,000 (US$57) per student per year. 5 http://peraturan.go.id/uu/nomor-14-tahun-2005.html. 6 The teacher certification process requires that teachers hold an undergraduate degree, submit a portfolio of their teaching experiences, and pass a competence test. Teachers are currently certified for life, with no recertification process in place. 7 Tunjangan Khusus Guru (teacher special allowance) is allocated for teachers assigned to special areas, including remote areas. For ease of reference, we use the term remote area allowance in this report. 1. compared with nonrecipients in the same schools. (BPS 2019). As a result, over one-third of the increase Likewise, the performance of students of certified in inequality from 2002 to 2012 can be explained by teachers did not differ from students of those who where one is born and who one’s parents are (World were not certified.8 Bank 2016). Disparity among rural and urban locations persists in education service delivery and outcomes. Indonesian students achieve relatively low Two-thirds of schools in remote areas are lacking levels of learning. In all international assessments teachers, while two-thirds of urban schools have in which the country has participated, the learning too many teachers (World Bank 2013b). Around 50 outcomes of Indonesian students rank at the bottom percent of the population age 15 and above in rural among participating countries.9,10 Furthermore, only areas has not completed or just completed elementary little progress has been achieved in student learning education, compared with 35 percent in urban areas outcomes over the past 20 years (OECD 2016; Beatty (BPS 2018). et al. 2018). The latest reading assessment of the Programme for International Student Assessment Teacher absence is a significant constraint in (PISA) shows that fewer than one in two Indonesian remote areas. In 2003, the first assessment of students demonstrates the basic reading skills needed teacher absence conducted in Indonesia found a to participate effectively and productively in life.11 In the nationwide absence rate of approximately 19 percent 2015 Trends in International Mathematics and Science among teachers in public primary schools.16 Although Study (TIMSS), Indonesian fourth grade students’ math this rate was reduced to 10 percent in 2014, in remote scores ranked 53 among 57 participating countries.12 areas, still one in five teachers was absent from Using the assumptions of improvement rates on PISA school.17 Furthermore, teacher absence in Indonesia tests from 2003 and 2015, the World Development is associated with higher student absences18; higher Report 2018 calculated that it would take Indonesia 48 dropout rates, particularly in remote areas19; as well as years to achieve the current Organisation for Economic lower student test scores.20 Co-operation and Development average score in The Government of Indonesia has issued policies mathematics and 73 years in reading, if education and resources to specifically focus on improving practices do not change (Beatty 2018; World Bank education service delivery in rural and remote 2018a). Clearly, “schooling ain’t learning”13 and years areas. The 2005 Law 14, the Minister of Education of education and enrollment figures do not necessarily Regulation 32 from 2007, and Government Regulation tell much about the quality of education provided. 74 from 2008 defined special areas (daerah khusus), Given the importance of education quality for individual which include remote, frontier, and disaster- and earnings and well-being and bearing in mind the role of conflict- prone areas. Teachers placed in these areas education quality for economic growth, 14 governments are entitled to several additional compensations around the world need to focus on achieving better and benefits, ranging from housing entitlement, student learning. Furthermore, it is noteworthy that automatic promotion, special promotion, job learning results vary substantially across the country, security and protection, scholarships, and priority with rural and remote areas lagging significantly behind for improving academic qualifications, certifications urban areas.15 and competencies. Most importantly, teachers have Education equity in poor rural and remote areas become eligible for allowances that double or triple remains a challenge (World Bank 2013b). Rural their base salary, if they meet certain requirements. For areas have consistently higher rates of poverty (14.1 example, a teacher who becomes certified is eligible to percent) compared with urban areas (8.2 percent), receive a Tunjangan Profesi (certification allowance), poorer connectivity, and lower quality of basic services which doubles his/ her base salary. Similarly, those 8 Toyamah et al. 2010; De Ree et al. 2018. 16 Usman, Akhmadi, and Suryadarma 2004; World Bank 2004; 9 The participating countries comprise the 34 Organisation for Chaudhury et al. 2006. Economic Co-operation and Development (OECD) member 17 ACDP 2014. countries, as well as several partner countries in Latin America, 18 Toyamah et al. 2010. Eastern Europe, Asia, and the Middle East and North Africa. 19 UNICEF 2012. 10 OECD 2016; World Bank 2018a. 20 Usman, Akhmadi, and Suryadarma 2004; Suryadarma et al. 2006. 11 OECD 2016. 12 Mullis et al. 2016. 13 Pritchett 2013; World Bank 2018a. 14 Hanushek and Woessmann 2007. 15 ACDP 2014; Stern and Nordstrum 2014. 2. INTRODUCTION who works in special areas are eligible to receive a As a follow up policy intervention, the Government Tunjangan Khusus, which range from IDR 1.5 million to of Indonesia with technical assistance from the double the teacher’s base salary. So if a certified civil World Bank has been implementing KIAT Guru servant teacher works in remote area, he or she could since 2016. KIAT Guru aims to improve teacher be entitled to a total income of up to three times his/ presence, teacher service performance, and student her base salary. While significantly increasing teachers’ learning outcomes in remote primary schools. Prior income might seem like a significant incentive for to implementation of several interventions, the World teachers to improve their performance, a World Bank Bank conducted a survey in the study areas, to achieve study on the impact of certified teachers on student better understanding of the challenges of primary learning outcomes does not find that this is the case education service delivery in remote areas. The study (De Ree et al., 2018). Similar outcome was identified areas consist of five districts across Indonesia, namely, for Tunjangan Khusus, where recipients had highest Ketapang, Landak, and Sintang (in West Kalimantan rate of absenteeism in comparison to non-recipients province) and West Manggarai and East Manggarai (Toyamah et al., 2010). (in East Nusa Tenggara (NTT) province). This report presents a detailed description of the survey findings, conducted in a total of 270 primary schools located in 235 very remote villages. Although the sampling, instruments, and research questions were generated to match the specific demands for KIAT Guru, the findings are general enough to inform the conditions of education in the study areas. 3. 4. 02 SCOPE of STUDY Scope of the Study: Locations, Instruments, and Sample The report is organized as follows. Section 2 describes the baseline survey administered in the study villages. Section 3 presents the schooling context. Section 4 discusses parental and community involvement in education. Section 5 analyzes teacher absence in the sample primary schools. Section 6 analyzes student learning outcomes. Section 7 concludes. Selection of Study Areas The study districts represent five of 122 disadvantaged districts in Indonesia. The selection of districts was based on the list of disadvantaged districts established by the Ministry of Villages, Disadvantaged Areas, and Transmigration in 2015. The list was narrowed through parameters set by the KIAT Guru project. Districts with very remote locations, conflict-prone, very low demand for education, very weak governance, and very high operational costs were excluded. Shortlisted districts had at least 40 primary schools in remote areas that fulfill the definition of eligible schools described below. Upon consultations with the Government of Indonesia at the national level, the list was further narrowed and visited to identify those having anecdotal problems of teacher absenteeism, whose district governments showed willingness to reform. The final list (map 1) includes three districts in West Kalimantan (Ketapang, Sintang, and Landak) and two districts in NTT (West Manggarai and East Manggarai). Schools eligible for selection in the study had a minimum Map 1. Participating Districts and Number of Schools in Each District Landak Sintang 51 88 Ketapang 59 Manggarai Manggarai Barat Timur 38 34 5. Table 1. Characteristics of the Study Districts Compared with Provincial and National Averages West Kalimantan NTT National Province West East Province Average Ketapang Landak Sintang Average Manggarai Manggarai Average Poverty rate (%) 10.99 12.32 10.07 7.87 19.35 27.71 22.19 10.86 Very remote villages 136 79 287 72 55 59 21 26 based on IDM (#) Very remote villages 51.91 51.30 82.23 48.31 32.54 33.52 13.96 16.28 based on IDM (%) Subdistricts 21 13 14 12 10 9 14 14 Villages 262 154 349 148 169 176 151 161 Average # villages per 12 12 25 12 17 20 11 12 subdistrict Total population (# 482,831 361,469 400,789 338,349 256,105 275,622 227,083 504,680 individuals) Average village 1,843 2,347 1,148 2,284 1,515 1,566 1,503 3,139 population (# individuals) Primary schools 526 457 429 305 259 329 368 289 Average # primary 2 3 1 2 2 2 2 2 school per village Sources: • Subdistricts and villages (2017): Central Bureua of Statistics https://www.bps.go.id/website/fileMenu/Perka-BPS-No55-Tahun-2017. pdf. • Individuals and households: SUSENAS March 2016 (own calculation). • School data: MoEC http://dapo.dikdasmen.kemdikbud.go.id/sp. • Poverty rate (2016): Central Bureau of Statistcs https://www.bps.go.id/website/pdf_publikasi/Data-dan-Informasi-Kemiskinan- Kabupaten-Kota-2016--.pdf. • Villages with IDM status (2015): MoV http://kedesa.id/id_ID/repository/indeks-desa-membangun-indonesia/. • National Socio-Economic Survey. Note: IDM = Individual Deprivation Measure; MoEC = Ministry of Education and Culture; NTT = East Nusa Tenggara; SUSENAS of 70 students, located at least a one-hour drive from The five districts have different provincial the district capital, and with at least three teachers who characteristics in terms of administrative units, received the remote area allowance. Table 1 presents but they all have more primary schools and more characteristics of the five study districts: poverty, remote schools compared with the national remoteness, administrative units, population, and average. Table 1 presents the average number of schools, with summary findings described below. villages classified as very remote according to the Village Development Index 2015.21 In the West Kalimantan Compared with other districts in Indonesia, the districts, at least 50 percent of the villages, and up to 82 five study districts have higher poverty rates. All percent of the villages in Sintang district, are classified five districts have higher poverty rates compared with as very remote. In West Manggarai and East Manggarai, the provincial and national rates, except Sintang, which approximately one-third of the villages are classified as has a lower poverty rate compared with the national very remote, which is significantly higher than in NTT (with average (11 percent in 2016). West Kalimantan districts only 14 percent of villages classified as very remote). are more prosperous than NTT districts and have a lower poverty rate than the national average. However, The average village population in the five districts is the three West Kalimantan districts are poorer than relatively small compared with the national level but other districts in that province. NTT has an average relatively large compared with the provincial level. poverty rate of 22 percent, compared with the national All five study districts have smaller populations than the poverty rate of 11 percent, with West Manggarai sitting average district nationally, but larger than the average district below the provincial average and East Manggarai above in their respective provinces. The three West Kalimantan it, with nearly 28 percent of its population living below districts, which have, on average, a population of about the official poverty line. 415,000 individuals, are much larger than the NTT districts, which have 265,000 individuals on average. Yet, the West Kalimantan survey districts are, per village, less populated 21 The Village Development Index was developed by the Ministry of Villages, Disadvantaged Areas and Transmigration (2015). than villages in the average district in the province. 6. SCOPE OF STUDY The Student Learning Assessment (SLA) captures Survey Instruments student competencies in literacy and numeracy, The survey comprises instruments to measure according to the standards set in the 2006 national teacher absence and student learning outcomes, curriculum. The SLA instrument is a grade-specific along with five questionnaires. The teacher test, which was developed based on frameworks and absence survey is based on Analytical and Capacity findings from international and national assessment Development Partnership’s instrument, which was tools. The assessment aims to capture basic to used for its 2014 survey, based on an instrument higher-order skills in reading and writing (in this case, developed by the World Bank for World Development Indonesian) and performing mathematical operations.24 Report 2004.22,23 Teacher presence is directly observed Tests were developed to contain a larger distribution of by enumerators during an unannounced school visit. lower grade–level questions, to capture a more normal It collects information on teacher location (in or out of distribution of student learning outcomes. The SLA was school, and in or out of class), teacher activity (teaching- administered to all the students in grades one to five in related or non-teaching-related) during school hours, the sample schools.25 of the 28,790 students registered and student absences from school. It generates data in grades one to five, 26,612 students—present on the to assess the rate of teacher absence from school, day enumerators administered the tests—undertook classroom, and teaching. the SLA Indonesian and math tests (table 2). Students Table 2. Student Learning Assessment: Sample Description West Kalimantan NTT Grade Gender All Areas Ketapang Landak Sintang West Manggarai East Manggarai Number of students registered in school 1 Male 3,192 633 511 1,080 485 483 Female 2,685 517 462 888 404 414 2 Male 2,924 564 484 925 452 499 Female 2,458 488 408 776 372 414 3 Male 3,068 615 465 1,024 492 472 Female 2,616 481 397 864 475 399 4 Male 3,080 552 557 938 541 492 Female 2,815 522 449 957 492 395 5 Male 3,126 552 520 1,015 533 506 Female 2,826 525 429 996 450 426 Number of students present in school on the day of the test 1 Male 2,802 537 436 949 431 449 Female 2,375 441 391 799 359 385 2 Male 2,641 491 434 838 402 476 Female 2,280 448 370 720 344 398 3 Male 2,838 551 410 973 459 445 Female 2,489 448 367 831 456 387 4 Male 2,884 492 513 897 516 466 Female 2,662 476 420 905 477 384 5 Male 2,918 495 479 964 507 473 Female 2,723 500 408 964 433 418 22 Chaudhury et al. 2006. for grades three through five. without a break, unless requested 23 Usman, Akhmadi, and Suryadarma 2004; Toyamah et al. 2010; by students, the math test then followed, with a time limit of 15 UNICEF 2012; ACDP 2014. minutes for grades one and two and 50 minutes for grades three through five. The differing time limits between the lower and upper 24 ASER 2014; Gove and Wetterberg 2011; Mullis et al. 2016; Platas grades were determined based on test item types. The items for et al. 2014; Uwezo 2012. the lower grades were largely on recognition (for example, letter 25 To avoid teachers preparing students for the test, it was and number recognition), which required less time to work on announced to teachers and students one day before administration. than the more complex items for the upper grades (for example, All participating students started with the Indonesian test with a reading comprehension and performing math operations). time limit of 25 minutes for grades one and two, and 45 minutes 7. in grades one and two took the test individually, that year. Students had not yet been taught—even less is, one person at a time, and were allowed 25 minutes so mastered—a significant part of the material for for the Indonesian test and 15 minutes for the math their current grade level. Therefore, 80 percent of the test. Students in grades three to five were allowed 45 questions for the Indonesian and math tests were minutes for the Indonesian test and 50 minutes for the based on the curriculum standards for one and two math test and took the tests in groups.26 grade levels below the respective grades at which the tests were administered. The remaining items (20 The SLA was a multiple-choice test, with three to four percent of the questions) were based on the curriculum possible answers for each question.27 There were 23 standards of the current grade level. and 30 questions in the Indonesian and math tests, respectively, for each grade-level test. Except for the The five questionnaires were adapted from tests for grade one, all the grade-level tests were previous surveys conducted in Indonesia.28 The developed with a large distribution of lower grade– questionnaires collect detailed information from village level questions, given that they were administered heads, school principals, teachers, school committees, during the middle of the first semester of the academic and parents of children attending primary school.29 Table 3. Study Participants and Respondents West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Subdistricts 33 8 7 7 6 5 Villages 235 59 38 82 27 29 Households 5,400 1,179 1,020 1,761 760 680 Registered students in grades 1-6 35,543 7,350 5,682 11,449 5,709 5,353 Registered students in grades 1-5 28,791 5,449 4,682 9,463 4,696 4,501 Students tested for SLA 26,613 4,879 4,228 8,840 4,384 4,282 Schools 270 59 51 88 38 34 Principals 270 59 51 88 38 34 PNS principals 268 58 51 87 38 34 Teachers listed in schools 2,293 508 370 700 385 330 Teachers surveyed 1,917 420 300 585 332 280 PNS teachers 755 140 133 240 137 105 Non-PNS teachers 1,162 280 167 345 195 175 Share non-PNS teachers (%) 39 33.33 44.33 41.03 41.27 37.50 Contract teachers 348 176 27 60 18 67 Honor teachers 814 104 140 285 177 108 Certified teachers 265 52 50 91 48 24 Noncertified teachers 1,652 368 250 494 284 256 School committees 268 58 50 88 38 34 Active 254 56 48 82 35 33 Non-active 14 2 2 6 3 1 Note: Number of subdistricts, villages, and schools in the sample. NTT = East Nusa Tengarra; PNS = civil servants; SLA = Student Learning Assessment. 26 The rationale behind the time differences in testing dependent cheating in group-setting tests, two versions of the test booklets on age is that students in lower grades are in the early phase of were developed, with variations in the sequencing of the same learning and are not fully able to read and write on their own; types of questions. Students sitting next to each other were thus thus, they require more intensive one-on-one assistance with test given different versions of the test booklet. instructions. For the students in the lower grades, administrators 27 Indonesian tests for grades one and two (only) provide four assisted by reading the instructions to them for each item possible answer options for 48 and 39 percent of the questions, presented and writing their answers on the answer sheet. The respectively. administrators received training on how to conduct this test, for example, without providing students with clues to the answers. 28 World Bank 2013a, 2015; ACDP 2014; Pradhan et al. 2014. Previously, this practice was undertaken by Stern and Nordstrum 29 The data in this report draw on the instruments that are (2014) and ASER (2014). Students in grades three and above are components of the quantitative surveys. In addition, qualitative normally used to reading and writing on their own; therefore, the data were collected in the pilot areas; however, these are not test was delivered in the usual way, requiring the students to read discussed in the report. the instructions and write the answers on their own. To avoid 8. SCOPE OF STUDY The village head questionnaire collected information populations of school-age children between districts— on village population characteristics, access to basic from 22 percent of the population in Ketapang to 46 energy and services, social cohesion, and degree of percent in East Manggarai. Children of primary school remoteness. The principal questionnaire collected age comprised approximately 40 percent of the comprehensive information on school operations, children ages 4 to 20 years. including the availability of key physical infrastructure, From an ethno-religious perspective, study student population characteristics, instruction villages are highly homogeneous and have a low processes, and results of student graduation exams. level of conflicts. The majority of the population Mirroring the principal questionnaire, the teacher belongs to the largest local religion and ethnic group. questionnaire sought information on teachers’ On average, in 76 percent of the villages, more than 80 backgrounds, education levels, experiences, living percent of the population belongs to the largest ethnic conditions, and activities in and outside school; teachers’ group, compared with an average of 51 percent of the allocation of time among different teaching-related villages with more than 80 percent of the population tasks; the salary and allowances they received; and belonging to the largest religion. Catholicism is the their motivations and sense of satisfaction. The school main religion. However, there is a notable difference committee questionnaire focused on the committee’s between the NTT districts, where Catholicism is the history, financial information, school management religion of nearly 90 percent of the population, and activities, and satisfaction with school quality. The West Kalimantan, where other religions are more evenly parent questionnaire collected information on parents’ spread among the residents. Islam and Protestantism socioeconomic background, degree of engagement with are the respective religions of 12-34 and 15-39 percent, the school, supervision of home study, participation in respectively, of residents across all five districts. On their child’s schoolwork, and student absence. average, approximately 11 percent of the villages (or 26 villages) reported having experienced local conflict Study Participants and Respondents in the past year, ranging from zero villages in West The study covers 270 primary schools located Manggarai to 16 percent of the villages in Landak. The in 235 remote villages across 33 subdistricts. main reasons given for these local conflicts were public Among these schools, 198 were in West Kalimantan policies, public service delivery, and economic matters. and 72 in NTT. Table 3 presents the sample size and Access to electricity, telecommunication, and population numbers in the study areas. Principals and internet varies widely across villages. The majority school committee representatives in 270 schools were of the villages (90 percent) have access to sources interviewed, along with 235 village heads. The sample of electrical power, but only very few (29 percent) schools comprised of 35,543 students. of 28,791 obtain power from state-owned electrical grid (PLN). students in grades one to five, 92 percent participated Access to PLN electricity ranged from 15 percent in the SLA.30 The survey also collected information on in West Manggarai to 44 percent in Ketapang. The 5,400 randomly selected parents of students who took types of cooking fuel used in the study villages were the SLA—four parents for each grade (for grades one to more province-specific, with all the villages in the NTT five only). If a school did not have a given grade level, four districts using firewood as cooking fuel and none using parents from other grade levels were selected instead. gas or liquefied petroleum gas. In West Kalimantan, 29 percent of the villages in Landak, 46 percent in Description of Study Villages Ketapang, and 54 percent in Sintang used gas to cook; Study villages had an average population of the remaining used firewood. The vast majority of the 1,400 individuals, only half of the national villages (all in the NTT districts) had access to mobile average (3,100 individuals per village). Among phone networks, although only 71 percent had access the five districts, Sintang had the lightest (1,000) and in Landak. Internet access was less widespread, with Landak has the densest (2,000) population (table 7 to 16 percent of the villages reporting access. A 4). On average, school-age children (ages 4 to 20) remarkable exception was West Manggarai, where 56 comprised 35 percent of the population in the five percent of the study villages had access to internet. districts. However, there were large differences in the The degree of remoteness of study villages varies in access to health centers and distance and travel time to key administrative and financial 30 The remaining 10 percent of the students were absent on the day the test was implemented. institutions. In all the districts, community health 9. Table 4. Characteristics of Study Villages West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Demographic characteristics Total population 1,396 1,576 2,097 999 1,328 1,328 School age (4-20) individuals 495 352 737 369 493 607 School age (4-20) individuals (% total population) 35.5 22.4 35.2 36.9 37.1 45.7 Pre-primary school age (4-6) 89 62 132 68 95 103 Primary school age (7-12) 193 178 264 153 196 202 Primary school age (7-12) (% total population) 13.9 11.3 12.6 15.3 14.8 15.2 Primary school age (7-12) (% school-age population) 39.1 50.6 35.8 41.6 39.8 33.3 Junior secondary school age (13-15) 108 61 172 78 100 149 Senior secondary school age (16-20) 104 51 169 69 102 153 Households/families 717 828 1,061 517 662 677 Use of basic energy and communication infrastructure (% villages with access) PLN electricity 29 44 32 26 15 21 Non-PLN electricity 90 92 89 88 100 86 Cooking fuel: gas or LPG 35 46 29 54 0 0 Cooking fuel: firewood 64 54 63 46 100 100 Mobile phone 90 90 71 93 100 100 Internet 17 12 16 13 56 7 Distribution of religions among residents (% village population) Islam 16.2 34.1 12.9 11.6 10.7 1.8 Christian - Protestant 23.6 15.2 38.8 36.7 1.1 3.5 Catholic 59.4 48.9 48.3 50.5 88.2 94.8 Buddha 0.34 1 0.03 0.88 0 0 Hindu 0.04 0.15 0 0 0 0 Confucius 0.01 0.02 0 0.010 0 0 Others 0.38 1.460 0 0 0 0 Community homogeneity and conflict Number of religions present in the village 2.46 2.78 2.820 2.77 1.31 1.48 % villages with more than 80% of the population having 51 53 11 39 88 100 the largest religion % villages with more than 50% of the population having 88 81 79 88 100 100 the largest religion % villages with more than 80% of the population of the 76 63 82 88 70 69 largest ethnicity % villages with more than 50% of the population of the 93 93 100 98 89 76 largest ethnicity Occurrence of local conflicts in village during past year 11 14 16 13 0 3 Conflict over public policies/services (% conflicts in past 35 38 17 36 0 100 year) Conflict over economic matters (% conflicts in past year) 23 13 33 27 0 0 Conflict over personal matters (% conflicts in past year) 15 25 33 0 0 0 Note: LPG = liquefied petroleum gas; NTT = East Nusa Tenggara; PLN = state-owned electrical grid. centers and health care staff are located less than one capital. The institution that is closest to the village hall hour from schools (table 5). Hospitals are much farther, is the subdistrict office, which is located on average 28 on average, approximately 100 kilometers and nearly kilometers and approximately 1.3 hours travel time four hours travel away. On average villages were 149 from the village. Neighboring district government kilometers and almost five hours away from the district offices may often be located closer—nearly one 10. SCOPE OF STUDY Table 5. Distance and Travel Time from School to Key Institutions West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Regional education office Distance (km) 124.8 205.3 87.0 132.1 77.1 78.1 Travel time (hours) 4.8 6.7 3.2 5.0 4.4 3.4 UPTD in subdistrict Distance (km) 41.6 37.0 48.7 56.0 20.0 27.4 Travel time (hours) 2.1 1.5 2.3 2.5 1.7 2.4 Nearest hospital Distance (km) 102.6 170.5 65.5 123.4 36.0 61.8 Travel time (hours) 3.9 5.2 2.7 4.7 2.6 2.9 Nearest community health Distance (km) 10.9 8.6 13.1 16.3 4.8 4.4 center Travel time (hours) 0.7 0.5 0.8 0.8 0.6 0.5 Nearest clinic/health care staff Distance (km) 5.5 4.8 7.3 2.3 4.2 13.9 Travel time (hours) 0.4 0.3 0.5 0.3 0.7 0.8 Nearest bank Distance (km) 52.0 49.9 44.5 82.6 24.4 21.2 Travel time (hours) 2.3 1.8 2.2 3.1 1.9 1.3 Nearest market Distance (km) 35.6 32.4 39.2 54.2 12.8 14.6 Travel time (hours) 1.7 1.3 2.0 2.2 1.3 1.1 Note: km = kilometers; NTT = East Nusa Tenggara; UPTD = Unit Pelaksana Teknis Daerah (district technical implementing unit at the subdistrict level). hour closer in West Manggarai and 1.2 hours closer in of the 235 villages surveyed, only 2 percent of Ketapang. Financial institutions are located closer to village heads were female, with the typical profile the study villages than the post office in all the districts being married male with a high school education. except Sintang. There is little difference in the distance The large majority of the villages (94 percent) were and travel time to banks and Automated Teller Machines headed by a village head or acting village head (table between provinces. In the NTT districts, cooperatives 7). Most of the villages with no head (10 of 13) were are closer than banks, and in West Kalimantan, credit located in Ketapang. The majority of the individual unions are located closer to villages. respondents to the village instrument (177 individuals, or 75 percent of the respondents) were village heads or A different type of remoteness is experienced in acting in the role. The typical village head was a married the West Kalimantan districts compared with the man in his mid-forties with a high school education (as NTT districts. Distance is slightly longer in the West the highest level of education attained) who resided in Kalimantan study districts than in the NTT districts, the village. There were only three female village heads varying from 14 kilometers in West Manggarai to 38 in total, one in Ketapang and two in Landak. In West kilometers in Sintang. The West Kalimantan districts Manggarai, 15 percent of the village heads resided in are located farther from key administrative and another village in the same subdistrict. Respondents financial institutions than the NTT study districts (table who were not village heads held the position of village 6, in annex A), but travel times for the West Kalimantan secretary (36 villages) or head of village affairs (22 districts are shorter or similar compared with the NTT villages). They were also typically married men with a study districts. For example, it takes about the same high school education and resided in the village, but time to travel to the subdistrict office in Sintang (38 they were slightly younger than the village heads, with kilometers) as in West Manggarai (14 kilometers), which an average age of 39 years. reflects differences in road infrastructure quality and topography.31 31 Several subdistrict offices in West Kalimantan are connected by the distance and travel time from subdistrict offices to villages, the Trans-Kalimantan Highway, whereas only a few subdistrict offices study villages are more difficult to reach in West Kalimantan in NTT are connected by large, paved roads. Similarly, subdistricts compared with NTT. This is largely due to the many rivers that must in West Kalimantan have better access to telecommunication be traversed by boat along roads between subdistrict offices and infrastructure than the NTT subdistricts. However, based on villages in West Kalimantan. anecdotes from the implementation team, and considering the 11. Table 7. Village Leadership Characteristics West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai # Villages 235 59 38 82 27 29 Type of respondent and characteristics Village head or acting village head (% 75 47 76 79 96 100 respondents) Village head or acting village head (#) 177 28 29 65 26 29 Other: secretary (% respondents) 15 24 18 17 4 0 Other: secretary (#) 36 14 7 14 1 0 Other: head affairs (% respondents) 9 29 5 4 0 0 Other: head affairs (#) 22 17 2 3 0 0 Duration in office (years) 3 3 3 3 3 5 Village with no village head (% respondents) 6 17 5 0 4 0 Village with no village head (#) 13 10 2 0 1 0 Village head - demographics and education Age (years) 44 45 42 41 47 47 % female 2 4 7 0 0 0 % married 97 100 97 97 88 100 % primary education 1 0 3 0 0 0 % junior secondary education 14 21 14 12 23 0 % senior secondary education 68 61 48 77 58 83 % university education 18 18 34 11 19 17 Other respondent - demographics and education Age (years) 39 37 49 38 41 - % female 5 10 0 0 0 - % married 98 100 100 94 100 - % primary education 3 3 11 0 0 - % junior secondary education 5 6 0 6 0 - % senior secondary education 83 87 78 76 100 - % university education 9 3 11 18 0 - Village head - residence location (% village heads) Village 92 86 86 98 85 97 Other village in subdistrict 6 11 7 2 15 3 Other subdistrict in district 2 4 7 0 0 0 Other respondent - residence location (% other respondents) Village 95 94 100 94 100 - Other village in subdistrict 3 3 0 6 0 - Other subdistrict in distict 2 3 0 0 0 - Note: NTT = East Nusa Tenggara. 12. SCHOOLING CONTEXT 03 Schooling Context School Characteristics School Availability Public primary schools are the most widely available type of educational institution in the five districts, with nearly a quarter of the schools conducting multi-grade classes. The implementation of the largest primary school construction program in the world happened in Indonesia between 1973 and 1978 (Duflo 2001). As shown in table 8, in West Kalimantan nearly all the study villages have at least one public primary school. In NTT, 85 percent of the villages have a public primary school. Other educational institutions are more diversely available across the survey districts. On average, 42 percent of the villages have at least one early childhood education facility, ranging from 26 percent in West Manggarai to 62 percent in East Manggarai. Kindergartens are scarcer in the study villages—there are none in the villages in East Manggarai; 25 percent of the villages in Ketapang have at least one kindergarten. On average, 46 percent of the villages have a junior secondary school, with variation across districts, from 33 percent in West Manggarai to 72 percent in East Manggarai. In general, senior high schools are only seldomly available, with 6 percent of the 235 villages having at least one such school. The study area includes 22 private primary schools, of which 20 are in NTT (table 55, in annex A). About 2 percent of the villages have an Islamic primary school. Nearly a quarter of the schools, ranging from 18 percent in the districts in NTT to 34 percent in Ketapang, employ multi-grade classes (table 9). These are defined as classes where a single teacher teaches students of two or more grades at the same time (Little 2006). Student Distribution The number of students per school is a bit lower than the national average. Table 9 presents the number of students in the study areas and their distribution in classes within schools. On average, there are six classes (rombongan belajar) per school–that is, one class per grade (kelas)–in the sample schools.32 These schools have a student-teacher ratio of 16 to 1, which is just slightly below national average of 17 to 1. The average number of students per school ranges from 111 in Landak to 157 in East Manggarai, well below 32 A grade (kelas) lasts an academic year and corresponds to a given level in the school system; there is a specific curriculum that students at this level are to be taught, and when they have not reached the required level, students may be required to repeat a grade. Class (kelompok belajar) corresponds roughly to the group of students who are physically located in a particular classroom and being taught simultaneously. 33 World Bank 2018b. 34 World Bank 2008. 13. Table 8. Availability of Education Facilities in the Villages West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Availability (% villages) 42 49 42 34 26 62 PAUD Number 1 1 2 2 1 1 Availability (% villages) 11 25 3 10 7 0 Kindergarden (TK) Number 1 1 5 1 2 Availability (% villages) 96 100 100 99 85 86 Public primary school Number 2 2 3 1 2 1 Availability (% villages) 16 17 3 1 44 48 Private primary school Number 1 2 1 1 1 1 Availability (% villages) 2 3 3 0 7 0 Islamic primary school Number 3 2 8 1 Availability (% villages) 46 47 53 37 33 72 Junior secondary school Number 1 1 1 1 1 1 Availability (% villages) 6 7 5 2 4 17 Senior secondary school Number 2 1 2 6 1 1 Note: Availability (% villages) refers to the share of study villages with at least one school type; number is conditional on availability in the village. NTT = East Nusa Tenggara; PAUD = Pendidikan Anak Usia Dini (early childhood education programs); TK = Tunjangan Khusus (special allowance). the national average of 191.33 However, 78 percent learning. Toyamah et al. (2010) and ACDP (2014) find that of Indonesian primary schools have fewer than 250 there is a direct correlation between the availability of students, and nearly 50 percent have fewer than 150.34 school facilities and teacher absence. Correspondingly, Overall, in the study areas, there were approximately 20 the sample schools lack universal access to key students per class (19-20 in West Kalimantan and 21-22 school facilities. Table 11 (in annex A) shows that the in NTT). sample schools are rather unequal in the availability of physical facilities. For teaching support facilities, Of the 35,543 students registered in the study there are significant differences in the availability of schools, there were larger gender disparities school libraries, ranging from 43 percent in Landak to in the earlier grades. In total, around 47 percent 91 percent in East Manggarai. Approximately 35 to 40 of students were female (table 10, in annex A). From percent of the schools across all the districts reported grades one to five, the largest disparities between the having a sufficient number of textbooks. numbers of male and female students were observed in East Manggarai (about two or three additional male Districts in West Kalimantan are better endowed students). Districts in West Kalimantan tended to have with buildings and other facilities, especially slightly larger classes in the lower grades than in the toilets, compared with districts in NTT. For upper grades, whereas in NTT districts the number of example, 96 percent of schools in Landak have a students was slightly higher in the upper grades, with teachers’ room, compared with 50 percent of schools in between one and five more students per grade. In the East Manggarai. Similarly, 26 percent of schools in West NTT districts, East Manggarai, in particular, had more Manggarai have a principal’s room, compared with 63 students on average than the other study districts, with percent of schools in Sintang. Overall, approximately 26 students per grade. 90 percent of the schools have toilet facilities, with lower percentages in NTT (79-85 percent). These toilet School Facilities facilities comprise the presence of toilets at schools regardless of whether they are reserved for teachers or Between 41 and 66 percent of the schools have students; it appears that teachers have toilets for their clean water; only 33 percent have access to own use more often than students do. Furthermore, electricity during school hours; and 45 percent female students in West Kalimantan are more likely have access to mobile phone signals. A common to have access to gender-specific toilets than male characteristic of remote schools in Indonesia is the students are. In West Kalimantan, on average, 70 lack of quality physical facilities to support teaching and percent of the sample schools have toilets for teachers, 14. SCHOOLING CONTEXT Table 9. Characteristics of Student Population in the Sample Schools West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai School size and class distribution by grade Average # students per school 132 125 111 130 150 157 Student-teacher ratio 16 14 16 16 15 16 Total # class groups 1,753 384 306 563 256 244 # Grade 1 classes 298 65 51 100 41 41 # Grade 2 classes 299 63 52 97 43 44 # Grade 3 classes 298 64 51 95 47 41 # Grade 4 classes 290 65 51 91 44 39 # Grade 5 classes 288 63 51 91 42 41 # Grade 6 classes 280 64 50 89 39 38 Average # classes per school 6.5 6.5 6.0 6.4 6.7 7.2 Average class size 20 19 19 20 22 21 Schools with multi-grade classes (%) 24 34 24 22 18 18 Student population # Students 35,543 7,350 5,682 11,449 5,709 5,353 # Male students 18,706 3,847 3,048 5,968 2,995 2,848 # Female students 16,837 3,503 2,634 5,481 2,714 2,505 Student graduation results, 2015/16 # Students in grade 6 21 20 20 21 25 24 # Male students in grade 6 10 10 10 10 12 11 # Female students in grade 6 11 10 10 11 13 13 # Graduates 21 20 20 21 25 24 # Male graduates 10 10 10 10 12 11 # Female graduates 11 10 10 11 13 13 Average UN score 149.0 113.7 112.3 160.7 187.2 194.7 Score in Indonesian 60.2 56.3 50.1 59.7 66.4 77.3 Score in math 54.8 50.2 45.2 52.9 64.7 72.0 Score in science 61.1 59.7 50.5 62.0 64.0 74.1 Note: NTT = East Nusa Tenggara; UN = United Nations. nearly 60 percent have toilets for female students, 89 million (US$6,360) in Landak and IDR 131 million and 50 percent have toilets for male students. In (US$9,360) in East Manggarai (table 12).35,36 Local comparison, in NTT, on average, 82 percent of the governments also contribute to school operational sample schools have toilets for teachers, 30 percent funds,37 except for the schools in Landak, which did have toilets for female students, and 21 percent have not receive funding from any local government. In the toilets for male students. other four study districts that reported having received financial support, district governments were the School Budget second most important source of school operational funds, although there were significant variations across All the schools receive operational funds from districts in the amounts the schools received. In West the central government. The amount received from Manggarai and East Manggarai, 16 and 12 percent of the central government under the School Operations the schools received IDR 33 million (US$2,360) and IDR Fund for academic year 2015/16 varied between IDR 23 million (US$1,640), respectively, from the district government. In Ketapang and Sintang, on average, 35 The exchange rate was approximately Rp 14,000 to US$1 at the approximately 90 percent of the schools received IDR date of publication. 36 This is in line with findings from Al-Samarrai and Cerdan-Infantes (2013): about half the public schools at the primary and junior secondary levels nationally reported not having received any 37 These are comprised of funds from the district-specific additional financial support from the district government in 2010. complement to the national School Operations Fund. 15. 10 million (US$715) from their district government. most frequent fees charged to parents are for school Al-Samarrai and Cerdan-Infantes (2013) find that uniforms, celebrations, and examinations. differences occur in the amount of resources allocated by district governments to schools for teacher salaries Language and Curriculum Utilized in compared with resources allocated for direct education Sample Schools support. Districts indeed tend to allocate a large part of Indonesian is the principal language used in their budgets to paying salaries––particularly to contract most of the schools in the study areas. More teachers––which leaves fewer resources available than 90 percent of the schools in West Kalimantan for direct support to schools. Provincial government and 70 percent in NTT use Indonesian (table 14, in funds are scarcer and not homogeneously distributed annex A). The remaining schools in NTT use Manggarai across districts. In NTT, only one school in the province as the main teaching language. In West Kalimantan, received additional provincial funding (IDR 18 million Malay (Ketapang) and Dayak (Sintang and Landak) are (US$1,285) for West Manggarai). In West Kalimantan, reported to be the principal languages of instruction in six schools in Ketapang and three schools in Sintang 3 to 7 percent of the schools, and by 2 to 9 percent of reported having received approximately IDR 5 million the teachers. (US$360) and IDR 13 million (US$930), respectively, per district, from the provincial government. The study found some discrepancies in the use of the national curriculum in the surveyed districts. Government funds comprise between 94.0 Most teachers (74 percent) teach at least four subjects, and 99.5 percent of school operational funds.38 and 14 percent teach one subject. For the curriculum, The remainder predominantly come from parent 99 percent of the principals reported that the 2004 contributions (on average, 1.33 percent), which vary curriculum is used in their school, whereas 94 percent widely in type and amount (table 13). Overall, the of the teachers reported using the 2006 curriculum. Table 12. Funding Sources for Sample Schools, Academic Year 2015/16 West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Source of school operation funds Central government 100 100 100 100 100 100 (%schools) Amount from central 108,695,824 105,881,608 88,737,600 108,972,376 119,043,496 130,656,944 government (Rp) Province government 4 5 0 7 3 0 (%schools) Amount from province 10,838,640 4,733,334 12,697,733 18,000,000 government (Rp) District government 53 92 0 89 16 12 (%schools) Amount from district 12,110,747 10,315,019 11,171,484 33,033,334 23,050,000 government (Rp) Village government 1 3 0 0 0 6 (% schools) Amount from village 2,400,000 3,000,000 1,800,000 government (Rp) Reported distribution of school operation funds by source (%) Government 97.7 97.3 99.5 98.3 97.6 94.3 Fees paid by parents 1.3 0.7 0.4 1.0 1.1 5.0 Community 0.1 0.2 0.0 0.0 0.3 0.4 contribution Other sources 0.8 2.0 0.1 0.6 1.0 0.4 Note: NTT = East Nusa Tenggara. 38 Here, “government funds” refers to national, provincial, district, and village funding sources. 16. SCHOOLING CONTEXT Table 13. Fees Charged to Parents, Academic Year 2015/16 West Kalimantan NTT West East All Areas Ketapang Landak Sintang Manggarai Manggarai Fees charged to parents Uniforms Number of schools 53 23 2 19 6 3 Amount (Rp) 1,146,943 550,870 90,000 1,773,053 88,333 4,573,334 Regular school payments/SPP/committee Number of schools 40 5 1 6 13 15 Amount (Rp) 4,534,075 27,000 25,000 2,459,500 3,001,846 8,494,800 Celebrations Number of schools 34 3 0 5 12 14 Amount (Rp) 218,176 19,333 280,000 249,000 212,286 Examinations Number of schools 24 8 2 7 6 1 Amount (Rp) 454,667 123,125 100,000 490,000 416,167 3,800,000 Student worksheets Number of schools 18 15 1 2 0 0 Amount (Rp) 673,222 452,333 8,000 2,662,500 Initial registration/admission fee Number of schools 11 4 0 0 3 4 Amount (Rp) 340,455 652,500 108,333 202,500 Activity fund (extracurricular) Number of schools 7 2 0 0 4 1 Amount (Rp) 64,429 22,500 23,000 314,000 Facilities/infrastructure construction fund Number of schools 8 3 1 1 0 3 Amount (Rp) 247,375 74,000 50,000 50,000 552,333 Others Number of schools 26 6 1 13 4 2 Amount (Rp) 337,423 94,500 2,000 544,846 117,250 326,000 Note: NTT = East Nusa Tenggara; SPP = social protection program. Only one percent of principals and teachers reported in East Manggarai reported having had no interruptions the use of 2013 curriculum. The data collection did during the academic year, with 244 effective teaching not ask additional questions that could explain these days. In contrast, 34 percent of the schools in Sintang discrepancies, and as such, this will be a good area of reported having had some interruptions, with an average investigations in future relevant studies. of 207 effective teaching days during the year. Instructional time in sample schools ranged from Teaching and Learning Time in Academic 26 to 33 hours per week. Teachers in sample schools Year 2015/16 spend on average 26 instructional hours every week About 25 percent of the sample schools reported teaching students in grade one. The weekly number that teaching and learning activities had been of instructional hours increases gradually with each interrupted at least once during the academic year. grade, reaching 31 hours in grade six. Schools in NTT Table 15 (in annex A) shows that, on average, the sample have slightly more instructional hours than those in schools had 226 effective teaching days during academic West Kalimantan, ranging from 27 hours in grade one year 2015/16, varying from 207 days in Sintang to 244 to 33 hours in grade six. days in East Manggarai. The number of interruptions varied quite significantly across the districts. The schools 17. School Supervision and Meetings In School principals or other staff also engaged in Academic Year 2015/16 several meetings with external stakeholders during academic year 2015/16. In most of the On average, about 90 percent of principals districts, approximately 90 percent of the schools reported having had a supervisory visit from had meetings between parents and teacher working the school superintendent or other officials groups during the year (figure 1). Meetings with during academic year 2015/16. Two-thirds of the education authorities and district and subdistrict teachers reported having been supervised by another technical implementing unit education agencies were education official while teaching during the year (table also common, with approximately 75 and 80 percent 16).39 Nearly all the principals (between 91 and 100 of the schools, respectively, reporting having had such percent, depending on the district) reported having meetings. Figure 2 shows the topics covered during had regular internal meetings between principals and these meetings, as reported by teachers. In these teachers (approximately six times during academic year discussions, the student learning process was the most 2015/16). This implies that there were approximately frequently discussed issue, followed by the curriculum, bimonthly internal meetings between principals and students’ grades, and teaching quality. teachers.40 A similar, slightly lower frequency of internal meetings was reported by teachers. Figure 1. School Meetings with External Stakeholders, 2015/16 100 90 Percent of Schools 80 70 60 50 40 30 20 10 0 With District With UPTD With With Teacher With Parents With Other Education Foundation Working school office Group All Areas Ketapang Landak Sintang West Manggarai East Manggarai Table 16. School Supervision and Meetings, Academic Year 2015/16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai School and teacher supervision Visit from superintendant/officials/foundation 88 80 80 90 100 97 Teacher supervision during teaching 67 71 62 64 67 73 Internal meetings between principals and teachers Principal report 95 97 94 91 100 100 Average # internal meetings (principal) 6 5 5 6 8 6 Teacher report 0.880 0.880 0.840 0.880 0.880 0.930 Average # internal meetings (teachers) 5 5 4 4 7 6 Participation of teachers in preparing school programs Entirely 58 60 32 48 71 88 Partially 22 27 24 28 14 9 Note: NTT = East Nusa Tenggara. 39 A teacher could be supervised by another teacher, the principal, 40 The school year runs from mid-July to mid-June in Indonesian the superintendent, or some other person working for a public public schools. educational institution. 18. SCHOOLING CONTEXT Figure 2. Topics Covered During Internal Meetings 70 Having Had an Internal Percent of Teachers 60 50 Meetings 40 30 20 10 0 Learning Curriculum Student Teaching Student Facilities/ Process Grades Quality Behavior Allowance All Areas Ketapang Landak Sintang West Manggarai East Manggarai civil servant (PNS) teachers—60 and 40 percent, Teachers and Principals respectively, over the entire sample. Among non-PNS Characteristics of Principals and Teachers teachers (1,162), 814 (about 42 percent of all the sample in Study Schools teachers) were school-contracted teachers; 302 (16 percent) were contract teachers; and the remaining 46 Almost all principals, but only 40 percent of the teachers (2 percent) had another employment status, teachers, were civil servants. Only two principals such as community or part-time schoolteacher. The in the sample survey were not civil servants. The study distribution of contract and school-contracted teachers sample was comprised of 245 principals and 14 acting varies widely between districts, but school-contracted principals, among the 270 primary schools surveyed. teachers were more common than contract teachers For the other 11 primary schools, respondents to the in the sample schools, except in Ketapang. The school- principal questionnaire were teachers or other school contracted teachers were hired by the schools and staff, who answered the questionnaire on behalf of an comprised approximately 27 percent of the teachers in absent or acting principal. More than 90 percent of the public primary schools, whereas the contract teachers schools are public schools, with three types of teacher were hired on fixed-term contracts on the government status: permanent, contract, and school-contracted payroll.42 According to World Bank (2013b), the share of teachers. Permanent teachers are tenured civil servants non-PNS primary school teachers increased from 25 to (PNS) hired by the central government, while contract 35 percent across Indonesia between 2006 and 2010. teachers are hired by district or provincial governments In 2010, nearly half of the Indonesian schools had under annual contracts. Meanwhile, schools hire school- between 20 and 40 percent non-PNS teachers, and a contracted teachers with a temporary employment quarter of the schools had more than 40 percent non- status. The study schools have 2,301 teachers, of whom PNS teachers.43 about 83 percent were surveyed using the teacher PNS teachers were predominantly male (60 questionnaire.41 In Ketapang, only about a third of the percent) and age approximately 44 years, whereas teachers were civil servants. Overall, the share of civil non-PNS teachers were mostly female (60 percent) servants in the sample schools was remarkably lower and age approximately 30 years. There was a clear than is commonly found in other studies of Indonesian difference between the demographic characteristics of schools. Chen (2011) finds that 70 percent of teachers PNS and non-PNS teachers. In comparison, World Bank were PNS, on average, from a sample of 400 public (2008) finds that female teachers constituted 55 percent primary schools located in 54 districts throughout of primary school teachers in Indonesia. Approximately the country, whereas World Bank (2008) reports that 95 percent of PNS teachers were married, compared approximately 52 percent of teachers in primary schools with 76 percent of non-PNS teachers. Similarly, 95 in remote areas were PNS. percent of PNS teachers were parents, compared with There were more non–civil servant teachers, contract 71 percent of non-PNS teachers. Among the teachers teachers, and school-contracted teachers than who were parents, PNS teachers had three children on 41 The remaining 17 percent of the teachers were not present 42 Suharti 2013. at school on the day the survey was implemented, because they 43 World Bank 2013b were not scheduled to teach on that day or they were absent (see section 5). 19. Table 17. Principal and Teacher Demographics West Kalimantan NTT All Areas Ketapang Landak Sintang West East Manggarai Manggarai Respondent status (#) Schools 270 59 51 88 38 34 Principal 245 55 38 82 38 32 Acting principal 14 2 10 1 0 1 Respondent on behalf of principal 11 2 3 5 0 1 PNS teachers 755 140 133 240 137 105 Non-PNS teachers 1162 280 167 345 195 175 Contract teacher 302 138 24 58 15 67 Honor teacher 814 104 140 285 177 108 Other employment status 46 38 3 2 3 0 Principal demographic characteristics Age (years) 48 47 49 47 51 51 Gender: male (%) 84 81 96 73 100 79 Marital status: married (%) 97 97 96 98 97 94 Marital status: single (%) 2 3 4 1 3 0 Marital status: other (%) 1 0 0 1 0 6 Has children (%) 98 98 96 99 95 100 Average # children 3 3 3 3 5 4 PNS teachers demographic characteristics Age (years) 44 43 44 45 45 45 Gender: male (%) 60 57 58 61 61 65 Marital status: married (%) 95 97 95 94 96 95 Marital status: single (%) 3 2 3 3 1 3 Marital status: other (%) 2 1 2 3 3 2 Has children (%) 95 95 93 95 96 96 Average # children 3 2 3 3 4 3 Non-PNS teachers demographic characteristics Age (years) 30 30 32 31 30 30 Gender: male (%) 40 36 45 39 36 45 Marital status: married (%) 76 78 74 78 74 74 Marital status: single (%) 22 21 24 18 24 25 Marital status: other (%) 2 1 2 4 2 1 Has children (%) 71 73 69 77 66 65 Average # children 1 1 1 1 1 1 Note: NTT = East Nusa Tenggara; PNS = civil servants. average, whereas non-PNS teachers had one child on however, the law has not yet been fully implemented average. Lastly, only 3 percent of PNS teachers and 2 for the sample schools. Principal and teacher percent of principals were single; 22 percent of non- educational attainment levels are presented in figure PNS teachers were single. As shown in table 17, the 3. Data from the teacher census show that only 14 typical principal in the sample schools was a married percent of primary school teachers in remote areas man approaching age 50, with three children. held a bachelor’s degree in 2010, compared with 27 percent of all primary school teachers nationally.44 Only slightly more than half of the principals and teachers in study schools held an undergraduate degree. The 2005 national Teacher Law requires that all teachers have a four-year bachelor’s degree; 44 World Bank 2013b. 20. SCHOOLING CONTEXT In the sample schools, similar shares of principals degree (or higher) in Ketapang and Sintang. (66 percent) and PNS teachers (60 percent) had the Less than one-third of teachers in the study schools required educational attainment (figure 3). Indeed, 18 have been certified. The rates of certification are percent of principals had only a high school diploma, rather low, especially considering that the 2005 Teacher whereas 29 percent of PNS teachers had this as their Law stipulates that all teachers teaching in Indonesian highest education level. Fifty percent of non-PNS schools must have completed the certification process teachers had a bachelor’s degree, on average. However, by 2015. Certification ensures that teachers possess this statistic masks important variation across districts, the proper competencies and provides them with a as only 32 percent of non-PNS teachers in Ketapang certification allowance equivalent to the base salary. (West Kalimantan) held the officially required degree, Table 18 shows that 34 percent of PNS teachers were and in West Manggarai (NTT), 78 percent held the same certified, and only 12 percent of non-PNS teachers were degree. In this respect, non-PNS teachers in the two certified. These findings could reflect a catch-up process NTT districts were highly qualified compared with those in recent years, given that two-thirds of the certified in other districts. More than 70 percent of non-PNS teachers have been certified since 2013. teachers in the NTT districts had at least a bachelor’s degree or higher. In contrast, approximately one-third of non-PNS teachers held the required bachelor’s Figure 3. Principal and Teacher Education Levels Principals PNS Teacher Non- PNS Teacher 100 90 80 70 60 50 40 30 20 10 0 High School Degree All Areas Ketapang Landak Sintang West Manggarai East Manggarai All Areas Ketapang Landak Sintang West Manggarai East Manggarai All Areas Ketapang Landak Sintang West Manggarai East Manggarai College Degree Bachelor, Master and Above Table 18. Teacher Certification Status All West Kalimantan NTT Areas Ketapang Landak Sintang West Manggarai East Manggarai PNS teachers Certified (% PNS teachers) 34 34 34 38 34 23 Certified in 2015-17 (% certified) 25 27 27 19 33 29 Certified in 2013-14 (% certified) 38 48 27 41 33 38 Certified in 2011-12 (% certified) 32 19 38 38 26 33 Certified in 2010 and earlier (% certified) 5 6 9 2 9 0 Note: NTT = East Nusa Tenggara; PNS = civil servants. 21. Figure 4. Principal and Teacher Longevity at Current School Principals PNS Teacher Non- PNS Teacher 100 80 60 40 20 0 All Areas Ketapang Landak Sintang West Manggarai East Manggarai All Areas Ketapang Landak Sintang West Manggarai East Manggarai All Areas Ketapang Landak Sintang West Manggarai East Manggarai 2015-2017 2011-2014 2006-2010 2005 and before Table 19. Principal and Teacher Work Experience West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Principal started at first school as principal (% principals) Principal’s first school is current school 70 56 78 75 68 68 2011-14 6 7 2 3 5 15 2006-10 7 15 2 6 8 3 2005 and earlier 17 22 18 15 18 15 PNS teacher started at first school (% PNS teachers) Teacher’s first school is current school 38 33 38 40 31 50 2015-17 0 0 0 0 0 0 2011-14 2 4 5 0 2 0 2006-10 11 8 17 8 16 8 2005 and earlier 49 56 41 53 51 42 Non-PNS teacher started at first school (% Non-PNS teachers) Teacher’s first school is current school 74 63 74 81 74 80 2015-17 1 0 1 1 1 2 2011-14 8 10 10 4 10 5 2006-10 11 16 8 8 10 11 2005 and earlier 6 11 7 5 5 2 Note: NTT = East Nusa Tenggara; PNS = civil servants. PNS teachers differ significantly, compared with On overall career experience, PNS teachers again principals and non-PNS teachers, in longevity at differ significantly from principals and non-PNS their current school. Figure 4 shows longevity at teachers. Among PNS teachers, approximately 60 current school, and table 19 presents overall experience percent worked in another school previously and 49 (years working). Sixty percent of PNS teachers have percent started working as teachers at their current continued to work in their current school since before school prior to 2005. Only 30 percent of principals and 2005. Between 20 and 30 percent of principals held 25 percent of non-PNS teachers had previously held their position at their current school for fewer than two their current respective positions in other schools. years, and 30 to 60 percent of principals did so for two Among non-PNS teachers who had previously worked to five years. Non-PNS teachers are the most recent in a school, 25 percent started before 2005 and 46 staff in the sample schools—29 and 32 percent worked percent started during 2006–10. in their current school for fewer than two years and two to five years, respectively. 22. SCHOOLING CONTEXT Table 20. Principals’ Characteristics (% Principals) All West Kalimantan NTT Areas Ketapang Landak Sintang West Manggarai East Manggarai Local language ability No or limited fluency 8 5 8 14 5 0 Average fluency 8 10 12 8 0 6 High or perfect fluency 84 85 80 78 95 94 Birth location compared with school location Same village as school 24 29 25 17 24 35 Other village, same subdistrict 30 32 27 27 45 18 Other subdistrict, same district 19 15 22 24 16 15 Other district, same province 16 10 24 11 11 32 Other province 11 14 2 20 5 0 Location of main residence compared with school location Same village as school 69 68 55 76 66 76 Other village, same subdistrict 26 31 29 23 32 18 Other subdistrict, same district 4 0 14 1 3 6 Other district, same province 0 0 2 0 0 0 Other province 0 2 0 0 0 0 # days spent in main residence last year 355 354 343 358 358 365 Reasons for living in main residence Owns a house there 69 66 78 59 68 85 Location of official residence 16 20 12 22 8 6 Spouse/children live there 39 61 31 40 34 12 Parents/relatives live there 9 17 6 13 3 0 Close to school 37 32 33 50 45 12 Other 10 10 2 17 11 6 Note: NTT = East Nusa Tenggara. Living Conditions Most principals and teachers live in the same village as their schools and spend a negligible Principals in the sample schools are predominantly amount of time and money to commute. As of local origin, and teachers are relatively well shown in tables 22 and 23 (in annex A), the median integrated into the study areas. Among the travel time to get to school for principals living in the principals, 84 percent speak the local language with same village as their school is five minutes, and the high fluency; nearly 75 percent were born in the same corresponding transportation cost is null. For principals district where they work; and 70 percent live in the same living outside the village, the median travel time is 30 village as their school (table 20). Among the teachers, minutes, and the median cost is IDR 8,000 (US$0.60). 81 percent speak (near-perfectly) the local language; In general, teachers have similar travel conditions as 80 percent were born in the same district where they their principals. However, teachers who live outside work; and 81 percent live in the village where the teach the village (20 percent of the teachers) where they (table 21). However, only 57 percent of the teachers teach spend approximately IDR 4,500 (US$0.30) on own a house in the village of their school, compared transportation (one way). with 69 percent of the principals. Several previous studies have found that teachers and principals who Activities at School and Outside School were born outside the province where their school was located had lower absence rates than those born in the In general, the teachers reported teaching more same province.45 than 90 percent of the scheduled hours. Most teachers worked at one school only. Table 24 lists the activities–– at school and outside school––in which teachers in the 45 Toyamah et al. 2010; ACDP 2014. study schools reported taking part. When surveyed, in the 23. Table 21. Teachers’ Characteristics (% Teachers) West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Local language ability No or limited fluency 1 0.170 1 0.080 0.080 0.060 Average fluency 0.090 0.150 0.110 0.080 0.050 0.080 High or perfect fluency 0.810 0.690 0.790 0.840 0.860 0.860 Birth location compared with school location Same village as school 39 39 39 40 31 42 Other village, same subdistrict 24 21 21 23 30 27 Other subdistrict, same district 17 14 24 17 19 13 Other district, same province 12 7 10 12 14 18 Other province 9 18 6 8 6 0 Location of main residence compared with school location Same village as school 81 81 72 86 80 82 Other village, same subdistrict 16 16 19 13 19 18 Other subdistrict, same district 2 2 8 1 1 0 Other district, same province 0 0 1 1 1 0 Other province 0 0 0 0 0 0 # days spent in main residence last year 359 355 357 361 359 361 Reasons for living in main residence Owns a house there 57 56 61 61 48 54 Location of official residence 14 15 12 17 16 5 Spouse/children live there 63 85 47 71 56 40 Parents/relatives live there 25 36 24 24 19 19 Close to school 32 41 23 28 44 25 Other 4 10 4 2 2 0 Note: NTT = East Nusa Tenggara. previous week teachers had taught 26 hours, on average, Teachers also reported spending time to give and of nearly 28 scheduled hours. However, there were sizable assess exams and homework. On average, there variations across districts, with the number of scheduled were approximately 11 primary school student exams teaching hours ranging from 26 hours in Landak and East conducted during the academic year, although there Manggarai to 33 hours in West Manggarai. Considering were significant variations across districts, from seven that students usually attend school six days a week exams in Landak to 14 in Ketapang. Homework is given in Indonesia, this implies that teachers in the sample daily by about 25 percent of the teachers, and weekly by areas teach, on average, between 4.3 and 5.5 scheduled more than 90 percent. Approximately 90 percent of the teaching hours daily. This differs from the number of teachers reported assessing homework themselves, weekly realized teaching hours, which ranged from 22 spending between three hours (Landak) and five hours hours in East Manggarai to 31 hours in West Manggarai. (East Manggarai) assessing homework each week. These self-reported weekly teaching hours are relatively On weekly time allocation, assessing daily exams and high compared with the average national teaching load. homework was the third most important teacher World Bank (2008) reports that approximately half of the activity, after teaching and preparing lesson plans. primary school teachers nationally have a workload of Other teaching-related tasks that occur less frequently fewer than 18 hours weekly. More recently, Suharti (2013) during the academic year include assessment of finds that nationally only 44 percent of teachers teach midterm and final exams (between four hours in Landak the minimum level of teaching hours required by law (24 and 11 hours in East Manggarai, on a monthly basis) hours), while 53 percent of teachers in rural areas, and 59 and teacher training and self-development (allocated percent in remote areas, work fewer than 18 hours each monthly, ranging from four hours in Landak to eight week. hours in Sintang). 24. SCHOOLING CONTEXT Table 24. Teachers’ Activities at School, Academic Year 2015/16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Teaching # schools at which teaching 1 1 1 1 1 1 Scheduled # teaching hours last week 28 28 26 27 33 26 Realized # teaching hours last week 26 26 23 27 31 22 Average # students in class daily 20 18 18 19 22 21 Exams and homework - frequency # exams conducted in academic year 11 14 7 13 10 10 Homework given daily (% teachers) 26 36 27 24 20 22 Homework given at least once a week (% teachers) 93 93 87 95 93 95 % homework assessed by teacher self 90 88 93 91 85 92 Weekly hours spent on teaching-related tasks Learning plan preparation 5 6 4 4 6 7 Teaching activities 18 17 17 20 14 20 Assessment of daily exams and homework 4 5 3 4 4 5 Remedial activities 2 2 1 2 2 2 Extracurricular activities 1 1 1 1 2 2 Monthly hours spent on teaching-related tasks Assessment of midterms and final exams 7 9 4 6 7 11 Teacher self-development and training 6 4 3 8 6 7 Research activities 0 0 1 0 0 0 Creation of teaching innovative learning tools 1 1 0 1 1 2 Additional roles in school (% teachers) Teacher has additional activities at school 68 75 51 66 67 81 Homeroom teacher 50 42 65 51 38 65 Extracurricular supervisor 37 38 10 43 38 42 Dapodik operator 11 11 11 13 13 6 Library supervisor 6 9 4 6 5 2 School committee administrator 2 1 1 3 1 4 Note: NTT = East Nusa Tenggara. Nearly all principals also teach, while 68 percent Manggarai, 31 percent of the surveyed teachers had of the teachers undertake roles other than attended training in the previous six months, and 43 teaching, with very different access to capacity percent had done so in the previous 12 months. development trainings. In addition to their main The majority of principals and about 70 percent principal activities, nearly all the principals had teaching of teachers also held another job. Among the scheduled in the previous week—on average 14 of 15 principals (table 25), 68 percent work in agriculture, hours (table 25). More than 90 percent of the principals spending between four hours (West Manggarai) and received education-related training sometime in the 24 hours (Landak and Sintang) in agricultural activities past three years. Teachers also took on other roles, monthly. A minority of the principals (3 percent) including homeroom teacher (50 percent of teachers), reported having an extra teaching job (outside school) extracurricular supervisor (37 percent), and principal in Ketapang and Landak. Similarly, agriculture is the education data operator (11 percent). On additional most common second work activity, occupying 54 teacher training, table 26 shows that about 8 percent of percent of the teachers. Eleven percent of the teachers teachers in Sintang had attended a training workshop work in non-agricultural pursuits, and 5 percent during the previous six months, and 18 percent had teach outside school. For example, teachers spent an done so during the previous 12 months. In East average of 32 hours in the previous month (about eight 25. Table 25. Principals’ Additional Activities: Trainings, Other Jobs, and Involvement in Local Organizations West Kalimantan NTT All Areas Ketapang Landak Sintang West East Manggarai Manggarai Activities at school Teaching at same school (% principals) 94 92 92 98 89 97 Scheduled # teaching hours (if teaching) 15 18 20 14 14 9 Realized # teaching hours (if teaching) 14 16 16 14 12 7 Ever received training in education (% principals) 94 90 94 97 92 94 Received training in education in past 3 years 90 85 90 94 92 82 (% principals) Other jobs, past month Teaching outside school (% principals) 3 8 4 0 0 0 # hours spent, if teaching outside school 6 5 8 Median monthly income from extra teaching job (Rp) 450,000 200,000 700,000 Working in agriculture (% principals) 68 61 75 72 58 73 # hours spent, if working in agriculture 18 18 24 24 4 8 Median monthly income from agriculture job (Rp) 400,000 500,000 294,667 500,000 187,500 170,833 Other non-agricultural job (% principals) 8 18 8 5 3 3 Hours spent, if other job 24 38 15 8 4 4 Median monthly income from other job (Rp) 708,333 1,125,000 1,080,000 500,000 1,250,000 500,000 Involvement in local organizations (% principals) Active in local organizations 64 81 48 69 55 56 Local government organization 13 20 20 6 8 15 Religious/youth/farmer organization 54 66 44 51 55 53 Political party or nongovernmental organization 0 0 2 0 0 0 Education/health/social organization 24 27 12 44 3 12 Note: NTT = East Nusa Tenggara. hours weekly) working in agriculture, compared with 26 additional median monthly income (on average, IDR hours teaching each week, on average. Although time 700,000), although there are significant variations spent by teachers in agricultural or other pursuits may between districts (table 25). Similarly, for teachers not appear to be overly time consuming or onerous, it who have an extra agricultural job, their additional could be argued that it is time taken away from teaching median monthly income varies from IDR 167,000 or leisure activities. In comparison, the amount of time (US$12) to IDR 437,000 (US$31) in West Manggarai spent on an additional job is remarkably higher than, and Sintang, respectively. For those who have an extra for instance, the amount of time spent on training and non-agricultural job, their additional median monthly self-development per month (five hours on average, income is more substantial—on average, IDR 500,000 table 26). This may be related to the low proportion (US$36) and up to IDR 833,000 (US$60) in Sintang. of teachers (34 percent) in the sample schools who Many of the principals and teachers participate in reported being certified. Indeed, De Ree et al. (2018) local organizations. Sixty-four percent of the principals find that an important effect of the certification are involved in one or more local organizations—religious, program and its certification allowance is to reduce the youth, or farmer organizations (84 percent); education, probability of teachers holding a second job. health, or social organizations (38 percent); and local Additional jobs do not provide principals and government organizations (20 percent) (table 25). Table teachers with significant additional income. 26 shows that about 55 percent of the teachers (1,048) Principals with an extra agricultural job received reported that they are involved in local organizations. additional median monthly income varying from less These teachers are mainly involved in religious, youth, than IDR 200,000 in the NTT districts to IDR 500,000 or farmer organizations (84 percent across all types of (US$36) in Sintang and Ketapang. The principals who organizations); education, health, or social organizations have an extra non-agricultural job receive the highest (24 percent); and local government organizations (20 26. SCHOOLING CONTEXT Table 26. Teachers’ Additional Activities: Trainings, Other Jobs, and Involvement In Local Organizations West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Education training (% of teachers) Attended training workshop in past 6 months 17 18 15 8 20 31 Attended training workshop in past 12 months 29 33 26 18 34 43 Other jobs, past month Teaching outside school (% teachers) 5 9 12 2 2 1 # hours spent, if teaching outside school 21 21 21 17 32 17 Median monthly income from extra teaching job (Rp) 200,000 158,333 275,000 180,000 300,000 143,750 Working in agriculture (% teachers) 54 46 61 62 41 60 # hours spent, if working in agriculture 32 35 34 39 16 24 Median monthly income from agriculture job (Rp) 300,000 333,333 266,667 437,500 166,667 191,667 Other non-agriculture job (% teachers) 11 21 16 8 7 2 Hours spent, if other job 42 43 37 43 52 31 Median monthly income from other job (Rp) 500,000 500,000 500,000 833,333 250,000 654,167 Involvement in local organizations (% teachers) Active in local organizations 55 62 48 50 55 60 Local government organization 20 16 20 19 17 29 Religious/youth/farmer organization 84 76 84 79 94 92 Political party or nongovernmental organization 1 2 3 0 1 2 Education/health/social organization 24 39 19 27 8 14 Note: NTT = East Nusa Tenggara. percent). Involvement in local organizations may affect may lead teachers to be more responsive to community teachers’ performance, for example, by increasing their demands for improvement in teaching outcomes, for workload. This would reflect the findings of ACDP (2014), example, through increased peer pressure to perform which reports high rates of absenteeism among teachers better. who are also involved in community organizations.46 Principals conducted teacher performance However, being more involved in their local communities evaluation in the majority of the schools. More Figure 5. Median Monthly Total Income (Idr) 46 In ACDP (2014), less than 1 percent of the teachers reported being involved in government programs as facilitators. 27. Table 27. Evaluation of Teachers by Principal, Academic Year 2015/16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Principal reporting (%) # Principals reporting having conducted teacher evaluation 195 44 38 57 28 28 Evaluation criteria Teacher discipline/behavior 87 86 74 93 89 93 Teaching capability 76 59 89 84 68 75 Teacher attendance 75 70 74 79 79 75 Performance or behavior of teacher’s students 58 50 55 60 61 71 Creativity outside the class 24 23 13 28 7 50 Others 53 48 39 49 71 68 Evaluation results communicated to teachers 98 98 97 98 96 100 Recognition of high-performing teachers Nothing 71 67 83 45 97 100 Oral praise 27 32 10 55 3 0 Appreciation certificate 0.1 4 0 0 3 0 Help with promotion and/or self-development opportunities 3 4 0 0.1 13 0 Promotion to become principal 2 4 0 4 0 0 Financial reward 4 2 6 8 0 0 Teacher reporting (%) # Teachers reporting having been evaluated 1,506 342 195 446 263 260 Evaluation criteria Teacher discipline/behavior 80 80 70 77 86 86 Teaching capability 62 71 57 63 59 56 Teacher attendance 70 72 60 73 70 71 Performance or behavior of teacher’s students 57 60 50 56 55 62 Creativity outside the class 24 27 15 20 26 33 Others 35 35 21 26 48 47 Evaluation results communicated by principal 67 65 68 65 62 78 Evaluation results considered fair and objective 97 96 97 98 96 98 Recognition of high-performing teachers Nothing 36 36 52 38 25 28 Oral praise 56 53 41 53 67 69 Appreciation certificate 0 2 0 0 1 0 Help with promotion and/or self-development opportunities 3 3 3 2 3 4 Promotion to become principal 0 0.1 0 0 0 0 Financial reward 3 5 0.1 4 5 1 Note: NTT = East Nusa Tenggara. than 70 percent of principals reported having evaluated criteria of principals as reported by teachers—teacher teachers in their school, and nearly 80 percent of discipline/behavior (80 percent), teacher attendance teachers reported having been evaluated by their (70 percent), teaching capability (62 percent), and the principal during academic year 2015/16 (table 27). The performance or behavior of students (57 percent). main evaluation criteria included teacher discipline or Approximately 25 percent of school principals behavior (87 percent), teaching capability (76 percent), reported having communicated the evaluation results attendance (75 percent), and the performance to teachers. In contrast, approximately 67 percent of or behavior of their students (58 percent). These teachers reported having received their evaluation percentages correspond with the main evaluation results from their principals, with 97 percent of these 28. SCHOOLING CONTEXT Table 28. Principals’ Salary Delivery Mode West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Means of receiving salary (% principals) Transferred to bank account 47 0 100 84 3 9 Cash from regional education office 3 9 0 1 3 6 Cash from subdistrict education office (UPPT) 14 11 0 1 51 30 Cash from school 32 81 0 12 38 39 Other 3 0 0 1 5 15 Distance, travel time, and cost from school to salary pickup location Salary is picked up outside village (% principals) 95 91 100 99 89 91 One-way distance (km) 52 42 47 89 19 19 Travel time (minutes) 147 99 134 221 106 91 Median transportation cost (Rp) 35,000 16,500 25,000 67,500 50,000 50,000 Frequency at which salary is picked up outside village (% principals) Monthly 97 96 100 94 100 100 Bimonthly 1 2 0 2 0 0 Quarterly 1 0 0 2 0 0 Other 1 2 0 2 0 0 Note:km = kilometers; NTT = East Nusa Tenggara. teachers considering the evaluation results to be fair There is more variation in teachers’ salary and objective. Seventy-one percent of principals did delivery modes. Among the teachers, 46 percent not recognize or reward high-performing teachers in receive their salary directly from the school, 26 percent any way, although 27 percent of principals reported through their bank accounts, 15 percent from the verbally praising high-performing teachers. However, subdistrict education office, and 13 percent from the 56 percent of teachers reported having been verbally district education office (table 29). More than half of the praised by their principals, and 36 percent reported teachers (55 percent) must travel outside the village to that they did not receive any particular recognition. pick up their salary. Among them, 75 percent do this every month and 17 percent do so quarterly. In a similar Incentives and Motivations of Principals dynamic as for principals, teachers travel between and Teachers 37 kilometers (West Manggarai) and 81 kilometers Salaries and Allowances (Sintang), for approximately 150 minutes, to reach their salary pickup location. The median transportation cost Almost all principals had to travel more than to the pickup location for teachers varies between IDR two hours to get their salaries. Nearly 50 percent 10,000 (US$0.7) in Ketapang and IDR 75,000 (US$5.4) of principals receive their salary through their bank in East Manggarai. accounts, 32 percent receive cash payments directly The incomes of principals and teachers differ from their school, and 14 percent receive cash payments widely according to their civil servant and from their local subdistrict education office (table 28). certification status. Principals and teachers in the Ninety-five percent of principals must travel outside their sample areas reported receiving a monthly median village to pick up their salary every month, traveling an income of about IDR 8.25 million (US$590) and about average of 52 kilometers. The distance ranges from 19 IDR 1.5 million (US$107), respectively. Figure 5 shows kilometers in the NTT districts to 89 kilometers in Sintang, the monthly median income received by principals with an average (one-way) travel time of approximately and teachers during the past year,47 according to their 150 minutes. The median (one-way) transportation costs for principals to reach their salary pickup location 47 The respondents reported their total income received during the past 12 months, which was then divided by 12 to present varies between IDR 16,500 (US$1.2) in Ketapang and IDR the corresponding monthly figures. Several of the principals and 67,500 (US$4.8) in Sintang. teachers did not receive their salary on a monthly basis. The self-reported income figures discussed here may be prone to measurement error, given that the teachers received salaries irregularly and in amounts that varied from one time to the next. 29. certification and contract (PNS or otherwise) status. In no additional allowances. There are also differences, the figure, important differences depend on teacher albeit smaller, in teachers’ base salaries. School- status (PNS or another category of teacher). In all the contracted teachers received a monthly median study areas, the median income of certified teachers base salary of IDR 0.6 million, with remarkably little is roughly similar to the median income of principals, differences extant between districts. Contract teachers around IDR 8.4 million (US$600). In 2018, nationwide, and PNS teachers received IDR 1.2 million and IDR 2.9 close to 1.9 million elementary and secondary school million, respectively. Principals and certified teachers teachers in Indonesia were recipients of the certification received a base salary of approximately IDR 3.8 million. allowance, with an annual budget of US$5.6 billion. The Principals and teachers may also receive same year, close to 69,000 elementary and secondary professional and remote area allowances. The school teachers were recipients of the remote area 2005 Teacher Law stipulates that certified teachers allowance, with an annual budget of US$183 million. receive a certification allowance equal to their base In Sintang, certified teachers receive slightly more salary. It also provides a remote area allowance, which than principals (median incomes of IDR 10.8 million is equal to the base salary for certified teachers. For (US$770) and IDR 10.1 million (US$720), respectively). noncertified teachers who have taught for at least two Next are noncertified PNS teachers, who receive an years and at least 24 hours weekly in an eligible school, average median monthly income of about IDR 4.6 the remote area allowance provides an additional million (US$330), ranging from IDR 3.4 million (US$243) allowance of approximately IDR 1.5 million (US$107) per in East Manggarai to IDR 6.3 million (US$450) in Sintang. month.48 The additional income allowance (tambahan Non-PNS and noncertified teachers have much lower penghasilan) is specifically for noncertified teachers. incomes, with little variation across districts. The Based on Presidential Decree 52/2009, the amount contract teachers’ monthly median income ranged of the additional income allowance for noncertified between IDR 0.85 million (US$61) (East Manggarai) and teachers is fixed at IDR 250,000 (US$18) per month. IDR 1.5 million (US$107) (Landak and West Manggarai) during the previous year. The school-contracted Nearly all the principals receive some additional teachers’ median monthly income was around IDR 0.55 allowance, on average, a monthly median of million (US$40). approximately IDR 3.5 million (US$250). About 70 percent of the surveyed principals receive the Differences in income for teachers are largely the certification allowance, varying from IDR 2 million result of differences in the amount of additional (US$143) per month in West Manggarai to IDR 3.7 allowances received. This can be observed by million (US$264) per month in Landak (table 30, in separately examining the base salary and additional Figure 6. Median Monthly Base Salary (IDR) allowances of teaching staff (figures 6 and 7). During the annex A). Approximately 31 percent of principals (80 past year, only principals, certified teachers, and PNS respondents) reported having received the remote teachers received an additional allowance, at monthly area allowance in the past 12 months. The monthly medians of IDR 3.5 million (US$250), IDR 3.6 million median amount varies dramatically between districts, (US$257), and IDR 0.96 million (US$69), respectively. Among other types of teachers, the majority received 48 Tomayah et al. 2010. 30. SCHOOLING CONTEXT Figure 7. Median Monthly Total Allowances (IDR) from IDR 0.83 million (US$60) on average in Ketapang certification allowance; the median amount these to IDR 2.8 million (US$200) in Sintang. No principal in teachers received was IDR 1.3 million (US$93). the East Manggarai sample schools reported receiving The median monthly amount of the remote area the remote area allowance during the past 12 months. allowance varied dramatically between districts. In Landak, principals reported receiving a monthly Approximately 16 and 6 percent of PNS and non-PNS median remote area allowance of IDR 2.7 million teachers, respectively, reported having received the (US$193) and median additional income allowance of remote area allowance in the past 12 months (table 31). IDR 0.1 million (US$7). For PNS teachers, the monthly remote area allowance The median monthly amount of the additional ranged from IDR 0.725 million (US$52) in Ketapang income allowance for noncertified teachers is to IDR 2.7 million (US$193) in Sintang. For non-PNS significantly less than the remote area allowance teachers, the remote area allowance ranged from IDR across all the districts, except East Manggarai.49 0.75 million (US$54) in East Manggarai, on average, to Table 31 (in annex A) shows that approximately 84 IDR 1.4 million (US$100) in Landak and Sintang. percent of PNS teachers and 45 percent of non-PNS The median monthly additional income allowance teachers received some additional allowance during for noncertified teachers was significantly academic year 2015/16. The monthly medians of the less than the remote area allowance across additional income allowance were approximately IDR all the districts. The additional income allowance 1.3 million (US$93) and IDR 0.2 million (US$14) for for noncertified teachers was received by 35 and 11 PNS and non-PNS teachers, respectively. In a similar percent of PNS and non-PNS teachers, respectively, in dynamic as for principals, the largest allowance for the sample areas (table 31). The exception was non- teachers was the certification allowance, followed by PNS teachers, who received IDR 0.75 million (US$54) the remote area allowance, and lastly the additional in remote area allowance and IDR 1 million (US$71) in income allowance for noncertified teachers. For all additional income allowance for noncertified teachers, three types of allowances, PNS teachers received on average, per month. In Landak, not one non-PNS amounts that were significantly greater than those teacher had received the additional income allowance received by non-PNS teachers. for noncertified teachers in the past 12 months. Approximately 32 percent of PNS teachers in The number of principals and teachers who the sample received the certification allowance. received the certification allowance increased The median amount varied from IDR 1.8 million in steadily between 2014 and 2016. Data on salary West Manggarai to IDR 3.3 million (US$129) in Landak and allowance delivery performance for teachers and (table 31). Only five non-PNS teachers received the principals during 2014–16 are presented in table 32 (in annex A). The vast majority of the principals and 49 The amount of the additional income allowance for noncertified teachers (combined, more than 90 percent) reported teachers reported here differs from the amount specified in Presidential Decree 52/2009. The amount reported may include having received the entire combined amount available the additional income allowance for noncertified teachers, which may have been funded through the local budgets of some district via the different allowances during 2014 and 2015. governments. 31. Table 33. Principals’ Opinions of Teachers and Students (% Principals) West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Opinion of teachers Teaching skills of school teachers Very bad/bad 0 2 0 0 0 0 Good/very good 73 72 75 61 87 85 School lacks teachers 71 68 81 86 45 52 Opinion of students Discipline/attendance of school students Very bad/bad 2 2 6 0 3 0 Good/very good 74 68 75 74 74 82 Ability level of school students Very bad/bad 9 3 12 13 8 6 Good/very good 39 39 43 34 42 38 Factors hindering students’ learning Lack of school facilities and infrastructure 66 71 65 68 55 65 Lack of parents’ awareness 50 42 35 61 45 59 Lack of teachers 32 29 43 47 5 12 Geographical factors 30 19 22 45 26 26 Parents’ economic situation 27 27 20 23 24 53 How to improve students’ learning Provide adequate school facilities/infrastructure 67 80 65 66 53 65 Improve teachers’ quality 43 53 27 51 42 32 Improve parents’ support to children 37 37 27 44 37 32 Increase the number of instructional hours 31 29 18 23 47 56 Provide enough teachers 27 24 39 38 8 12 Improve cooperation between school, parents, and 24 22 10 36 18 21 village government Note: NTT = East Nusa Tenggara. During the 2016 financial year, the share of principals Perceptions, Challenges, and Satisfaction and teachers who received the entire amount of all of Principals and Teachers three allowances (certification allowance, remote area allowance, and additional income allowance for Most of the principals reported that teachers in noncertified teachers) was slightly lower, around 60- their schools have good or very good teaching 80 percent, which may be related to the timing of the skills, but that their schools do not have enough survey. teachers. Table 33 reports on the principals’ perceptions of teachers and students. Between 61 percent (Sintang) The relatively high share of teachers and principals and 87 percent (West Manggarai) of principals consider who received the total amount of their allowance that teachers in their school have good or very good is somewhat unusual, as suboptimal allowance teaching skills. However, on average, approximately 71 delivery in Indonesia is commonly reported. For percent of principals also reported that their schools example, Tomayah et al. (2010) report that about 60 lacked enough teachers, although this varied widely percent of teachers who were eligible for the remote across the districts—45 to 52 percent of principals in area allowance did not receive the full amount of the the NTT districts have this opinion, compared with 68 to allowance, with wide district variation. For all three 86 percent of principals in the West Kalimantan districts. types of allowance, a higher share of teachers (69 to Results from multivariate OLS regressions (Table 58, in 82 percent) received the full amount during the 2016 annex A.) indicate that teacher shortage is more likely financial year compared with the lower share evident to be reported by principals and teachers who work for principals (62 to 76 percent). 32. SCHOOLING CONTEXT Table 34. Challenges Experienced by Teachers (% Teachers) West Kalimantan NTT All areas West East Ketapang Landak Sintang Manggarai Manggarai Challenges experienced by teachers and affecting performance Inadequacy of school infrastructure/facilities 88 85 90 89 86 91 Inadequacy of school infrastructure/facilities discourages 68 66 71 69 67 67 performance Inadequate teaching tools/learning materials 87 83 89 89 86 91 Inadequate teaching tools/learning materials discourages 74 70 74 76 73 73 performance Insufficient salary 77 76 86 76 68 85 Insufficient salary discourages performance 40 38 43 41 35 44 Lack of discipline from students 68 65 74 61 76 70 Lack of discipline from students discourages performance 53 52 52 52 50 63 Lack of attention of students during class 68 69 71 62 71 71 Lack of attention of students during class discourages 57 55 55 58 55 61 performance Lack of interest from students’ parents 58 70 66 52 52 51 Lack of interest from students’ parents discourages 64 63 69 66 55 66 performance Students are often absent 57 60 70 51 51 58 Students are often absent discourages performance 53 48 53 51 56 62 Teacher shortage 50 45 68 68 22 34 Teacher shortage discourages performance 65 61 73 68 51 56 Lack of interest in education from community 41 48 49 34 41 40 Lack of interest in education from community discourages 56 55 64 58 45 54 performance Lack of study and training opportunities 41 49 44 33 39 44 Lack of study and training opportunities discourages 64 68 68 62 60 60 performance Irregular salary 38 54 37 23 27 55 Irregular salary discourages performance 45 41 60 39 43 47 House is far from school 23 19 29 17 26 33 House is far from school discourages performance 48 38 52 57 44 47 Too many other tasks in addition to teaching 21 25 22 19 16 24 Too many other tasks in addition to teaching discourages 45 44 49 45 40 45 performance Irregular payment and lack of transprency of special 14 19 13 14 11 14 allowance (TK) Irregular payment and lack of transprency of special 31 29 28 33 29 34 allowance (TK) discourages performance Irregular payment and lack of transprency of professional 7 10 5 6 5 6 allowance (TP) Irregular payment and lack of transprency of professional 30 36 29 22 13 50 allowance (TP) discourages performance Problems with student’s parents 7 8 6 6 6 8 Problems with student’s parents discourages performance 52 62 53 62 16 52 Disruption/excessive demand from officials outside the 6 9 4 4 4 8 school Disruption/excessive demand from officials outside the 55 56 45 65 42 52 school discourages performance Note: The share of teachers who report that a challenge discourages performance is conditional on having reporting the challenge in question. NTT = East Nusa Tenggara; TK = Tunjangan Khusus (special allowance); TP = Tunjangan Profesi (professional allowance). 33. in schools which have relatively few assigned teachers discipline and attention is a challenge, and 57 percent and relatively more enrolled students. Furthermore, the reported that student absence is a challenge. For OLS results show that schools who suffer from higher roughly half these teachers, student-related challenges teacher absence are more likely to report that teacher discourage performance. Parents’ lack of interest in shortage is a problem. their children’s educational process is a challenge for 58 percent of the teachers, and lack of community Notably, staff shortages are a common problem interest is a challenge for 41 percent. reported by schools in remote areas across Indonesia. For example, World Bank (2008) reports that 93 Overall, teachers are relatively satisfied with percent of remote schools in the sample area claimed the appreciation they have received from the they lacked personnel. However, those results contrast central government for their role. As shown in markedly with the results from the study sample table 35 (annex A), 35 percent of teachers rate their schools, as shown in table 9. The sample schools satisfaction at 6 or 7, on a scale from 1 to 7 (7 being demonstrate a low student-teacher ratio of 16 to 1. the most satisfied). However, notably, 15 percent of teachers are very dissatisfied, rating their satisfaction In general, principals hold a rather favorable at 1. Interestingly, there is little difference between the opinion of the students in their schools. However, districts in teachers’ overall satisfaction with the central there is a clear difference between the share of government’s appreciation of their work. However, principals who consider the discipline and attendance disaggregating between PNS and non-PNS teachers, of their students as good or very good (74 percent) a clear divide emerges. PNS teachers are much more and the share who rate the ability of their students as satisfied with the appreciation shown by the central good or very good (39 percent) (table 33). According government—53 percent of them are highly satisfied, to the principals, the main factors hindering student and only 19 percent have a low or very low satisfaction learning are related to school environment and level. Among non-PNS teachers, 23 and 47 percent parental awareness, namely, lack of school facilities have high and low to very low levels of satisfaction, and infrastructure (66 percent), lack of parental respectively. awareness and support for children in their education (50 percent), lack of available teachers (32 percent), and Teachers appear to be satisfied with the geographical factors (30 percent). Consequently, the performance of the district education offices main suggestions provided by principals on potential in the organization of primary schools. Among measures to improve student learning are related teachers, 38 percent provided a satisfaction rate of to school facilities (67 percent), teacher quality (43 6-7, whereas only 7 percent provided a rating of 1, and percent), parental support for children (37 percent), there is little difference between the five districts of the and the number of instructional hours (31 percent). study (table 35, in annex A). Slightly fewer PNS teachers expressed dissatisfaction than non-PNS teachers—4 The most significant challenge reported by and 9 percent, respectively, gave a satisfaction rate teachers is related to the inadequacy of school of 1. Forty-seven percent of PNS teachers expressed infrastructure and learning facilities. Eighty- a highly satisfied rate, compared with 32 percent of eight percent of the teachers reported this was the non-PNS teachers. Teachers reported that they are case (table 34). For two-thirds of these respondents, similarly satisfied with the performance of the village this challenge discourages teacher performance. The government and community members (as a joint inadequacy of teaching tools was also acknowledged category), who help in the organization of school by 87 percent of teachers, 75 percent of whom find it management, and with the level of appreciation of a challenge that discourages performance. Teachers’ teachers’ roles by community members involved in working conditions are also a challenge, especially the school management. However, there appear to be low salary levels (77 percent), although only two-fifths notable differences between districts in the level of of the teachers find it a challenge important enough satisfaction with community members’ appreciation to discourage teacher performance. Salary irregularity of their roles as teachers. Fewer teachers in Landak affects 38 percent of the teachers in the sample. (about 35 percent, PNS and non-PNS) rated their Teacher shortages affect half the teachers in the sample, satisfaction level at 6-7 on this criterion, compared with two-thirds of whom reported that it discourages their teachers in other districts, nearly half of whom rated performance. Teachers reported several challenges their satisfaction at 6-7. related to students and their behavior. Sixty-eight percent of the teachers reported that students’ lack of 34. SCHOOLING CONTEXT Most teachers were satisfied with their income, In summary, in line with the significant differences with PNS having higher level of satisfaction in salary and allowances, PNS teachers reported compared to non-PNS. Overall, 53 and 20 percent being more satisfied than non-PNS teachers with of PNS and non-PNS teachers, respectively, reported the support from higher authorities and financial a high level of satisfaction (6-7), again with large incentives they receive. It remains to seen whether differences between districts (table 36, in annex A). such differences also lead to differences in performance Among PNS teachers, 38 percent in Landak and 62 quality between PNS and non-PNS teachers. Overall, percent in Sintang expressed high satisfaction. Among principals and teachers overwhelmingly reported that non-PNS teachers, only 10 percent in East Manggarai learning is affected by poor school facilities, which both and 28 percent in Ketapang reported being highly groups recognize as their most important challenge. satisfied with their salary/honorarium. Approximately Principals and teachers also reported lack of parental 13 percent of teachers overall reported a very low interest in their children’s education as a significant level of satisfaction (1). However, this figure comprises challenge to student learning. This could be due to the only 3 percent of PNS teachers, yet 20 percent of lack of relevant information available to parents about non-PNS teachers. Between districts, there are larger their children’s learning progress (such as ability to differences, especially in the share of teachers with a learn, behavior at school, and provision of homework), very low level of satisfaction (1), which ranges from 14 and lack of awareness about how to participate more percent in Ketapang to 38 percent in East Manggarai. actively in their children’s education process (such as reading with their child, helping with homework, When asked about their ideal salary, the majority making sure their child does not work during school of the teachers in the sample schools preferred to hours, and ensuring that children are fed). This study have higher than their current salary. Sixty-three could contribute to effect changes in this important percent of PNS teachers and 6 percent of non-PNS topic. Interestingly, principals and teachers hold teachers reported this to be the case. Twenty percent very different perceptions on student behavior and of the teachers reported that their current salary was discipline. Most of the principals expressed relative satisfactorily ideal, and 20 percent of teachers in the satisfaction with student behavior, whereas most of the NTT districts would ideally prefer a higher salary. The teachers reported it to be a significant challenge. share of PNS teachers who would prefer a higher salary ranges from 5 percent in Sintang to 17 percent in East Manggarai. Among non-PNS teachers, this share ranges from 4 percent in Landak to 21 percent in East Manggarai and West Manggarai. 35. 36. PARENT AND COMMUNITY INVOLMENT IN EDUCATION 04 Parent and Community Involvement in Education Parents Parent Background Most parents are caretakers of students in the study schools, practice the predominant religion in their village, and use a language other than Indonesian at home. Table 37 shows the socioeconomic characteristics of parents of students in the sample schools. The students overwhelmingly (97 percent) reside with their parents. Only 181 caretakers, or 3 percent of the 5,400 caretakers surveyed, are guardians of students. Sixty-three percent of student guardians are the children’s grandparents, and 26 percent are uncles or aunts. Unsurprisingly, the religion of the parents in the sample is by-and-large representative of the predominant religions in the villages in which they reside, with Catholicism being the main religion of the parents in the sample. However, there is a difference between the NTT districts, where Catholicism is the religion of 86 and 99 percent of the parents, and West Kalimantan, where there is also a significant presence of Islam and Protestantism, with between 14 and 36 percent of parents practicing Islam and between 13 and 39 percent practicing Protestantism. Only a minority of the parents use the Indonesian language to communicate with their child at home, this being more common in West Kalimantan than in the NTT districts. In West Kalimantan, between 58 and 82 percent of parents use the Dayak language, and between 7 and 33 percent use the Malay language, as the preferred language of communication with their child. In NTT, 89-93 percent of parents use the Manggarai language at home; the remainder use other local languages. Most of the parents attended primary school as their highest level of education. In the West Kalimantan districts, between 53 and 59 percent of parents attended primary school as their highest level of education; 19 to 21 percent attended junior secondary, and 14 to 16 percent attended senior high school (table 37). In the NTT districts, about 75 percent of parents attended primary school as their highest level of education, 13 to 15 percent attended junior secondary school, and 8 to 10 percent attended senior high school. In Sintang, 11 percent of parents never attended school. Across all five districts, between 7 and 11 percent of parents are unable to read and write (using the Roman alphabet). 37. Table 37. Parents’ Background Information (% Parents) West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Main caretaker Parent 97 97 98 97 97 95 Guardian (wali murid) 3 3 2 3 3 5 Grandparents 63 50 85 57 65 73 Uncle/aunt 26 30 15 27 26 24 Brother/sister 6 5 0 12 4 3 Other family 4 10 0 5 0 0 Other but not family 2 5 0 0 4 0 Demographic characteristics (#) Age of parents (years) 37 37 37 36 39 40 Age of guardian (years) 49 46 54 47 52 54 Average household size (#) 5 5 5 5 6 5 Average # children 3 2 3 2 3 3 Religion Islam 18 36 16 14 14 1 Christian - Protestant 21 13 32 39 0 0 Catholic 61 51 51 47 86 99 Others 0 0 0 0 0 0 Main language used to communicate with child Bahasa Indonesia 7 9 8 9 1 1 Malay 12 33 13 7 0 0 Dayak 54 58 78 82 0 0 Manggarai 24 0 0 0 89 93 Others 3 1 1 2 10 6 Highest education level attended No education 6 5 5 11 2 2 Primary school 60 59 55 53 74 72 Junior secondary school 19 21 23 19 15 13 Senior secondary school 13 14 16 14 8 10 University 2 2 1 3 1 3 Literacy ability Able to read and write Roman alphabet 84 83 87 80 90 86 Able to read and write non-Roman alphabet 2 1 1 4 0 4 Able to read and write multiple alphabets 3 7 1 3 0 1 Unable to read or write 9 7 8 11 8 8 Employment status and sector Worked in past month 98 98 98 98 98 99 Work in agriculture 85 80 92 82 88 90 Work in industry and construction 7 7 4 9 6 4 Work in trade and services 8 12 4 9 6 5 Employment - position Self-employed 18 23 19 17 18 9 Self-employed with unpaid labor 53 31 49 60 64 68 Self-employed with paid labor 2 4 2 2 2 2 Private employee 20 33 28 13 10 16 Freelancer 5 8 1 5 5 3 Unpaid worker 1 1 1 1 1 1 Government employee 1 1 0 2 1 1 38. PARENT AND COMMUNITY INVOLMENT IN EDUCATION West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Household asset ownership House 95 93 97 94 96 98 Land for house 93 92 97 89 95 97 Poultry, livestock, or fish 70 72 78 67 63 69 Sofa 4 10 3 3 3 1 Table 56 60 61 43 77 52 Fridge 12 28 7 13 1 0 TV 49 73 58 57 15 12 Car 3 5 2 4 0 0 Motorcycle 56 84 73 66 10 9 Bicycle 14 25 15 15 3 0 Note:NTT = East Nusa Tenggara. Nearly all the parents work—between 80 and 90 to get to school, on average, and the majority does not percent in agriculture, and the remaining are pay any transportation costs to get there. homogeneously distributed between industry According to their parents, students in the sample and services. Most of the parents work in unpaid self- areas spend a meaningful amount of time on employment—ranging from 31 percent in Ketapang to schooling and learning, attend school nearly 68 percent in East Manggarai (table 37). Across all five every day, and study at home. The parents reported districts, between 9 and 23 percent work in paid self- school attendance as being relatively high—on average, employment. Most of the non-self-employed parents children attend 5.62 of 5.87 school days per week (table are private employees: 28 to 33 percent in the West 38). Approximately 60 percent of the parents (ranging Kalimantan sample districts and 10 to 16 percent in the from 36 percent in West Manggarai to 72 percent in NTT sample districts. Landak) reported that their child studies at home every Most of the parents own their house and the land on day, with the remainder reporting that their child only which it was built, and between 63 and 78 percent sometimes studies. Only 1 to 5 percent of the children own poultry, livestock, or fish. On average, parents never study at home, according to their parents. in the West Kalimantan districts, especially in Ketapang, The parents reported that they are relatively own several household appliances, such as a television involved in supporting their children at home. The (73 percent of parents in Ketapang), motorcycle (84 vast majority, about 80 percent, reported helping their percent), fridge (28 percent), bicycle (25 percent), or sofa child study at home for about 48 minutes per day (on (10 percent). In NTT, the parents in the sample are less average) during the previous week. About 34 percent of likely to own these types of assets—the most commonly the parents reported that their child had received help owned asset is a table (77 percent in West Manggarai and in their studies from someone else during the previous 52 percent in East Manggarai). On average, only 12-15 week. The help was not remunerated and lasted for 33 percent of parents in the NTT districts own a television, minutes daily, on average. Around half the parents in the and about 10 percent own a motorcycle. sample areas reported reading––sometimes or often–– their child’s textbooks. In the West Kalimantan districts, Child Support and Activities at Home approximately 13 percent of the parents reported Most of the students live near the study schools, never reading their child’s textbooks, compared with 24 but they did not attend any form of early to 27 percent in the participating NTT districts. Parents childhood education. However, this varies across in the West Kalimantan districts also ask their children districts, from 68 percent in Ketapang to 95 percent in to study more frequently (more than five days a week), West Manggarai. In Ketapang, 17 percent of the children compared with parents in the NTT districts (around had attended playgroup, and 16 percent had attended four days per week). The overwhelming majority of kindergarten (table 38). In general, students tend to live the parents, from 86 percent in West Manggarai to relatively close to their school, about 600 meters on 96 percent in Sintang and East Manggarai, reported average. It takes the children around 10 to 15 minutes knowing the subjects that their children did not master. 39. Table 38. Child’s Education and Parent Involvement West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Early childhood education attended (% parents reporting) Playgroup/Kelompok Bermain (KB) 7 17 6 3 4 6 Kindergarten (TK) 10 16 7 14 2 1 Raudhatul Athfal (RA) 0 0 0 0 0 0 Did not attend PAUD 83 68 88 83 95 92 Child school attendance in past week Average # school days in past week 5.9 5.8 5.9 5.8 6 6 Average # school days attended in past week 5.6 5.5 5.7 5.6 5.6 5.7 Distance to school Median distance from house to school (km) 0.3 0.3 0.3 0.3 0.5 0.5 Median travel time from house to school (minutes) 10 10 10 5 10 15 Median transportation cost from house to school (Rp) 0 0 0 0 0 0 Study at home (% parents reporting) Child studies at home every day 58 61 72 57 36 59 Child studies at home sometimes 39 37 26 40 58 39 Child never studies at home 3 1 2 3 5 2 Child is helped by caretaker for home study 82 88 81 82 78 76 Child received help from caretaker in past week 74 79 74 77 67 66 Time spent daily by caretaker in past week, if help from 48 45 47 45 55 52 caretaker (minutes) Child was helped by other for home study in past week 34 40 28 29 31 46 Time spent daily by other in past week, if helped by other 33.48 30.22 34.86 30.67 38.38 38 (minutes) Caretaker paid other to help child, if helped by other 0 0 2 2 0 0 Amount paid to other to help child per visit, if paid (Rp) 27,521 83,300 32,650 19,375 1,000 0 Parent awareness of child’s performance at school (% parents reporting) Frequency of reading child’s textbooks Never 16 12 14 12 27 24 Rarely 19 23 19 17 14 19 Sometimes/often 53 52 60 56 43 46 Child has no books 7 6 3 7 11 6 Parent cannot read 6 7 4 8 5 5 Parent knows subjects that the child does not master 93 92 91 96 86 96 Average # days in week parent asks child to study 5 6 5 6 4 5 Note: KB = Kelompok Bermain (playgroup); km = kilometers; NTT = East Nusa Tenggara; PAUD = Pendidikan Anak Usia Dini (early childhood education programs). Child participation in work was not very common hours weekly). Children work in family businesses in in the sample areas, although the majority of the 19 percent of the households in Ketapang, 14 percent children help with household chores. About two- in West Manggarai and East Manggarai, 10 percent in thirds of the parents reported that their child helped Sintang, and 5 percent in Landak. On average, children with household chores during the past month (table 39). spend approximately eight hours per week working Children’s help with household chores is more common in family businesses, with (one-way) travel taking in the NTT districts: 85 to 90 percent of parents reported approximately 20 minutes to undertake this work. that their child helps with chores for approximately five Children working for pay is not very common in the hours weekly, compared with 55 to 72 percent in the sample areas, with the number of paid weekly hours West Kalimantan districts (approximately two to four worked varying from four in Landak to eight in Sintang 40. PARENT AND COMMUNITY INVOLMENT IN EDUCATION Table 39. Child Participation in Paid, Unpaid, and Household Works West Kalimantan NTT All areas West East Ketapang Landak Sintang Manggarai Manggarai Paid work Child has worked for pay in past month 3 8 1 2 4 3 # weekly hours in past month, if worked for pay 8 8 4 8 8 6 Median monthly income from work, if worked for pay (Rp) 30,000 40,000 15,000 30,000 22,000 11,250 One-way travel time to go to work, if worked for pay 16 14 16 13 19 20 (minutes) Unpaid, family work Child has worked for family business in past month 12 19 5 10 14 14 # weekly hours in family business past month, if worked 8 8 7 7 6 11 for family One-way travel time to go to family business, if worked 19 18 18 19 17 25 for family (minutes) Household chores Child has helped with household chores in past month 67 72 55 56 85 90 # weekly hours spent doing household chores, if chores 4 2 4 4 5 5 Note: NTT = East Nusa Tenggara. Table 40. Parents’ Expectations of Child’s Education West Kalimantan NTT All areas West East Ketapang Landak Sintang Manggarai Manggarai Expectations - child’s achievement at school Attend school everyday 40 38 25 44 36 61 Be able to go to the next grade 36 34 45 32 41 28 Get high scores/be class champion/join competitions 47 56 51 43 47 38 Be able to read/write/count 43 42 35 39 57 51 Others 6 16 3 4 3 3 None 1 1 0 2 2 0 Expectations - child’s highest education level Graduate from primary school 1 2 0 1 1 3 Junior secondary school 3 3 2 2 3 4 Senior secondary school 19 17 20 17 16 31 College/university 43 38 42 42 48 46 It is up to the child 16 16 19 14 24 12 As high as possible 18 24 17 25 8 4 Expected actions in case of grade retention Give advice or reprimand the child verbally 90 89 90 89 92 90 Give physical punishment 3 2 2 1 9 4 Give nonphysical punishment 2 3 2 2 3 1 Ask teacher/classroom teacher/principal 6 9 5 4 7 7 Do nothing 6 4 6 8 5 5 Others 3 8 2 2 2 1 Note: NTT = East Nusa Tenggara. 41. and Ketapang. The monthly income received by children that the committees would monitor and provide input who work for pay ranges from IDR 34,500 (US$2.5) in about school operations (including programs, budget Landak to IDR 81,000 (US$5.8) in Ketapang. plans, facility improvement, and teacher training), while formally involving parents and communities in the Parents’ Expectations for Their Children functioning of their children’s schools. Parents in the sample areas expect their children Most of the schools in the sample areas have an to be able to read, write, and count; attend school active school committee, and most of the school every day; and move to the next grade in school. In committees manage one school. Among the 270 West Manggarai, nearly 60 percent of parents reported schools included in the study, 254 have an active school that they expected their children to be able to read, committee, 14 have an inactive committee, and two write, and count (table 40). In Ketapang and Landak, appear to have no current committee (table 41, annex A). the predominant expectation of the parents is that of the 241 schools with an active committee (for which their children would achieve high scores, be the class we have information on their date of establishment), champion, and join competitions. In Sintang, parents 20 percent were established in 2016–17, 63 percent in expect daily school attendance from their children, in 2010–15, 14 percent in 2005–09, and 4 percent before addition to high performance. Across all five districts, 2005. The overwhelming majority of the active school between 28 and 45 percent of parents expect that committees (93 percent) manage only one school. For their children will go to the next grade. the functioning of school committees, 15 percent have articles of association or bylaws; 12 percent received All the parents in the survey sample expect their funds for activities from the school during academic year children to progress beyond graduating from 2015/16; and 2 percent (five committees) were provided primary school. Approximately 43 percent of parents with an office space by the school. expect their children to reach university. On average, approximately 20 percent of parents expect their children to reach senior high school, ranging from 16 percent in Committee Management West Manggarai to 31 percent in East Manggarai. Between 12 and 24 percent of parents leave expectations of their Most of the school committee respondents had children’s highest education level to their children. been serving for five years, with the majority of them having graduated from senior secondary The survey asked parents what they would do if school. Most of the respondents were chairpersons, and their child were to be held back in a grade. About a few were vice-chairpersons, secretaries, members, and 90 percent of the parents reported that they would give treasurers. On average, the respondents had occupied advice to or reprimand their child verbally, as required. their positions in the school committee for approximately In contrast, 6 percent of the parents reported they would five years (table 42). The highest education levels of do nothing or ask the teaching personnel about it. Nine the committee respondents were senior secondary and 4 percent of the parents in West Manggarai and education (36 percent), junior secondary education (27 East Manggarai, respectively, reported that they would percent), primary education (24 percent), and university- use physical punishment with their children. In the other level education (6 percent). About 9 percent of the districts, this proportion was lower—around 2 percent. respondents were also administrators or members of other school committees. School Committees Most of the school committee members were democratically selected. The school committee Committee Background and Establishment questionnaire asked about the composition of the school committees and how committee members Since 2002, school committees have been formally were selected. According to Chen (2011), school established as the institution representing committees are required to be chaired by a community communities at the school level. This has been a representative from outside the school and should consequence of Indonesian Ministry of Education Decree have at least nine members elected from among No.044/U/2002 on the Education Board and School parents, community leaders, education professionals, Committees and following the principles of school- the private sector, education associations, teachers, based management. The objective of this decree was nongovernmental organizations, and village officials. that school committees would support improvements in Overall, committee administrators in the sample schools educational service delivery. In particular, it was expected 42. PARENT AND COMMUNITY INVOLMENT IN EDUCATION Table 42. School Committee Management (% Committee Respondents) All West Kalimantan NTT Areas Ketapang Landak Sintang West Manggarai East Manggarai Respondent characteristics Type of respondent Chairperson 86 79 94 89 94 73 Vice chairperson 6 11 2 4 0 12 Secretary 4 5 2 1 6 9 Treasurer 1 0 2 1 0 0 Member 4 5 0 5 0 6 Duration of respondent in current role (years) 5 3 5 3 8 6 Education level of respondent Did not graduate primary school 6 7 2 6 11 0 Primary school 24 30 19 26 23 21 Junior secondary school 27 34 25 24 29 24 Senior secondary school 36 21 52 35 34 39 University 6 5 0 7 3 12 Package a/b/c 2 2 2 1 0 3 Respondent is member/administrator of 9 16 8 4 9 12 committees for other schools Committee management establishment Selection of committee administrators Selection through meeting 88 88 81 87 91 97 Appointed by school (principal and teachers) 10 11 19 11 6 0 Appointed by previous member/ 1 0 0 1 0 3 administrators Other 1 2 0 1 3 0 Who attended selection meeting Principal 93 96 87 96 94 91 Teachers 95 96 87 97 100 94 Parents 88 92 97 82 90 81 Village officials/public figure 72 78 56 77 81 63 Previous committee members 49 33 38 72 45 38 Selection of committee chairperson during meeting: Deliberation without voting (consensus) 24 29 31 27 16 9 Voting 75 71 64 73 84 91 Appointed by principal 1 0 5 0 0 0 Selection of other committee administrators during meeting: Deliberation without voting (consensus) 29 31 26 39 16 25 Voting 52 51 46 40 65 72 Appointed by principal 3 0 13 3 0 0 Appointed by chairperson 9 12 3 11 13 3 Appointed by others 1 0 3 1 3 0 Only chairperson in the committee structure 5 6 10 6 3 0 were predominantly selected through a committee meetings were reported to have been attended by meeting process (88 percent of the committees in the school principals (93 percent of the committees), sample); a small portion were appointed directly by the teachers (95 percent), parents (88 percent), village school (10 percent). Committee administrator selection officials and leaders (72 percent), and previous 43. All West Kalimantan NTT Areas Ketapang Landak Sintang West Manggarai East Manggarai Committee’s administrators/members 18 9 31 22 11 9 receive salary Median monthly salary (Rp) 100,000 100,000 100,000 100,000 58,334 50,000 Incentive sources: Parents 4 0 0 0 25 33 School budget 9 0 7 11 25 0 School Operations Fund 82 80 93 78 75 67 Other sources 4 20 0 0 25 0 Do not know 4 0 0 11 0 0 Note: NTT = East Nusa Tenggara. committee members (49 percent). During these parental involvement in school affairs in the study selection meetings, the chairperson was selected by areas tends to be limited to interactions with teachers vote or consensus, whereas other administrators or principals on issues related to their own children, were selected by vote, designated by consensus, as reflected in similar findings by other studies.50 or appointed by the chairperson. Five percent of Among those who visited their child’s school during the committees in the sample schools only have a academic year 2015/16, 17 percent discussed their chairperson, without other committee members. child’s exam results with the principal, 23 percent Interestingly, Pradhan et al. (2014) test the effectiveness with their child’s classroom teacher, and 10 percent of various school committee institutional reforms and with another teacher. Fifteen, 12, and 6 percent of find that the election of committee members does not the parents discussed their child’s overall learning lead to improvements in student learning, although it development with the principal, classroom teacher, does increase community awareness. These findings or another teacher, respectively. Approximately 11 suggest that community awareness alone does not percent of the parents reported having discussed their provide committees with the legitimacy and power child’s discipline and/or attendance at school with the needed to improve educational service delivery. principal, 9 percent with the classroom teacher, and 5 percent with another teacher. In the NTT districts, Eighteen percent of the committees reported the parents discussed—with the principal (20 to 33 that the administrator and chairperson receive a percent), the classroom teacher (13 percent), and other salary for their role. In West Kalimantan, the median teachers (10 to 13 percent)—how they or the school monthly salary for the administrator or chairperson committee could contribute to their child’s education. is IDR 100,000 (US$7); in the NTT districts, it is much lower, at IDR 50,000 (US$3.5) to IDR 60,000 (US$4.3). The majority of the school committees reported The incentives for the committee administrator and having at least one annual meeting to discuss chairperson are from the School Operations Fund for various topics. Nearly four-fifths of the school 82 percent of the school committees that offer a salary committees held at least one meeting with the principal, to their administrator, and from the school budget for 9 parents, or the principal and parents together during percent of the school committees. In the NTT districts, academic year 2015/16. During academic year 2015/16, funds for administrator and/or chairperson incentives 35 percent of the committees reported having only met come from parents in two schools. with the principal, and among these committees, 48 percent had met in the previous month. The committee Involvement of Parents and respondents reported that they covered topics such Committees at School as the preparation of student evaluations (84 percent of the committees), suggestions and complaints from Parental and Committee Involvement at parents (83 percent), school budget and financial School resources (77 percent), student learning outcomes (76 percent), student discipline and behavior (76 percent), Parents in study schools were actively involved and teacher discipline and behavior (68 percent) (table in school affairs. More than four-fifths of the parents in the sample survey visited their child’s school during academic year 2015/16 (table 43). However, 50 Chen 2011; Vernez, Karam, and Marshall 2012. 44. PARENT AND COMMUNITY INVOLMENT IN EDUCATION Table 43. Parents’ Involvement in School, 2015/16 (% Parents) West Kalimantan NTT All areas West East Ketapang Landak Sintang Manggarai Manggarai Parents went to school in 2015/16 82 82 80 82 85 82 Discussion with principal, if went to school Exam results of child 17 19 13 23 12 12 Child’s overall learning development 15 13 10 24 10 9 Child’s discipline/attendance at school 11 9 8 15 7 10 Teacher performance/teaching quality in school 5 4 4 7 5 3 Committee/parents’ contribution 13 5 2 13 20 33 Others 3 2 4 3 4 4 Discussion with classroom teacher, if went to school Exam results of child 23 31 17 27 14 14 Child’s overall learning development 12 11 8 19 12 6 Child’s discipline/attendance at school 9 8 6 13 7 7 Teacher performance/teaching quality in school 3 2 2 3 4 0 Committee/parents’ contribution 7 2 1 7 13 13 Others 1 1 1 1 2 2 Discussion with other teacher, if went to school Exam results of child 10 13 4 13 7 9 Child’s overall learning development 6 6 2 9 7 4 Child’s discipline/attendance at school 5 4 2 8 5 4 Teacher performance/teaching quality in school 2 2 1 3 4 1 Committee/parents’ contribution 5 1 0 5 13 10 Others 1 1 1 1 2 1 Note: NTT = East Nusa Tenggara. 44, in annex A). These meetings occurred exclusively In short, the parents and school committees at the initiative of the principal in 40 percent of the appear to be involved in their children’s school committees. affairs, based on their self-reports. This finding differs from Vernez, Karam, and Marshall (2012), Almost all school committees provided suggestions who find minimal involvement of school committees and feedback to the schools, with the majority of and parents in school affairs and consider that “both them being implemented by the schools. Twenty-one expressed an attitude of non-interference with school percent of the school committees held internal meetings matters and deference to school staff.” Furthermore, during academic year 2015/16, and 91 percent of the Vernez, Karam, and Marshall (2012) do not find that the internal meetings generated suggestions or feedback for committees were actively involved in school decision the schools. The suggestions and feedback generated by making and activities. In their study, focus groups internal committee meetings predominantly concerned suggested that the figure of 44 percent of principals who rehabilitation of infrastructure and furniture (63 percent), reported school committee participation in decision need for improvement in teacher and/or student making was overestimated. In the Vernez, Karam, and discipline (46 percent), teaching and learning processes Marshall (2012) sample of 400 public primary schools (29 percent), and need for improvement in teacher across Indonesia, the principals solely considered quality (17 percent). of those committees that submitted school committees as intermediary platforms for suggestions to their schools, 81 percent reported that informing parents of school decisions. the school had implemented some of their suggestions, in particular, those concerning need for improvements in Parent Satisfaction the discipline of teachers and/or students, rehabilitation of school infrastructure (33 percent), and teaching and The majority of the parents reported being learning processes (23 percent). satisfied with the quality of education offered in 45. their village. Eighteen percent rated their satisfaction Compared with the aforementioned report on overall as very good (satisfaction rate of 7, on a scale from 1 parent satisfaction with education,51 there appears to to 7), and 65 percent as good (satisfaction rate of 4, 5, be less parent satisfaction in the sample areas when or 6) (figure 8). The quality of education at their child’s rating the learning outcomes of their children in math school (during academic year 2015/16) was rated as and Indonesian—approximately 24 percent of the very good by 10 percent of the parents, and as good parents rated the outcomes as bad, and 5 percent by 79 percent of the parents. Compared with academic rated them as very bad. year 2014/15, the quality of education in 2015/16 was considered better or similar by 26 and 66 percent of Committee Satisfaction the parents, respectively. According to Chen (2011), The school committees in the study areas were parents’ paradoxical satisfaction with the quality of relatively satisfied with educational and school education at their child’s school, in light of their limited quality. In particular, the school committees were interaction and involvement with the school, casts highly satisfied with community and parental support doubt on whether increasing school accountability to to schools and the quality and behavior of teachers parents in Indonesia will be effective, given the strong (figure 9). However, their satisfaction toward average sense of community and related tendency of people student learning outcomes during 2015/16 was more not to complain openly or express dissatisfaction. mitigated—44 percent found it satisfying, and 45 Parents’ satisfaction with their child’s classroom percent found it unsatisfying. Nearly half of the school teacher is also relatively high, with 21 percent committees found that support from the district highly satisfied and 53 percent satisfied. However, and subdistrict education offices was unsatisfying, about 20 percent responded that they did not know and nearly three-fourths were dissatisfied about the whether they were satisfied with their child’s classroom physical condition of school facilities, mirroring the teacher; therefore, they could not give an answer. dissatisfaction of principals and teachers. Figure 8. Parent Satisfaction with Education Quality and Learning Outcomes Very Good Good Bad Very Bad Do not know Figure 9. School Committee Satisfaction with Education and School Quality District/subdistrict education office support to school Very Good Good Bad Very Bad Do not know 51 Chen 2011. 46. 05 TEACHER ABSENCE Teacher Absence Definitions and Statistics Definition and Measurement This report presents measures of teacher absence from class and teacher absence from school. It reports teacher absence from class— that is, the share of classes observed without teachers. Students being left without a teacher is arguably the biggest problem facing Indonesian schools. Additionally, several other measures of teacher absence are reported, to provide a more complete picture of teacher absence in line with previous studies on this subject.52 Teacher absence from school is defined as the number of teachers who were not at school on the day of the visit. Teacher absence from teaching is defined as the number of teachers who were not in the classroom, although they were present at the school.53 Class Absence, or Classes Observed without Teachers During unannounced visits to the sample schools, enumerators directly observed 1,705 classes, of which nearly a quarter were without a teacher. Table 45 shows that 398 classes (23 percent on average) were observed without teachers. This share varies widely across districts, from 14 percent of the classes in Sintang to 32 percent in Ketapang. However, the absent teacher physically returned to the classroom before the end of the enumerator’s observation in 327 classes (19 percent of all observed classes). Previous studies of teacher absence in Indonesia have found relatively high rates of teacher absence, despite a slight improvement in recent years.54 For instance, Usman, Akhmadi, and Suryadarma (2004) found that almost one in five (19 percent) teachers in Indonesian public primary schools were absent from classrooms. However, Toyamah et al. (2010) subsequently found a reduced overall teacher absence rate of 14 percent. In particular, lower teacher absence was found to be directly related to more regular supervision of schools, higher salaries, and teachers’ overall sense of improved welfare. In comparison, this report demonstrates that the teacher 52 ACDP 2014. 53 To allow comparisons of teacher absence rates with previous studies, all teacher absence numbers are expressed as a proportion of all teachers who were reported scheduled to be teaching during the period of observation. In the sample, based on principal reports, 1,687 teachers were reported scheduled to teach during the observation, which is close to the number of classes observed on the day of the survey. 54 ACDP 2014; Chaudhury et al. 2006. 47. Table 45. Classes Observed with No Teacher West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Classes observed without teacher # classes observed 1,705 367 301 545 249 243 Classes observed without teacher (#) 398 118 81 79 65 55 Classes observed without teacher (%) 23 32 27 14 26 23 Classes without teacher, returned before end observation (#) 327 87 75 67 58 40 Classes without teacher, returned before end observation (%) 19 24 25 12 23 16 Students’ activities during observation Activities of students in classes with teacher Learning session in class 88 89 85 85 92 95 Group discussion 2 2 3 1 0 1 Individual work 9 7 11 12 5 4 Exam/test 1 1 0 1 2 1 No clear structured activities 1 1 1 1 1 0 Activities of students in classes without teacher Learning session in class 3 3 1 1 3 11 Group discussion 6 4 6 4 6 11 Individual work 48 51 59 47 32 45 Exam/test 1 1 0 0 0 2 No clear structured activities 41 41 30 46 58 31 All students were absent 2 1 4 3 0 0 Note: NTT = East Nusa Tenggara. absence rate remains at 23 percent in remote areas. In Teacher Absence from School another related study, UNICEF (2012), which focuses on the province of Papua, found a rate of teacher absence On the day of the survey, 2,210 teachers (and from school of 37 percent, and the rate was nearly 50 principals) were scheduled to work; 421 of those percent in the highland districts. Most recently, ACDP scheduled to be teaching at the time of the (2014) found a national rate of teacher absence from observation were absent from school (table 46). school of 10 percent, with the rates in remote areas This finding is similar to the finding of ACDP (2014) of hovering at close to 20 percent, and a teacher absence approximately 20 percent teacher absence from school rate from class of approximately 13 percent. in remote areas. Absent teachers were absent for eight days on average since their last attendance—ranging In 85 to 90 percent of the classes with a teacher from three days in Sintang to 11 days in Landak. Less present, students were involved in a learning than half of the teachers who were found to be absent session. In these classes, between 4 percent (East had already been absent for more than two days. Manggarai) and 12 percent (Landak) of the students were working individually (table 45). In classes without Among those teachers who were absent from a teacher, students were observed to be engaged in school when they were scheduled to be present, individual work in 48 percent of the classes; students they were reported as going on assignments, were not involved in clearly structured activities in 41 being sick or on leave, or having unknown percent of the classes. In approximately 2 percent of the reasons. About 30 percent were reported to be classes observed without a teacher, all the students had working on school-related assignments (table 46). In left the school before the enumerator arrived (ranging the West Kalimantan districts, about a quarter of the from none in the NTT districts to 4 percent of the classes absent teachers were absent for this reason. In the NTT observed without teachers in Landak). At the end of the districts, this share was higher (40 percent). The second enumerators’ visits, 4 percent of all the observed classes most significant reasons for teacher absence from (71 classes) were still unattended by any teacher. school were sickness (14 percent) and other reasons (14 percent). On average, there was no known reason 48. TEACHER ABSENCE Table 46. Teacher Absence from School West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai # Teachers Total # teachers listed in schools (including 2,293 508 370 700 385 330 principals) # teachers scheduled to be present at school 2,210 466 365 677 376 326 # teachers scheduled to be teaching 1,687 364 289 537 251 246 Teacher absence from school # teachers absent from school 421 91 105 82 77 66 Teacher absence from school (% teachers 25 25 36 15 31 27 scheduled to be teaching) Duration of teacher absence since last attendance Mean # days of absence from school 8 8 11 3 10 8 Median # days of absence from school since 2 2 1 1 2 2 last attendance Teacher absence is justified in writing to 75 79 84 85 64 58 principal (% absent teachers) Reasons for teacher absence from school (% absent teachers) Working on school-related assignments 30 23 26 24 42 36 Sick 14 9 12 12 18 18 Taking care of sick family member 10 15 10 12 7 5 Working on non-school-related assignments 9 14 8 5 4 12 Going to college/further education 5 5 3 11 1 2 Late arrival 5 3 3 6 12 5 Scheduled teaching hour is not yet started 3 2 5 7 0 2 Early leave 1 1 0 0 4 0 Scheduled hour is already finished 0 1 0 0 0 0 On leave 3 4 3 1 1 5 Others 14 18 17 12 9 12 Do not know 7 3 14 9 1 5 Location of absent teachers Same village as school location 25 23 23 21 27 32 Different village within the subdistrict 14 16 18 13 14 8 Subdistrict where district capital is located 16 16 20 22 16 3 Different subdistrict within the district 27 26 16 26 32 38 Different district within the province 8 3 8 9 8 14 Other province 2 7 1 1 1 0 Other country 1 2 0 0 0 2 Do not know 7 5 14 9 1 5 for teacher absence for 7 percent of the teachers, meetings or participating in training sessions (Usman, with Landak recording a survey high of 14 percent. Akhmadi, and Suryadarma 2004). In ACDP (2014), the The principal had received a written justification for main reason given for absence is official duties outside three-quarters of the cases of teacher absence. In the school (26 percent). Significantly, the results show comparison, in 2003, another study found that 45 that absent teachers were mainly resident in another percent of absent teachers had been absent without subdistrict other than the subdistrict where the school any known reason; 36 percent had been sick or on or district capital is located (27 percent) or in the same official leave; and the remaining 19 percent had been village as the school (25 percent). on official duties outside the school, such as attending 49. Teacher Absence from Teaching Determinants of Teacher Absence Among the 25 percent of teachers absent from This subsection provides some additional insights classrooms, 5 percent were replaced by other into the teacher and school characteristics that teachers who were not scheduled to teach the are associated with teacher absence. It presents observed classes. In focusing on teachers who were results from simple ordinary least squares (OLS) scheduled to teach, we found on average a teacher regressions of a dummy (equal to 1 for teachers absence rate (from teaching) of 25 percent, ranging absent from teaching) on a set of teacher and school from 16 percent in Sintang to 34 percent in West characteristics derived from the descriptive analysis Manggarai. Only 8 percent of the teachers were found presented in the previous sections. to be in school but not observed teaching—ranging As explanatory variables, the analysis uses from 2 percent in East Manggarai to 16 percent in several teacher and principal characteristics. West Manggarai. According to the principals’ reports, These include dummy variables for principals, female 1,687 teachers were scheduled to teach on the day teachers/principals, PNS teachers, certified teachers, of the observation. The teachers who were observed having at least a bachelor’s degree, remote area teaching, but who were not scheduled to do so, were allowance receipt, additional income allowance for most likely substitute teachers replacing absent noncertified teachers receipt, having an extra job, high teachers. To identify the number of classes that were satisfaction (> 4) with their salary/honorarium, and left without (substitute) teachers, we compared the having been evaluated by the principal in 2015/16. The number of teachers scheduled to teach and found last two variables are used only in the OLS regressions teaching with the number of teachers found teaching, run using teachers as the subsample, as this information regardless of whether they were scheduled to teach. is only pertinent for teachers. The analysis also uses Among those who were scheduled to teach, 75 percent the number of years of seniority at the current school were indeed observed in class teaching. However, as as a teacher characteristic. seen in table 47, there were 1,354 teachers observed in Additional variables include school characteristics. class teaching on the day of the survey, which amounts The analysis includes dummy variables for schools with to 80 percent of those who were scheduled to teach. toilets for teachers, schools with electricity, and schools This finding implies that, among the teachers who were that reported having been visited by school supervisors scheduled to teach but not observed teaching by the during 2015/16. Additional continuous variables include enumerators (25 percent), 5 percent were replaced school distance to the district education office and the by another teacher, and 20 percent of the (scheduled) shares of PNS teachers, teachers with a bachelor’s classes remained without a teacher. degree, teachers who have been in their current school The teachers who were absent from teaching were for more than five years, certified teachers, teachers predominantly also absent from school. Two-thirds receiving a remote area allowance, and teachers with of the teachers who were observed to be absent from high satisfaction (> 4) with their salary/honorarium. We teaching were reported to be involved in school-related run additional regressions on schools with an active administrative activities; the remainder were on a break committee as a subsample, include dummies for schools or involved in non-school-related activities. where the committee chairperson was selected by vote In summary, teacher absence is rather high in during meetings attended by parents, and dummies the study areas. The enumerators observed that for committees that reportedly held separate meetings approximately one in four classes was without a teacher. with the principal and parents during 2015/16. All the One-fourth of the teachers who were scheduled to be regressions include district fixed effects to control for teaching during the observation were absent from differences in local education policy. school, and one-third of those teachers were reported Principals, PNS teachers, male teachers, and to be absent because of school-related assignments. those who have worked at the school for at Approximately 25 percent of the teachers were absent least five years are associated with absence from teaching, and 20 percent of the scheduled classes from teaching. The results of the OLS regressions were without a teacher. These data contrast with the are presented in table 48. The regression results are self-reported teacher and principal information on merely correlations and cannot be interpreted as causal teaching, which suggests that teachers spend a lot of effects. For example, it is remarkable, and in line with time on teaching and other activities. the findings of UNICEF (2012), that the survey findings 50. TEACHER ABSENCE Table 47. Teacher Absence from Teaching West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Teachers found in class teaching # Teachers found in class teaching 1,354 260 219 472 198 205 Teachers found in class teaching (% teachers scheduled 80 71 76 88 79 83 to be teaching) Absence from teaching - teachers scheduled to teach # Teachers scheduled to be teaching 1,687 364 289 537 251 246 # Teachers found in class teaching while scheduled to 1,258 247 200 453 166 192 teach Teachers found in class teaching (% teachers scheduled 75 68 69 84 66 78 to be teaching) # Teachers absent from teaching while scheduled to 429 117 89 84 85 54 teach Teachers absent from teaching (% teachers scheduled to 25 32 31 16 34 22 be teaching) # Teachers absent from school while scheduled to teach 288 66 76 53 45 48 Teachers absent from school (% teachers scheduled to 17 18 26 10 18 20 be teaching) # Teachers found in school but not in class while 141 51 13 31 40 6 scheduled to teach Teachers found in school but not in class (% teachers 8 14 4 6 16 2 scheduled to be teaching) Activities of teachers absent from teaching (% teachers absent from teaching) School-related administrative activities 66 62 67 69 66 72 Break/non-school-related activities 34 38 33 31 34 28 Note:NTT = East Nusa Tenggara show that being a principal is positively and significantly Democratically elected school committee chair is associated with absence from teaching duties, correlated with better teacher presence in class. controlling for all other characteristics. Interestingly, Selection of the school committee chairperson—by the female teachers and teachers regularly evaluated by vote of attendees at selection meetings,55 as opposed the principal are significantly less likely to be absent to being appointed by the principal or selected through from teaching duties; this finding is robust across the consensus—is significantly negatively associated with different specifications. Other individual characteristics teacher absence from teaching, conditional on being are not significantly associated with absence from scheduled to teach. Another study in Indonesia showed teaching. An exception is that teachers who have been that democratically elected school committee increased evaluated by the principal are less likely to be absent, community awareness (Pradhan et al. 2014). It is as seen in columns (3) and (4), which present the plausible that this increased awareness affects teachers’ results (in the subsample of teachers) of regressions behavior, but this survey did not collect additional data. of teacher absence from teaching, conditional on being scheduled to teach. 55 Attendees at selection meetings may include the principal, teachers, parents, village officials, community members, and previous committee members. 51. Table 48. OLS Regressions of Teacher Absence on Selected Teacher and School Characteristics (1) (2) (3) (4) Principal 0.272*** 0.276*** - - (0.055) (0.054) Female -0.052** -0.051** -0.050** -0.049** (0.023) (0.023) (0.024) (0.024) PNS 0.042 0.040 0.034 0.033 (0.027) (0.026) (0.029) (0.029) Certified -0.001 0.001 -0 0.001 (0.039) (0.039) (0.041) (0.041) Bachelor or above -0.019 -0.016 -0.031 -0.027 (0.024) (0.024) (0.025) (0.025) # Years in current school -0 -0 -0 -0 (0.002) (0.002) (0.002) (0.002) Receive special allowance -0.044 -0.044 -0.029 -0.029 (0.041) (0.041) (0.043) (0.044) Receive tamsil -0.022 -0.019 -0.011 -0.009 (0.025) (0.025) (0.025) (0.026) Extra job -0.021 -0.022 -0.016 -0.018 (0.023) (0.023) (0.024) (0.024) Born in the same district as school -0.026 -0.020 -0.030 -0.023 (0.043) (0.043) (0.046) (0.046) Salary satisfaction is high - - -0.002 -0.002 (0.025) (0.025) Evaluated by principal - - -0.056** -0.051* (0.028) (0.029) Distance to district education office -0 -0 -0 -0 (0) (0) (0) (0) Toilet is available for teachers 0.001 0.007 0.006 0.010 (0.027) (0.027) (0.027) (0.027) Electricity is available at school -0.023 -0.018 -0.028 -0.024 (0.027) (0.027) (0.027) (0.027) Share of PNS teachers -0.183* -0.167 -0.189* -0.187* (0.099) (0.102) (0.106) (0.110) Share of teachers with minimum bachelor’s degree 0.081 0.075 0.072 0.068 (0.073) (0.073) (0.073) (0.073) Share of teachers with minimum 5 years at school 0.147 0.133 0.183* 0.172* (0.097) (0.094) (0.098) (0.097) Share of certified teachers 0.073 0.065 0.097 0.093 (0.103) (0.102) (0.105) (0.105) Share of teachers receiving special allowance -0.037 -0.033 -0.060 -0.057 (0.053) (0.053) (0.057) (0.057) Share of teachers highly satisfied with their salary 0.013 0.010 0.018 0.018 (0.072) (0.070) (0.080) (0.079) Supervisor visited school -0.073 -0.057 -0.052 -0.040 (0.046) (0.045) (0.047) (0.047) 52. TEACHER ABSENCE (1) (2) (3) (4) Committee chairperson is selected by voting - -0.069** - -0.053* (0.027) (0.028) Committee had meetings with parents and principal - 0.024 - 0.022 (0.029) (0.031) Constant 0.360*** 0.355*** 0.361*** 0.353*** (0.101) (0.099) (0.105) (0.103) Observations 1,578 1,569 1,440 1,432 R-squared 0.087 0.093 0.054 0.056 Note: Standard errors are in parentheses. All regressions include district dummies. All respondents from the TAS instrument are considered in columns (1) and (2), whereas only teachers are considered in columns (3) and (4). Only schools with an active coommittee are considered in columns (2) and (4); all school are considered in the remaining columns. PNS = civil servants; TAS = Teacher Absence Survey. *** p < 0.01, ** p < 0.05, * p < 0.1. 53. 54. STUDENT ABSENCE AND LEARNING OUTCOMES 06 Student Absence and Learning Outcomes Student Absence There is a discrepancy between the official student absences and enumerators’ observations. Only 8 percent of the students were absent when the enumerator visited, according to the schools’ attendance books (table 49, in annex A). However, the number of students found in class during the observations shows a higher student absence rate of 14 percent, on average. This difference was particularly significant in Ketapang, where 9 percent of registered students were officially absent, but 24 percent of them were not present in class. Likewise, the rate of student presence differs from the official figures and parent information, suggesting that student absence is a substantial problem that may affect teacher performance and student learning. Student absence rates decrease with each increase in grade. In the NTT districts, and East Manggarai in particular, there were fewer differences between official and observed student absence rates than in the West Kalimantan districts (table 49, in annex A). In approximately one-third of all the classes observed, all the students were present on the day of the enumerators’ visit. In grade one, full student presence ranged from 15 percent of the observed classes in West Manggarai to 34 percent in East Manggarai. This figure is rather low and again contrasts with official student attendance records and parents’ reports that their children attended nearly all scheduled school days. There are more registered male students than female students across all grades, and male students have a slightly higher absence rate than female students. Male and female absence rates are 9 and 7 percent overall, on average (table 57, in annex A). Student absence rates decrease slightly as grades increase—for female students, from 11 percent in grade one to 5 percent in grade six, and for male students, from 12 to 7 percent. Outcome Levels The test results indicate whether students had mastered the grade- level competencies they were supposed to have acquired when they graduated to their current grade level. First, student test results are expressed as percentages of correct answers, ranging from zero to 100, as seen in table 50. The tests were multiple-choice, with three or four possible answers for each question. Therefore, a student who answered each question randomly has an expected score of 25 to 33 percent. 56 Mullis et al. 2016. 55. Table 50. Student Test Scores: Descriptive Statistics West Kalimantan NTT Grade Subject All Areas Ketapang Landak Sintang West Manggarai East Manggarai Bahasa Mean 32.72 38.81 26.23 36.14 26.98 30.3 SD 22.65 22.62 20.16 25.4 18.64 18.87 1 Median 30.43 34.78 26.09 34.78 26.09 26.09 Min 0 0 0 0 0 0 Max 100 100 95.65 100 95.65 95.65 Mean 44 49.24 36.01 44.88 42.01 45.86 SD 22.59 23.18 21.28 25.45 18.85 18.11 2 Median 43.48 47.83 34.78 43.48 39.13 43.48 Min 0 0 0 0 0 0 Max 100 100 100 100 100 95.65 Mean 28.26 34.16 25.33 31.58 19.87 25.92 SD 15.84 14.17 13.89 15.31 15.61 15.91 3 Median 26.09 34.78 26.09 30.43 17.39 26.09 Min 0 0 0 0 0 0 Max 78.26 69.57 69.57 78.26 65.22 69.57 Mean 35.32 39.96 32.91 39.02 27.99 33.38 SD 14.73 12.67 12.86 13.6 15.66 15.7 4 Median 34.78 39.13 34.78 39.13 30.43 34.78 Min 0 4.35 0 4.35 0 0 Max 78.26 78.26 78.26 73.91 69.57 78.26 Mean 35.3 38.05 32.19 38.86 28.8 34.51 SD 14.24 13.19 12.53 14.05 13.59 14.86 5 Median 34.78 39.13 30.43 39.13 26.09 34.78 Min 0 4.35 4.35 0 0 0 Max 78.26 73.91 73.91 78.26 73.91 73.91 Math Mean 33.5 44.2 26.61 39.07 22.52 26.5 SD 24.33 22.81 22.53 25.85 18.68 20.55 1 Median 33.33 46.67 23.33 40 16.67 23.33 Min 0 0 0 0 0 0 Max 100 100 100 100 100 96.67 Mean 43.31 51.65 35.62 45.93 35.95 43.03 SD 23.42 21.92 19.59 25.13 22.02 22.23 2 Median 36.67 50 33.33 43.33 30 36.67 Min 0 0 0 0 0 0 Max 100 100 100 100 100 100 Mean 30.38 36.55 27.1 33.21 22.66 28.37 SD 16.83 14.07 15.26 16.03 18.16 17.23 3 Median 30 36.67 26.67 33.33 20 26.67 Min 0 0 0 0 0 0 Max 73.33 70 66.67 73.33 73.33 70 Mean 29.72 32.62 28.23 31.53 25.76 28.82 SD 10.79 10.27 9.21 9.6 12.41 11.55 4 Median 30 33.33 26.67 30 26.67 30 Min 0 0 0 0 0 0 Max 70 66.67 56.67 70 60 63.33 56. STUDENT ABSENCE AND LEARNING OUTCOMES West Kalimantan NTT Grade Subject All Areas Ketapang Landak Sintang West Manggarai East Manggarai Math Mean 30.7 32.7 30.36 32.67 27.1 28.34 SD 10.63 10.36 9.31 10.38 10.62 11.17 5 Median 30 33.33 30 33.33 26.67 30 Min 0 0 0 3.33 0 0 Max 66.67 63.33 63.33 63.33 66.67 63.33 Note:NTT = East Nusa Tenggara; SD = standard deviation. Table 51. Student Test Scores in Indonesian and Math, by Parent Education Grade 1 Grade 2 Grade 3 Grade 4 Grade 5 Indonesian Overall 33.49 44.83 28.62 34.45 32.58 Parents have no education 28.91 41.180 27.54 31.94 32.65 Parents have primary education 31.25 42.96 27.51 33.3 31.41 Parents have junior secondary education 37.5 47.88 30.78 36.96 35.35 Parents have senior secondary education 36.72 49.3 30.63 37.93 35.05 Parents have university education 50.38 56.37 36.41 42.43 39.35 Math Overall 34.41 44.23 30.82 29.55 32.03 Parents have no education 33.15 41.74 29.96 28 32.02 Parents have primary education 32 42.52 29.94 29.1 31.43 Parents have junior secondary education 39.12 47.7 33.12 30.6 32.99 Parents have senior secondary education 37.54 48.27 32.33 30.7 33.78 Parents have university education 46.78 50.48 35.02 34.53 35.94 Overall, the students have slightly higher scores grade three, the scores in Indonesian are quite low, as in Indonesian than in math, and female students low as the benchmark score of 25 percent for random averaged higher scores than male students. This guessing, varying from 20 in West Manggarai to 34 in difference is consistent with the PISA scores (OECD Ketapang. From grade three onward, the maximum 2016). Overall, there is a positive correlation between scores obtained are also far below 100. In grades four parent education levels and student scores (table and five, scores range from approximately 28 in West 51). Table 56 (in annex A), shows student test scores Manggarai to nearly 40 in Ketapang, with lower standard by grade and gender. Female students have higher deviations, suggesting more homogeneity in students’ average scores than male students in Indonesian and ability in Indonesian in these grades compared with math and across all grades, which is consistent with students in grades one and two. The scores in math the findings from the TIMSS56 and the Progress in display similar regional patterns of relative performance International Reading Literacy Study. 57 to those in Indonesian. Students in Ketapang have the highest scores across all grades, whereas students in The averages in Indonesian and math are low, West Manggarai and Landak have the lowest scores. with some regional patterns. In grade one, students The math scores in grades three to five are the lowest obtained an average score of 32.7 in Indonesian, scores of all the grades, only marginally higher than the varying from 26.2 in Landak to 38.8 in Ketapang. Test benchmark random guess score of 25 percent. There scores in grade one vary from zero to 100, except in is also a lower standard deviation in math scores in Landak, West Manggarai, and East Manggarai. Scores in grades four and five. grade two range from 36 in Landak to 49 in Ketapang. In 57 Mullis et al. 2012. 57. Second, students’ results are also classified by 36 percent in Ketapang and 57 percent in West grade-level competencies based on the 2006 Manggarai) are unable to recognize letters. Between curriculum standards, to present learning outcomes 43 percent of the students in West Manggarai and 62 in a simple and meaningful way for community percent in Ketapang have a basic understanding of and local education stakeholders.58 Classifications letters but no basic reading competency. Only 3 percent are assigned based on students’ test scores, which are of grade one students in Sintang attained the grade one defined by the number of questions answered correctly level in Indonesian. In grade two, between 2 percent of on the test (tables 52 and 53). These classifications are the students in East Manggarai and 8 percent in Sintang applied to the Indonesian and math test results. There are and Landak have no understanding of letters. On four classifications of competencies: whether a child is (1) average, 80 percent of the students in grade two have unable to recognize letters/numbers, (2) able to recognize only a basic understanding of letters and no reading letters/numbers but lacking basic competencies, (3) below competency. Approximately 11 percent of the students current grade-level competencies, or (4) at or above current in grade two have reached the level of grade one in grade-level competencies. The classification is expected Indonesian—ranging from 5 percent of the students in to contribute to raising stakeholders’ awareness of their West Manggarai and Landak to 17 percent in Ketapang. children’s learning achievements as well as providing Most students in grades three to five reached information about the general quality of teaching and reading and writing competency levels that are two learning in their school. This information should stimulate grades below their current grade level. By grade three, concrete actions toward improvement in student learning all the students reached a basic understanding of letters; outcomes. Further explanation of the classifications of however, the majority of the grade three students (between student competencies is provided in annex B. 57 percent in Ketapang and 74 percent in Landak) acquired The results are in stark contrast with the high only the grade one level in Indonesian. In West Manggarai, proportion of parents who reported being satisfied 20 percent of the students in grade three acquired only or very satisfied with their child’s learning outcomes. a basic understanding of letters and no reading ability. In Tables 52 and 53 (in annex A) show the proportion of Ketapang, East Manggarai, and Sintang, between one-fifth students who acquired each of the four classifications of and two-fifths of the students reached the grade two level. student competency in Indonesian and math. Confirming No students in grade three in the sample schools had the findings in table 50, these tables show that student yet reached the grade three level in Indonesian. Similarly, learning outcomes are low. In Indonesian, barely any the overwhelming majority of the students in grade four students reached the competency level corresponding to (between 81 percent in West Manggarai and 93 percent the grade they currently attended, which may be partially in Landak) reached the grade two level—they were two explained by the timing of the survey. In math, between grades behind in their competencies. Almost one-fifth 6 and 13 percent of the students in grades one to three of the students in grade four in West Manggarai have no reached the level of the grade they currently attended. basic reading ability in Indonesian. In the sample areas, Nevertheless, in all grades for Indonesian and math, the an average of 4 percent of the students in grade five (up majority of the students are two grade levels behind to 9 percent of the students in West Manggarai) remain their current grade in their competencies, implying that excluded from further learning due to only having a basic these students have not yet mastered competency of the understanding of letters. Between 75 percent of the previous grade level from which they had graduated. students in grade five in Sintang and 87 percent in Landak only reached the grade three level in Indonesian, whereas Most students in grade one were unable to an average of 17 percent across all five districts reached recognize letters, and most students in grade two the grade four level. had no basic reading comprehension. In Indonesian, the vast majority of the students in grade one (between The math test results were similar to those for the Indonesian test: overall, the majority of the students in each grade are, on average, two grade levels behind in their ability. However, in grade 58 This classification framework for interpreting and reporting test results has been adopted from the model applied by several one, there are more students who reached grade citizen-led assessments, such as the Annual Status of Education Report and Uwezo (Plaut and Jamierson Eberhardt 2015). The one competency in math compared with Indonesian, citizen-led assessment movement, which was initiated by Pratham, including up to 20 percent of grade one students in is an attempt by civil society organizations to gather evidence on learning–specifically basic literacy and numeracy–and use it for two Ketapang and Sintang. The proportion of students who main purposes: to increase awareness of low learning outcomes and stimulate actions that are intended to address the learning reached the level of their current grade decreased gap (Plaut and Jamierson Eberhardt 2015). 58. STUDENT ABSENCE AND LEARNING OUTCOMES rapidly thereafter, reaching zero by grade three. of students whose parents were also surveyed using the A percentage of students barely learned anything parent instrument and students who attend schools with across all the grades, with shares varying in different an active committee. Again, the regression results are grades. The largest share of students who exhibit merely correlations and cannot be interpreted as causal only a basic competency in math is in grade five. The effects. Columns (1) and (4) present regressions on highest competency level that grade five math students student characteristics for the Indonesian and math test reached is the grade three level. scores, respectively. There are common correlates for Indonesian and math and correlates that are significantly associated with only one or the other. A parent’s own Determinants of Student Learning participation in early childhood education is a strong This subsection presents the results of OLS positive correlate of their child’s learning outcomes in regressions of student test scores based on a set Indonesian and math, as are a mother’s education level of student, parent, and school characteristics. above junior secondary school and parent satisfaction The student characteristics include gender, early with their child’s outcomes in Indonesian or math. The childhood education participation, whether they live level of a father’s education is a strong correlate of the with their parents, and whether they report being Indonesian scores, whereas only a father’s education helped by parents when studying at home. The parent above senior secondary level is statistically significantly characteristics include mother and father education associated with the math scores. Parents who reported levels across the entire sample. In the subsample of that they help their child study at home is significantly students whose parents were also surveyed using and positively associated with students’ math scores, the parent instrument, we add dummy variables but not with their Indonesian scores. Notably, adding for students whose parents reported helping their school characteristics (columns (2)-(3) and (5)-(6)) does child study at home and parents who reported high not affect the correlations of the student-level variables satisfaction the levels with student outcomes in with test scores. Indonesian or math (very good or good). Although the regressions find some positive The school characteristics included in the associations between teacher characteristics regressions are comprised of the same variables and students’ test scores, teacher absence has no as for the teacher absence regressions and association. Teacher absence, whether measured as variables for teacher characteristics and district absence from teaching (conditional on being scheduled fixed effects. That is, the regressions include shares to teach)59 or as the share of classes observed without of PNS teachers, teachers with a bachelor’s degree, a teacher (columns (3) and (6), respectively), is not teachers in their current school for more than five statistically significantly associated with student test years, certified teachers, teachers who received a scores. Positive correlations with student scores, remote area allowance, and teachers expressing high in Indonesian and math, occur with teachers with a satisfaction (> 4) with their salary/honorarium. The bachelor’s degree, the share of certified teachers in the regressions also include dummies for schools where school, and having a school committee that has met with the school committee chairperson was selected by the principal and parents in the previous academic year. vote during meetings attended by parents, and where Interestingly, seniority at the school level, measured by the school committee reported holding (separate the share of teachers who have taught for more than or joint) meetings with the principal and parents five years at the school, is negatively correlated with during 2015/16. We also include teacher absence test scores. There are positive correlations between rates (defined as teachers absent from teaching or several school characteristics, such as the share of classes observed without a teacher, depending on teachers receiving a remote area allowance and the the specification), as well as district fixed effects in share who are highly satisfied with their salary. all the regressions, to control for differences in local conditions and education policy in particular. Several of the parent characteristics are positively associated with students’ test scores, but school characteristics are not. Table 54 presents the results of OLS regressions for the Indonesian (columns (1) to (3)) and math (columns (4) to (6)) scores for the samples 59 Similar results are obtained when using unconditional teacher absence from teaching. 59. Table 54. OLS Regressions of Selected Student and School Characteristics on Student Test Scores Variable (1) (2) (3) (4) (5) (6) Attended PAUD/ECED 7.348*** 7.050*** 6.882*** 7.652*** 7.491*** 7.205*** (0.939) (0.951) (0.949) (0.916) (0.926) (0.925) Living with parents -1.944 -2.460 -2.517 -0.954 -1.573 -1.652 (2.248) (2.237) (2.237) (2.194) -2.177 (2.180) Parents help at home -0.275 -0.360 -0.353 1.756*** 1.798*** 1.800*** (0.635) (0.633) (0.633) (0.619) (0.616) (0.617) Mother education: SD 0.459 0.599 0.598 0.862 1.064 1.035 (1.112) (1.109) (1.109) (1.083) (1.078) (1.079) Mother education: SMP 3.063** 2.985** 2.906** 2.682** 2.912** 2.724** (1.309) (1.310) (1.309) (1.275) (1.273) (1.273) Mother education: SMA or above 5.574*** 5.724*** 5.675*** 3.708*** 3.997*** 3.861*** (1.410) (1.413) (1.412) (1.374) (1.373) (1.374) Father education: SD 2.311* 2.323* 2.315* 0.159 0.121 0.141 (1.261) (1.255) (1.255) (1.232) (1.223) (1.224) Father education: SMP 3.628*** 3.761*** 3.738*** 0.367 0.522 0.506 (1.405) (1.401) (1.401) (1.371) (1.363) (1.365) Father education: SMA 4.078*** 4.010*** 4.023*** 2.458* 2.487* 2.537* (1.449) (1.445) (1.445) (1.415) (1.407) (1.409) Satisfied with learning outcome 6.211*** 5.919*** 5.909*** 6.314*** 6.055*** 6.059*** (0.533) (0.534) (0.534) (0.519) (0.519) (0.520) Teacher absence rate - 1.403 -0.532 - 4.109*** 0.999 (1.263) (1.169) (1.226) (1.138) Share of PNS teachers - -3.124 -3.309 - -1.724 -1.945 (2.067) (2.070) (2.007) (2.012) Share of teachers with minimum bachelor’s degree - 5.220*** 5.209*** - 4.652*** 4.608*** (1.008) (1.008) (0.979) (0.980) Share of teachers with minimum 5 years at school - -4.894** -4.732** - -9.962*** -9.425*** (1.915) (1.908) (1.861) (1.857) Share of certified teachers - 9.561*** 9.765*** - 12.25*** 12.45*** (2.095) (2.100) (2.033) (2.040) Share of teachers receiving special allowance - -0.207 -0.348 - 1.627** 1.408* (0.783) (0.783) (0.763) (0.763) Share of teachers highly satisfied with their salary - 1.527 1.441 - 2.891** 2.883** (1.312) (1.317) (1.279) (1.285) Committee chairperson is selected by voting - 0.913 0.719 - 1.191** 0.798 (0.592) (0.580) (0.576) (0.565) Committee had meetings with parents and principal - 2.114*** 2.148*** - 1.467** 1.557** (0.631) (0.630) (0.612) (0.612) Constant 26.31*** 22.96*** 23.71*** 24.31*** 21.87*** 22.94*** (2.560) (2.958) (2.967) (2.495) (2.876) (2.888) Observations 4,998 4,963 4,963 5,011 4,975 4,975 R-squared 0.101 0.113 0.113 0.113 0.132 0.130 Note: Standard errors are in parentheses. All regressions include district dummies. Columns (1) to (3) report results from regressions of Indonesian scores; columns (4) to (6) report results from regressions of math scores. All regressions are run on the sample of students whose parents are also surveyed using the parent instrument and who go to schools with an active committee. In columns (2), and (4), teacher absence is the share of teachers absent from teaching conditional on being scheduled to teach. In columns (3) and (6), teacher absence is the share of classes observed without teachers. ECED = early childhood education; PAUD = Pendidikan Anak Usia Dini (early childhood education programs); PNS = civil servants; Sekolah Menengah Atas (senior high school); SD = standard deviation; SMA = SMP = Sekolah Menengah Pertama (junior high school). It is important to keep in mind that in the regression we are at partial correlations. So teacher absence is positively correlated with math scores, holding all variables constant. It is possible that this is due to some correlation between teacher absence and one or some of the other correlates included in the regression. *** p < 0.01, ** p < 0.05, * p < 0.1. 60. 07 CONCLUSION Conclusions This report presents the educational context of five districts Landak Sintang located in remote Indonesia, namely, Ketapang, Landak, and Sintang in the province of West Kalimantan, as well as West Manggarai and East Manggarai in the province of NTT. The Ketapang report is based on a comprehensive survey of primary schools and Manggarai Manggarai their personnel (principals, teachers, and committees) as well as an assessment of teacher absence, student learning outcomes, and a Barat Timur detailed parent survey. The survey finds that On average, the study schools are located five hours away from poor infrastructure district capitals, and 2.3 hours away from financial institutions. This seriously hampers long traveling time affects principals and teachers who need to travel education service to district capitals for administrative and logistical purposes, or to delivery. retrieve their salaries on a regular basis. Improving infrastructure – better roads, telecommunication, and electricity – is likely to contribute to improving the quality of education in remote Indonesia, since it increases the attractiveness of these areas for better qualified personnel and makes communication, money withdrawal, and supervision easier. Teacher presence is The study areas are characterized by high teacher absence compared remarkably low in the with average estimates for schools across the country, but they are five study districts, similar to estimates for remote schools. Enumerators observed that suggesting that there is approximately one in four classes did not have a teacher. One in four a dire need and ample teachers who were scheduled to be teaching during the observation room for improvement. was absent from school, with one-third of these teachers reported to be absent due to school-related assignments. Overall, 20 percent of the scheduled classes did not have a teacher, and approximately 25 percent of the scheduled teachers were absent from teaching. The Government of Indonesia should address this serious problem, as it directly affects whether students learn in school. In addition, various ways to improve teacher presence need to be tested, including strengthening teacher monitoring, enforcing teacher evaluation and invoking sanctions for underperforming teachers. It should be noted, however, that OLS regression results demonstrate that, all else being equal, teacher absence is not significantly correlated, or positively correlated, with student test scores. Thus, ensuring that teachers are present and involved in teaching might not automatically lead to improved learning outcomes among students. 61. There may be a Parents and communities would benefit from participating in setting service need to reset parent standards alongside principals and teachers, correspondingly participating in and community the evaluation of teacher performance against these standards, and having expectations of clear channels for effectively voicing their concerns about the outcomes the quality of the of teacher services. On the supply side, principals and teachers cited the educational services lack of active involvement of parents in their children’s education as an their children receive, important factor hindering student learning. Being held (more) accountable particularly in regards by parents and communities may lead principals and teachers to become to teacher presence. more motivated to improve their performance, including on attendance. Another area for Barely any students had reached the ability level corresponding to the improvement would grade they currently attended—the majority were two grade levels be the provision behind. In contradiction with the observed low quality of education in of more accurate the study areas, principals, parents, and school committees reported information to high satisfaction levels with teacher performance and student learning parents and school outcomes. This is in stark contrast with the share of students who committees on actual demonstrated no basic ability in Indonesian and/or math. This situation levels of student may be influenced by a limitation of the quantitative survey, which may learning outcomes. be subject to bias due to respondents answering what they think is the “right” answer. Additional qualitative data may identify this as a potential issue.60 Nevertheless, the regression results suggest that parents’ satisfaction with their children’s learning outcomes is a strong correlate of student scores. If this is the case, then student learning outcomes may improve by providing parents, school committees, and communities with relevant and targeted information on student learning development and the service quality levels expected from teachers. In addition to social Providing additional support, monitoring, and evaluation of teachers accountability, may improve their motivations and efforts to improve their presence performance-based and service performance. However, these social pressures may not payment of teacher be enough to affect lasting changes in teachers’ behavior. As such, allowance should pecuniary mechanisms may provide stronger incentives, particularly for be considered as a underperforming teachers. As mentioned in the introduction section, means to improve the Government of Indonesia has provided eligible teachers working in teacher performance. remote areas with a Tunjangan Khusus, at a substantial amount ranging from IDR 1.5 million up to one times teacher’s monthly base salary. However, recipients of Tunjangan Khusus turned out to have higher absenteeism rate compared to non-recipients (Toyamah, et al., 2010). As such, mechanisms to make Tunjangan Khusus more effective in directly incentivizing teacher presence or service performance should be tested. A 2017 World Bank survey in 100 schools in ten districts, including five districts covered in this study, indicated that principals and teachers preferred performance-based over seniority-based determinants of pay and promotion (Perez-Alvarez, et al. 2019). 60 As undertaken in Vernez, Karam, and Marshall (2012). 62. There are important Many teachers in the remote schools surveyed do not possess higher differences between education degrees, but there are substantial variations across the districts. the five districts, For example, although more than 70 percent of non-PNS teachers in the especially in terms of sample schools in the two NTT districts have at least one university degree, school management, overall nearly 67 percent of non-PNS teachers have only a high school working conditions diploma as their highest level of education. However, across all the districts, for principals and these differences appear to have marginal influence on student learning teachers, teacher outcomes and teacher performance. characteristics and Education degrees Education degrees efforts, parental involvement, and student learning 70% non-PNS teachers in the sample 67% non-PNS teachers have only a high achievements. schools in the two NTT districts have at least one university degree school diploma as their highest level of education The findings suggest The qualifications and status of non-PNS teachers, who represent the the need to upgrade majority of teachers in remote areas, should be improved. UNICEF (2012) the qualifications finds higher absenteeism among non-PNS and local teachers. There and teaching skills are also important differences in teachers’ salaries and honorariums, of principals and depending on their status. Non-PNS teachers receive very low salaries teachers, focusing and are relatively dissatisfied with their salary levels. Further, efforts are on districts where needed to raise the motivation of teachers operating in remote areas, a large number in particular through ensuring smooth implementation of teacher of education certification and the remote area allowance. Alternative approaches personnel are found should be investigated as well, since, at the national level, these factors with insufficient were not shown to have a large effect on improving learning. qualifications. 63. 64. ANNEX A. TABLES Annex A. Tables Table 6. Distance and Travel Time from the Village Hall to Administrative and Financial Institutions West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Subdistrict Distance (km) 28.4 28.3 24.3 37.8 14.4 19.8 office Travel time (hours0 1.3 1.1 1.0 1.6 1.5 1.4 District Distance (km) 149.1 268.3 102.6 129.2 94.3 78.3 government Travel time (hours0 4.8 6.8 3.2 4.6 4.8 3.5 office Nearest district Distance (km) 139.7 200.5 103.6 166.3 71.6 76.4 government Travel time (hours0 4.9 5.6 3.5 5.9 3.8 3.5 office Post office Distance (km) 53.8 45.4 48.9 71.7 34.3 46.2 Travel time (hours0 2.3 1.5 1.9 2.9 2.3 2.4 Bank Distance (km) 53.4 44.0 47.4 80.3 28.5 26.9 Travel time (hours0 2.2 1.4 1.8 3.2 1.9 1.4 ATM Distance (km) 56.8 54.1 47.2 76.4 32.7 42.6 Travel time (hours0 2.3 1.7 1.8 3.0 2.2 2.1 Cooperative Distance (km) 42.1 36.9 40.6 62.9 16.9 22.2 Travel time (hours0 1.6 1.2 1.7 2.3 1.4 1.1 Credit union Distance (km) 26.8 20.2 22.8 18.6 39.1 70.0 Travel time (hours0 1.2 0.6 1.2 0.9 3.7 3.1 Table 10. Gender Distribution of Students, by Grade West Kalimantan NTT All Grade Note: Gender teller machine; km = kilometers; NTT = East Nusa Tenggara. ATM = automated Areas West East Ketapang Landak Sintang Manggarai Manggarai All 23 23 19 23 24 26 1 Male 12 13 10 13 13 14 Female 11 11 9 11 11 12 All 21 21 18 20 24 27 2 Male 11 11 10 11 13 14 Female 10 10 8 9 11 13 All 22 21 17 22 26 26 3 Male 12 11 9 12 13 14 Female 10 10 8 10 13 11 All 22 20 20 22 28 27 4 Male 12 10 11 11 15 15 Female 11 10 9 11 13 12 All 22 20 18 23 26 27 5 Male 12 10 10 12 14 15 Female 11 10 8 11 12 13 All 22 19 20 21 26 25 6 Male 11 9 10 10 13 12 Female 11 10 10 11 13 13 Note: NTT = East Nusa Tenggara. 65. Table 11. Availability of Key School Facilities (Percentage of Sample Schools) West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Buildings Principal room 48 51 47 63 26 32 Teachers’ room 87 86 96 97 87 50 Sports equipment/field 99 100 96 99 100 100 Additional physical facilities Toilets 91 90 96 97 79 85 Toilets - only for teachers 64 66 80 65 50 50 Toilets - only for female students 50 56 61 53 32 35 Toilets - only for male students 41 51 59 40 29 15 Clean water 54 66 59 52 42 41 Electricty during school hours 30 36 18 40 24 24 Mobile phone signal 45 44 27 28 68 91 Teaching supporting facilities Library 54 58 43 48 47 91 Textbooks in sufficient number 39 37 35 42 39 41 Note: NTT = East Nusa Tenggara. Table 14. Instructional Language, Curriculum, and Teaching Load, Academic Year 2015/16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Main teaching language - principal report (% schools) Indonesian 86 92 94 92 63 74 Malay 1 5 0 1 0 0 Dayak 4 3 6 7 0 0 Manggarai 9 0 0 0 37 26 Curriculum used in 2015/16 - principal report (% schools) 2013 curriculum 1 2 0 1 0 0 2006 curriculum (education unit level) 2 5 2 2 0 0 2004 curriculum (competence based) 99 100 98 100 95 100 Curriculum used in 2015/16 - teacher report (% teachers) 2013 curriculum 1 1 1 1 0 0 2006 curriculum (education unit level) 94 94 93 95 93 94 2004 curriculum (competence based) 8 8 9 7 11 6 Subjects taught (% teachers) Teachers teaching 1 subject 14 13 12 11 21 17 Teachers teaching 2-3 subjects 12 6 6 7 23 22 Teachers teaching 4 subjects or more 74 81 82 83 56 61 Note: NTT = East Nusa Tenggara. 66. ANNEX A. TABLES Table 15. Instructional Time, Academic Year 2015/16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Effective teaching days Effective teaching days 226.38 235.37 228.18 206.56 240.24 243.970 Interruption in teaching (% schools) 25 34 31 34 3 0 Weekly teaching hours (average #) Grade 1 26 26 26 25 27 28 Grade 2 26 27 26 26 27 29 Grade 3 29 29 28 28 31 32 Grade 4 31 32 30 30 32 33 Grade 5 31 32 31 30 32 33 Grade 6 31 33 31 30 32 33 Note: NTT = East Nusa Tenggara. Table 22. Principals’ Living Conditions: Median Distance, Travel Time, and Transportation Cost from Home to School West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Overall Distance from house to school (km) 0.5 1 6 0.200 0.280 0.320 Travel time from house to school (minutes) 5 10 20 5 8.5 5 Transportation cost from house to school (Rp) 0 1,500 3,000 0 0 0 Principals living in same village as school Distance from house to school (km) 0.150 0.230 0.5 1 1 0.150 Travel time from house to school (minutes) 5 5 5 3 3 5 Transportation cost from house to school (Rp) 0 1,000 0 0 0 0 Others Distance from house to school (km) 8 11 12 7 3 3.5 Travel time from house to school (minutes) 30 30 40 30 29 30 Transportation cost from house to school (Rp) 8,000 7,500 10,000 5,000 6,000 10,000 Note: km = Kilometer; NTT = East Nusa Tenggara. 67. Table 23. Teachers’ Living Conditions: Median Distance, Travel Time, and Transportation Cost from Home to School West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai Overall Distance from house to school (km) 0.300 0.400 0.5 0.200 0.160 0.5 Travel time from house to school (minutes) 5 5 10 5 6.5 10 Transportation cost from house to school (Rp) 0 900 0 0 0 0 Teachers living in same village as school Distance from house to school (km) 0.200 0.300 0.200 0.200 1 0.300 Travel time from house to school (minutes) 5 5 5 5 5 10 Transportation cost from house to school (Rp) 0 0 0 0 0 0 Others Distance from house to school (km) 5 7 15 4 3 3 Travel time from house to school (minutes) 30 20 45 30 30 30 Transportation cost from house to school (Rp) 4500 4500 8000 4500 0 3650 Note: km = Kilometer; NTT = East Nusa Tenggara. Table 30. Principals’ Allowance West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Total allowance # principals receiving it 252 57 44 83 38 30 % principals receiving it 97 100 92 100 100 91 Monthly median (Rp) 3,575,525 3,058,575 4,064,868 4,605,900 3,125,000 2,766,113 Professional allowance (TP), past 12 months # principals receiving it 182 46 32 53 29 22 % principals receiving it 70 81 67 64 76 67 Monthly median (Rp) 3,257,880 3,079,287 3,706,908 3,602,400 1,850,967 2,920,000 Special allowance (TK), past 12 months # principals receiving it 92 8 10 60 12 2 % principals receiving it 36 14 21 72 32 6 Monthly median (Rp) 2,678,825 90,625 2,707,200 2,834,425 2,609,500 156,250 Additional income allowance (Tamsil), past 12 months # principals receiving it 110 35 11 43 15 6 % principals receiving it 42 61 23 52 39 18 Monthly median (Rp) 408,333 350,000 100,000 775,000 833,333 350,833 Note:NTT = East Nusa Tenggara; TK = Tunjangan Khusus (special allowance); TP = Tunjangan Profesi (professional allowance). 68. ANNEX A. TABLES Table 31. Teachers’ Allowance West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai PNS teachers # teachers receiving total allowance 631 130 96 221 114 70 % teachers receiving it 84 93 72 92 83 67 Median monthly total allowance (Rp) 1,333,333 1,227,083 952,099 2,694,500 1,299,533 433,333 # teachers receiving professional allowance (TP) 244 47 41 89 44 23 % teachers receiving it 32 34 31 37 32 22 Median monthly professional allowance (Rp) 2,898,700 2,800,000 3,333,333 3,133,333 1,782,458 2,023,711 # teachers receiving special allowance (TK) 144 18 13 80 27 6 % teachers receiving it 19 13 10 33 20 6 Median monthly special allowance (Rp) 2,144,583 100,000 2,549,342 2,733,333 1,000,000 736,250 # teachers receiving additional allowance (Tamsil) 265 65 16 87 62 35 % teachers receiving it 0.350 0.460 0.120 0.360 0.450 0.330 Median monthly additional income allowance (Rp) 400,000 380,000 100,000 747,500 683,333 62,500 Non-PNS teachers # teachers receiving total allowance 519 186 37 100 136 60 % teachers receiving it 45 66 22 29 70 34 Median monthly total allowance (Rp) 200,000 300,000 37,500 221,667 164,167 329,167 # teachers receiving professional allowance (TP) 5 2 0 1 2 0 % teachers receiving it 0 1 0 0 1 0 Median monthly professional allowance (Rp) 1,375,000 1,437,500 - 1,500,000 955,833 - # teachers receiving special allowance (TK) 81 21 3 29 20 8 % teachers receiving it 7 8 2 8 10 5 Median monthly special allowance (Rp) 1,057,500 1,050,000 416,667 1,375,000 1,025,000 708,333 # teachers receiving additional allowance (Tamsil) 130 70 0 12 25 23 % teachers receiving it 11 25 0 3 13 13 Median monthly additional income allowance (Rp) 366,667 366,667 - 500,000 100,000 1,080,000 Note:NTT = East Nusa Tenggara; PNS = civil servants; TK = Tunjangan Khusus (special allowance); TP = Tunjangan Profesi (professional allowance). 69. Table 32. Principals’ and Teachers’ Allowance Delivery, 2014-16 West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Professional allowance (TP) # principals receiving it in 2014 149 34 30 46 21 18 Entirely (% receiving) 88 71 87 96 95 94 # teachers receiving it in 2014 182 28 35 69 28 22 Entirely (% receiving) 84 54 91 91 82 91 # principals receiving it in 2015 167 39 32 53 23 20 Entirely (% receiving) 89 85 88 94 78 95 # teachers receiving it in 2015 207 40 40 72 31 24 Entirely (% receiving) 90 85 93 96 74 100 # principals receiving it in 2016 180 42 35 54 27 22 Entirely (% receiving) 62 76 66 65 37 55 # teachers receiving it in 2016 248 48 46 83 45 26 Entirely (% receiving) 69 67 65 86 47 65 Special allowance (TK) # principals receiving it in 2014 76 11 12 46 4 3 Entirely (% receiving) 97 100 100 96 100 100 # teachers receiving it in 2014 189 28 27 116 10 8 Entirely (% receiving) 95 93 93 96 90 100 # principals receiving it in 2015 75 8 11 48 5 3 Entirely (% receiving) 93 100 100 94 80 67 # teachers receiving it in 2015 194 33 14 122 16 9 Entirely (% receiving) 94 97 93 97 75 89 # principals receiving it in 2016 85 1 11 63 9 1 Entirely (% receiving) 76 0 100 79 33 100 # teachers receiving it in 2016 224 41 18 111 41 13 Entirely (% receiving) 82 76 89 90 66 69 Additional income allowance (Tamsil) # principals receiving it in 2014 116 28 15 58 9 6 Entirely (% receiving) 95 82 100 100 89 100 # teachers receiving it in 2014 332 116 26 109 39 42 Entirely (% receiving) 90 85 96 94 97 86 # principals receiving it in 2015 120 30 14 59 10 7 Entirely (% receiving) 91 77 100 100 80 71 # teachers receiving it in 2015 340 121 28 104 42 45 Entirely (% receiving) 90 87 96 95 83 89 # principals receiving it in 2016 117 29 11 62 14 1 Entirely (% receiving) 66 72 100 65 36 0 # teachers receiving it in 2016 374 122 23 117 82 30 Entirely (% receiving) 74 66 91 88 66 67 Note:NTT = East Nusa Tenggara; PNS = civil servants; TK = Tunjangan Khusus (special allowance); TP = Tunjangan Profesi (professional allowance). 70. ANNEX A. TABLES Table 35. Teachers’ Reported Satisfaction (% Teachers) West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai a. Satisfaction with central government appreciation of teachers’ role Very low (1) 15 16 16 10 17 20 Low (2-3) 21 20 28 22 16 20 Average (4-5) 29 33 30 28 29 26 High (6-7) 35 31 26 40 38 33 PNS Very low (1) 6 6 11 2 7 6 Low (2-3) 13 11 17 11 13 15 Average (4-5) 28 30 29 26 28 28 High (6-7) 53 53 43 61 53 50 Non-PNS Very low (1) 21 21 20 15 25 28 Low (2-3) 26 24 37 29 17 23 Average (4-5) 30 34 31 30 30 26 High (6-7) 23 20 13 26 27 23 b. Satisfaction with district education office performance in organizing primary schools Very low (1) 7 8 7 5 9 7 Low (2-3) 20 19 25 19 20 18 Average (4-5) 35 36 38 33 32 38 High (6-7) 38 37 30 43 39 37 PNS Very low (1) 4 4 6 1 5 5 Low (2-3) 16 20 18 13 20 10 Average (4-5) 33 27 39 33 32 37 High (6-7) 47 49 37 54 43 47 Non-PNS Very low (1) 9 10 8 8 12 9 Low (2-3) 22 18 31 24 20 22 Average (4-5) 36 41 38 34 31 38 High (6-7) 32 31 24 34 37 31 c. Satisfaction with appreciation of people around the school of role as a teacher Very low (1) 4 2 5 3 6 6 Low (2-3) 15 13 20 13 11 18 Average (4-5) 30 28 40 34 23 25 High (6-7) 51 57 35 50 60 51 PNS Very low (1) 2 1 4 0 3 4 Low (2-3) 12 13 15 12 7 11 Average (4-5) 30 26 43 31 23 27 High (6-7) 56 59 38 57 67 57 Non-PNS Very low (1) 5 2 5 5 8 7 Low (2-3) 16 14 25 14 13 21 Average (4-5) 30 29 37 36 23 24 High (6-7) 48 55 33 45 55 47 71. West Kalimantan NTT All Areas Ketapang Landak Sintang West Manggarai East Manggarai d. Satisfaction with performance of government and people in the village in helping organizing school Very low (1) 6 6 5 4 11 8 Low (2-3) 22 25 29 17 23 21 Average (4-5) 36 35 40 38 32 34 High (6-7) 36 35 26 42 33 37 PNS Very low (1) 4 4 4 1 9 5 Low (2-3) 19 18 27 14 23 16 Average (4-5) 37 32 39 38 34 39 High (6-7) 40 46 30 47 34 39 Non-PNS Very low (1) 8 7 5 6 13 10 Low (2-3) 24 28 31 19 24 23 Average (4-5) 35 36 41 38 30 31 High (6-7) 33 30 23 38 33 35 Note:NTT = East Nusa Tenggara; PNS = civil servants. 72. ANNEX A. TABLES Table 36. Teachers’ Reported Satisfaction with Their Salary (Percentage of Teachers) West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Satisfaction with salary/honorarium received as teacher in current school Very low (1) 13 11 14 11 14 20 Low (2-3) 26 26 35 23 24 25 Average (4-5) 27 27 27 28 24 31 High (6-7) 33 36 24 37 38 24 PNS Very low (1) 3 5 3 0 4 6 Low (2-3) 15 16 27 11 16 8 Average (4-5) 28 27 32 26 23 38 High (6-7) 53 52 38 62 57 48 Non PNS Very low (1) 20 14 23 19 21 28 Low (2-3) 33 31 41 32 29 35 Average (4-5) 27 27 22 29 25 27 High (6-7) 20 28 13 20 25 10 Ideal salary/honorarium Much Lower 1 1 1 1 0 2 Lower 2 1 2 1 2 4 Equal to current salary 19 15 15 21 27 18 Higher 66 70 77 68 54 58 Much higher 12 14 5 9 17 19 PNS Much Lower 0 0 0 0 1 1 Lower 2 0 2 2 1 4 Equal to current salary 25 16 24 28 33 23 Higher 63 71 68 65 55 55 Much higher 9 13 6 5 10 17 Non PNS Much Lower 1 1 1 1 0 3 Lower 2 1 3 1 2 3 Equal to current salary 16 14 8 17 23 14 Higher 67 69 83 70 54 60 Much higher 14 15 4 12 21 20 Note:NTT = East Nusa Tenggara; PNS = civil servants. 73. Table 41. School Committee Background West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai # Schools having an active committee 254 56 48 82 35 33 # Schools having an inactive committee 14 2 2 6 3 1 (Active) school committee characteristics (% active committees) Current committee establishment 2016-17 20 22 19 27 6 10 2010-15 63 69 55 65 52 69 2005-09 14 5 23 6 32 17 Before 2005 4 4 2 3 10 3 Committee only manages this school 93 95 88 95 97 91 Committee has article of association/bylaws 15 16 13 17 6 18 Committee received funds for activities from 12 13 7 20 6 7 school in 2015/2016 Median funds received from school (Rp) 326,000 300,000 200,000 301,000 2,300,000 300,000 School provides office space for committee 2 4 2 1 0 3 74. ANNEX A. TABLES Table 44. School Committee Activities, 2015/16 (% Committees) West Kalimantan NTT Pilots Areas West East Ketapang Landak Sintang Manggarai Manggarai Meetings of committee with principal and parents Meetings of committee with principal and parents, 78 75 70 87 71 85 2015/16 Meetings organization initiative Always principal 40 27 50 51 28 29 Always committee 9 7 6 6 20 14 Sometimes principal, sometimes committee 51 66 44 44 52 57 Topics covered during meetings Preparation of students evaluation 84 90 88 82 83 75 Suggestions and complaints from parents 83 80 84 85 83 79 School budget and financial resources 77 73 72 75 96 75 Student discipline and behavior 76 83 69 72 71 89 Student learning outcomes 76 85 72 73 79 71 Teacher discipline and behavior 68 73 69 69 58 64 Recruitment of teachers 55 56 44 51 63 68 Curriculum and teaching methods 49 54 44 62 38 25 Contribution of the committee/parents 44 41 38 31 68 68 Others 83 80 84 82 79 89 Meetings of committee with principal only Meetings of committee with principal, past month 35 39 40 49 11 9 Meetings of committee with principal, 2015/16 48 54 52 56 37 27 Meetings organization initiative: Always principal 38 23 42 46 54 11 Always committee 18 17 13 15 31 33 Sometimes principal, sometimes committee 44 60 46 39 15 56 Internal committee meetings Internal committee meeting, past month 14 18 17 15 3 12 Internal committee meeting, 2015/16 21 27 20 21 11 24 Internal meeting generated suggestion/feedback for 91 93 100 94 50 88 school Suggestions from committee to school Rehabilitation of infrastructure and furniture 63 71 56 63 50 57 Discipline improvement of teachers and/or students 46 57 56 38 50 29 Teaching and learning process 29 43 56 13 0 14 Teacher quality improvement 17 7 33 13 50 14 Purchase of learning tools 10 14 22 0 0 14 Teacher welfare improvement 8 14 11 0 50 0 Others 40 21 44 56 0 43 School implemented some suggestions from committee 81 93 89 75 50 71 Suggestions from committee implemented by school Rehabilitation of infrastructure and furniture 33 46 25 33 0 20 Discipline improvement of teachers and/or students 54 62 63 42 100 40 Teaching and learning process 23 31 38 17 0 0 Teacher quality improvement 5 8 13 0 0 0 Purchase of learning tools 10 15 13 0 0 20 Teacher welfare improvement 10 15 0 8 100 0 Others 41 15 25 75 0 60 Note:NTT = East Nusa Tenggara. 75. Table 49. Student Absence and Reasons, by Grade All West Kalimantan NTT Areas Ketapang Landak Sintang West Manggarai East Manggarai Overall # Registered students 35543 7350 5682 11449 5709 5353 Official student absence rate (%) 8 9 9 7 9 7 Observed student absence rate (%) 14 24 13 13 11 8 # Classrooms 1702 366 301 543 249 243 % Classrooms with all students present 35 35 37 39 26 36 Grade 1 # Registered students 6195 1373 982 2058 884 898 Official student absence rate (%) 11 11 13 11 13 7 Observed student absence rate (%) 19 27 17 19 16 8 # Classrooms 289 64 51 94 39 41 % Classrooms with all students present 25 20 20 31 15 34 Grade 2 # Registered students 5663 1257 895 1739 859 913 Official student absence rate (%) 9 11 8 8 9 7 Observed student absence rate (%) 16 27 8 16 12 10 # Classrooms 289 62 52 91 41 43 % Classrooms with all students present 37 29 40 41 37 35 Grade 3 # Registered students 5848 1231 863 1908 974 872 Official student absence rate (%) 8 10 10 6 9 7 Observed student absence rate (%) 13 24 12 10 13 9 # Classrooms 290 61 51 92 45 41 % Classrooms with all students present 36 34 45 39 22 34 Grade 4 # Registered students 6015 1187 1004 1897 1039 888 Official student absence rate (%) 7 9 8 6 7 6 Observed student absence rate (%) 14 27 15 12 8 6 # Classrooms 280 60 49 89 43 39 % Classrooms with all students present 36 37 47 39 19 36 Grade 5 # Registered students 6048 1179 942 2012 984 931 Official student absence rate (%) 7 9 8 5 8 6 Observed student absence rate (%) 11 20 13 8 11 7 # Classrooms 281 59 49 90 42 41 % Classrooms with all students present 37 41 37 42 26 32 Grade 6 # Registered students 5774 1123 996 1835 969 851 Official student absence rate (%) 6 6 9 4 5 7 Observed student absence rate (%) 11 15 12 11 8 6 # Classrooms 273 60 49 87 39 38 % Classrooms with all students present 42 48 33 45 36 45 76. ANNEX A. TABLES Table 52. Classification of Student Competency in Indonesian West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Grade 1 (% students) BMH - no understanding of letters 47 36 56 43 57 51 BMKD - understand letters but no basic competency 51 62 44 55 43 49 KD1 - grade 1 level 1 2 1 3 0 0 Grade 2 (% students) BMH - no understanding of letters 6 5 8 8 5 2 BMKD - understand letters but no basic competency 80 74 85 73 90 88 KD1 - grade 1 level 11 17 5 14 5 9 KD2 - grade 2 level 3 5 2 4 1 1 Grade 3 (% students) BMKD - understand letters but no basic competency 7 2 6 3 20 8 KD1 - grade 1 level 64 57 74 63 65 66 KD2 - grade 2 level 29 41 20 34 15 26 KD3 - grade 3 level 0 0 0 0 0 0 Grade 4 (% students) BMKD - understand letters but no basic competency 6 1 4 1 17 9 KD2 - grade 2 level 89 91 93 91 81 86 KD3 - grade 3 level 6 8 3 8 3 4 KD4 - grade 4 level 0 0 0 0 0 0 Grade 5 (% students) BMKD - understand letters but no basic competency 4 1 3 1 9 6 KD3 - grade 3 level 80 79 87 75 85 79 KD4 - grade 4 level 17 20 10 23 6 16 KD4 - grade 5 level 0 0 0 0 0 0 Note:NTT = East Nusa Tenggara. 77. Table 53. Classification of Student Competency in Math West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Grade 1 (% students) BMH - unable to recognize numbers 37 17 49 28 56 47 BMKD - understand numbers but no basic competency 50 62 44 52 41 48 KD1 - grade 1 level 13 21 7 20 3 5 Grade 2 (% students) BMH - unable to recognize numbers 6 2 7 7 9 4 BMKD - understand numbers but no basic competency 49 36 64 42 60 51 KD1 - grade 1 level 41 58 27 44 29 42 KD2 - grade 2 level 4 4 2 7 2 3 Grade 3 (% students) BMKD - understand numbers but no basic competency 12 2 13 7 30 16 KD1 - grade 1 level 82 90 84 86 66 80 KD2 - grade 2 level 6 8 3 7 3 4 KD3 - grade 3 level 0 0 0 0 0 0 Grade 4 (% students) BMKD - understand numbers but no basic competency 3 1 2 1 11 5 KD2 - grade 2 level 93 93 96 95 87 91 KD3 - grade 3 level 4 6 2 4 3 4 KD4 - grade 4 level 0 0 0 0 0 0 Grade 5 (% students) BMKD - understand numbers but no basic competency 19 13 18 14 27 26 KD3 - grade 3 level 81 87 82 86 73 74 KD4 - grade 4 level 0 0 0 0 0 0 KD4 - grade 5 level 0 0 0 0 0 0 Note: NTT = East Nusa Tenggara. 78. ANNEX A. TABLES Table 55. Primary School Availability at the Subdistrict Level # Villages # Villages # Private Subdistrict # Villages # PS # Public PS w/ 1 PS w/ 2 PS PS All areas 0 235 270 184 51 248 22 Ketapang 0 59 59 59 0 58 1 Jelai Hulu 8 8 8 0 8 0 Kendawangan 9 9 9 0 9 0 Manis Mata 8 8 8 0 8 0 Marau 4 4 4 0 4 0 Sandai 6 6 6 0 6 0 Simpang Dua 4 4 4 0 4 0 Simpang Hulu 10 10 10 0 10 0 Sungai Laur 10 10 10 0 9 1 Landak 38 51 25 13 51 0 Air Besar 12 17 7 5 17 0 Jelimpo 5 7 3 2 7 0 Mempawah Hulu 7 10 4 3 10 0 Menjalin 3 4 2 1 4 0 Menyuke 4 4 4 0 4 0 Ngabang 3 4 2 1 4 0 Sebangki 4 5 3 1 5 0 Sintang 82 88 63 19 87 1 Kayan Hilir 17 17 12 5 17 0 Kayan Hulu 13 13 12 1 13 0 Ketungau Hilir 5 5 5 0 5 0 Ketungau Hulu 16 16 12 4 16 0 Ketungau Tengah 11 13 8 3 13 0 Sepauk 12 13 9 3 12 1 Tempunak 8 11 5 3 11 0 M. Barat 27 38 15 12 29 9 Boleng 4 6 2 2 5 1 Komodo 2 4 0 2 4 0 Kuwus 5 7 3 2 3 4 Macang Pacar 5 6 4 1 5 1 Ndoso 5 7 2 3 7 0 Welak 6 8 4 2 5 3 M. Timur 29 34 22 7 23 11 Elar 5 5 5 0 4 1 Kota Komba 2 4 0 2 3 1 Lamba Leda 6 6 5 1 3 3 Rana Mese 9 11 6 3 7 4 Sambi Rampas 7 8 6 1 6 2 Note: The total numbers of schools and villages in each subdistrict, pilot subdistricts. PS = primary school. 79. Table 56. Mean Student Test Scores, by Grade and Gender West Kalimantan NTT Grade Subject All Areas West East Ketapang Landak Sintang Manggarai Manggarai Overall Bahasa 32.720 38.810 26.230 36.140 26.980 30.300 1 Math 33.5 44.200 26.610 39.070 22.520 26.5 Bahasa 44 49.240 36.010 44.880 42.010 45.860 2 Math 43.310 51.650 35.620 45.930 35.950 43.030 Bahasa 28.180 33.930 25.170 31.430 20.070 25.940 3 Math 30.320 36.430 27.360 32.930 23.030 28.100 Bahasa 34.240 38.870 31.670 37.970 27.040 32.290 4 Math 29.420 32.490 27.570 31.5 25.370 28.260 Bahasa 32.440 35.220 29.900 35.350 26.330 32.020 5 Math 31.920 34.390 31.200 34.170 27.680 29.510 Female students Bahasa 33.730 40.25 26.630 38 27.320 30.560 1 Math 32.470 43.210 26.130 38.470 21.25 24.610 Bahasa 46.540 52.240 37.390 47.770 45.640 47.200 2 Math 43.960 52.770 35.880 47.390 36.430 41.880 Bahasa 29.560 35.580 26.380 33.550 21.880 26.060 3 Math 31.230 36.600 28.660 34.520 24.770 27.980 Bahasa 35.870 40.300 32.910 39.440 29.020 33.710 4 Math 29.840 32.890 27.400 31.980 26.030 28.440 Bahasa 34.010 36.710 31.240 36.830 28.270 32.900 5 Math 32.310 35.030 31.520 34.330 28.440 29.160 Male students Bahasa 31.880 37.620 25.870 34.570 26.700 30.120 1 Math 34.380 45.010 27.040 39.570 23.570 28.170 Bahasa 41.810 46.5 34.830 42.400 38.900 44.740 2 Math 42.75 50.630 35.390 44.690 35.530 43.990 Bahasa 26.960 32.600 24.080 29.620 18.260 25.820 3 Math 29.520 36.290 26.200 31.570 21.300 28.200 Bahasa 32.730 37.480 30.650 36.490 25.210 31.130 4 Math 29.020 32.100 27.710 31.010 24.75 28.120 Bahasa 30.980 33.710 28.760 33.860 24.670 31.240 5 Math 31.570 33.740 30.930 34.010 27.040 29.810 Note: NTT = East Nusa Tenggara. 80. ANNEX A. TABLES Table 57. Student Absence, by Grade and Gender West Kalimantan NTT All Areas West East Ketapang Landak Sintang Manggarai Manggarai Overall Registered female students 16,837 3,503 2,634 5,481 2,714 2,505 Absent female students 1,179 288 222 343 192 134 Registered male students 18,706 3,847 3,048 5,968 2,995 2,848 Absent male students 1,673 406 310 430 299 228 Grade 1 Registered female students 2,851 626 457 940 408 420 Absent female students 307 70 59 96 49 33 Registered male students 3,344 747 525 1,118 476 478 Absent male students 389 87 73 127 69 33 Grade 2 Registered female students 2,615 601 401 800 388 425 Absent female students 220 61 33 63 35 28 Registered male students 3,048 656 494 939 471 488 Absent male students 272 72 38 77 46 39 Grade 3 Registered female students 2,691 557 406 864 476 388 Absent female students 179 45 37 48 31 18 Registered male students 3,157 674 457 1,044 498 484 Absent male students 292 78 48 65 58 43 Grade 4 Registered female students 2,865 571 451 952 495 396 Absent female students 179 41 32 56 32 18 Registered male students 3,150 616 553 945 544 492 Absent male students 254 66 46 66 43 33 Grade 5 Registered female students 2,908 583 430 999 460 436 Absent female students 154 42 28 38 32 14 Registered male students 3,140 596 512 1,013 524 495 Absent male students 258 61 50 55 46 46 Grade 6 Registered female students 2,907 565 489 926 487 440 Absent female students 140 29 33 42 13 23 Registered male students 2,867 558 507 909 482 411 Absent male students 208 42 55 40 37 34 Note: NTT = East Nusa Tenggara. 81. Table 58. OLS Regressions of School, Principal, and Teacher Characteristics on Reported Teacher Shortage at School Variable (1) (2) # Teachers at school -0.110*** -0.208*** (0.016) (0.014) # PNS teachers at school -0.016 -0.070*** (0.018) (0.016) # Of students at school 0.002*** 0.005*** (0.001) (0.001) # Ff classes at school 0.004 -0.018 (0.032) (0.026) Teacher presence (# present teachers/#teachers) at -0.244* -1.010*** school (0.142) (0.135) Age of principal -0.004 -0.001 (0.003) (0.003) Principal is female 0.041 0.061 (0.071) (0.066) Principal is married -0.102 -0.177 (0.144) (0.131) Age of teacher -0.001 (0.003) Teacher is female -0.050 (0.049) Teacher is married 0.064 (0.068) Teacher is not civil servant -0.110*** (0.035) Constant 1.879*** 4.636*** (0.282) (0.303) Adjusted R2 0.203 0.166 Number of observations 259 1,918 Note: Standard errors are in parentheses. All regressions include district dummies. The dependent variable is binary (1=teacher shortage , 0=no reported teacher shortage). Column 1 contains results on principals while column 2 displays results from teacher regressions. *** p < 0.01, ** p < 0.05, * p < 0.1. 82. ANNEX B. STUDENT COMPETENCY CLASSIFICATIONS Annex B. Student Competency Classifications The classifications were assigned based on student test scores, which are defined by the number of questions answered correctly on the test. These classifications are applied for the results of the Indonesian and math tests. There are four classifications of competencies for whether a child is (1) unable to recognize letters/numbers, (2) below basic competencies, (3) below current grade-level competencies, or (4) at or above current grade-level competencies. A student is classified as unable to recognize letters/numbers—the lowest level of competency—when s/he is unable to answer all the letter and number recognition questions in the Indonesian and math tests. This state of learning outcomes can only be detected for students in grades one and two, who are given letter and number recognition questions. A student is classified as having below basic competency when s/he is able to recognize letters and numbers but unable to answer at least more than half of the total number of questions that are two grade levels below the current grade-level standards in the test.61 A student is then classified as having below current grade level competency when s/he is unable to answer at least more than half of the total number of questions at their current grade-level standards in the test. Lastly, a student is classified as having at or above current grade level competency when s/he can answer more than half of the total number of questions at their current or higher grade-level standards in the test, which is the ideal learning outcome expected from students. For type 3 and 4 classifications, a student’s specific grade-level competency is thus determined by the number of questions at the highest grade-level standards that s/he can answer correctly, whereby s/he is able to answer at least more than half of the highest grade-level questions provided in the test. 61 The exception to this rule is grade one Indonesian, whereby students are classified in this type of competency when they cannot answer at least one set of grade one–level reading comprehension questions. This exception was applied because there were only two sets of reading comprehension questions in the grade one Indonesian test. 83. 84. REFERENCES References ACDP (Education Sector Analytical and Capacity Development Partnership). 2014. Study on Teacher Absenteeism in Indonesia 2014. Jakarta, Indonesia: ACDP. Al-Samarrai, Samer, and Pedro Cerdan-Infantes. 2013. “Where Did All the Money Go? Financing Basic Education in Indonesia.” In Education in Indonesia, edited by D. Suryadarma and G. Jones, 109–38. Singapore: ISEAS–Yusof Ishak Institute. ASER (Annual Status of Education Report). 2014. Annual Status of Education Report (Rural) 2013. New Delhi, India: ASER Centre. BPS. 2018. Statistik Indonesia: Statistical Yearbook of Indonesia 2017. Jakarta: Badan Pusat Statistik. BPS. 2019. Profil Kemiskinan di Indonesia September 2018. Jakarta: Badan Pusat Statistik. Beatty, Amanda, Emelie Berkhout, Luhur Bima, Thomas Coen, Menno Pradhan and Daniel Suryadarma. 2018. “Indonesia Got Schooled: 15 Years of Rising Schooling and Flat Learning Profiles.” RISE Working Paper 18/026, Research on Improving Systems of Education, Oxford, United Kingdom. Bjork, Christopher, and Dewi Susanti. "Community Participation and Teacher Accountability: Improving Learning Outcomes in Remote Areas of Indonesia." Unpublished manuscript, last modified 11 June, 2019. Microsoft Word file. Chaudhury, Nazmul, Jeffrey Hammer, Michael Kremer, Karthik Muralidharan, and Hasley F. Rogers. 2006. “Missing in Action: Teacher and Health Worker Absence in Developing Countries.” Journal of Economic Perspectives 20 (1): 91–116. Chen, Dandan. 2011. “School-Based Management, School Decision-Making and Education Outcomes in Indonesian Primary Schools.” Policy Research Working Paper 5809, World Bank, Washington, DC, https://openknowledge. worldbank.org/handle/10986/3572. De Ree, Joppe, Karthik Muralidharan, Menno Pradhan, and Halsey Rogers. 2018. “Double for Nothing? Experimental Evidence on an Unconditional Teacher Salary Increase in Indonesia.” Quarterly Journal of Economics 133 (2): 993– 1039. Duflo, Esther. 2001. “Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment.” American Economic Review 91 (4): 795–813. Gaduh, Arya, Menno Pradhan, Jan Priebe, and Dewi Susanti. "Scores, Camera, Action? Incentivizing Teachers in Remote Areas." Unpublished manuscript, last modified 28 March, 2019. Microsoft Word file. Gove, Amber, and Anna Wetterberg, eds. 2011. The Early Grade Reading Assessment: Applications and Interventions to Improve Basic Literacy. Research Triangle Park, NC: RTI Press. Hanushek, Eric A., and Ludger Woessmann. 2007. “The Role of Education Quality for Economic Growth.” Policy Research Working Paper 4122, World Bank, Washington, DC, https://openknowledge.worldbank.org/ handle/10986/7154. 85. Little, Angela W., ed. 2006. Education for All and Multi-Grade Teaching: Challenges and Opportunities. London: Springer. Ministry of Villages, Disadvantaged Regions and Transmigration. 2016. Peraturan Menteri Desa, Pembangunan Daerah Tertinggal, dan Transmigrasi Republic Indonesia. Nomor 2, Tahun 2016. Tentang Indeks Membangunan Desa. Jakarta, Indonesia: Menteri Desa, Pembangunan Daerah Tertinggal, dan Transmigrasi. Mullis, Ina V. S., Michael O. Martin, Pierre Foy, and Kathleen T. Drucker. 2012. PIRLS (Progress in International Reading Literacy Study) 2011 International Results in Reading. Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Lynch School of Education, Boston College. Mullis, Ina V. S., Michael O. Martin, Pierre Foy, and M. Hooper. 2016. TIMSS (Trends in International Mathematics and Science Study) 2015 International Results in Mathematics. Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Lynch School of Education, Boston College. OECD (Organisation for Economic Co-operation and Development). 2016. PISA 2015 Results in Focus. Paris: OECD, Program for International Student Assessment. Perez-Alvarez, Marcello, Jan Priebe, and Dewi Susanti. 2019. “Teacher Accountability and Pay-for-Performance Schemes in (Semi-) Urban Indonesia: What do Education Stakeholders Think?” Unpublished manuscript, last modified 25 January, 2019. Microsoft Word file. Platas, Linda M., Leanne R. Ketterlin-Gellar, Aarnout Brombacher, and Yasmin Sitabkhan. 2014. Early Grade Mathematics Assessment (EGMA) Toolkit. Research Triangle Park, NC: Research Triangle Park Press. Plaut, Daniel, and Molly Jamierson Eberhardt. 2015. Bringing Learning to Light: The Role of Citizen Led Assessments in Shifting the Education Agenda. Washington, DC: Results for Development Institute. Pradhan, Menno, Daniel Suryadharma, Amanda Beatty, Maisy Wong, Armida Alishjabana, Arya Gaduh, and Rima Prama Artha. 2014. “Improving Educational Quality through Enhancing Community Participation: Results from a Randomized Field Experiment in Indonesia.” Policy Research Working Paper 5795, World Bank, Washington, DC, https://openknowledge.worldbank.org/handle/10986/3559. Pritchett, Lant. 2013. The Rebirth of Education: Schooling Ain’t Learning. Baltimore, MD: Brookings Institution Press. Stern, Jonathan, and Lee Nordstrum. 2014. Indonesia 2014: The National Early Grade Reading Assessment and Snapshot of School Management Effectiveness Survey Report and Findings. Research Triangle Park, NC: Research Triangle Park Press. Suharti. 2013. “Trends in Education in Indonesia.” In Education in Indonesia, edited by Daniel Suryadarma and Gavin W. Jones, 15–52. Singapore: ISEAS–Yusof Ishak Institute. Suryadarma, Daniel, Asep Suryahadi, Sudarno Sumarto, and F. Hasley Rogers. 2006. “Improving Student Performance in Public Primary Schools in Developing Countries: Evidence from Indonesia.” Education Economics 14 (4): 401–29. Toyamah, Nina, Bambang Sulakson, Meuthia Rosfadhila, Silvia Devina, Sirojuddin Arif, Stella Aleida Hutagalung, Eduwin Pakpahan, and Asri Yusrina. 2010. Teacher Absenteeism and Remote Area Allowance: Baseline Survey. Jakarta, Indonesia: SMERU Research Institute. UNICEF (United Nations Children’s Emergency Fund). 2012. We Like Being Taught: A Study on Teacher Absenteeism in Papua and West Papua. Jakarta, Indonesia: UNICEF. Usman, Syaikhu, Akhmadi, and Daniel Suryadarma. 2004. When Teachers Are Absent: Where Do They Go and What Is the Impact on Students? Jakarta, Indonesia: SMERU Research Institute. Uwezo. 2012. Are Our Children Learning? Annual Learning Assessment Report. Kampala, Uganda: Uwezo. Vernez, Gorges, Rita Karam, and Jeffrey H. Marshall. 2012. Implementation of School-Based Management in Indonesia. Santa Monica, CA: Rand Corporation, sponsored by the World Bank. World Bank. 2004. World Development Report 2004: Making Services Work for Poor People. Washington, DC: World Bank and Oxford University Press. 86. REFERENCES World Bank. 2008. Teacher Employment and Deployment in Indonesia: Opportunities for Equity, Efficiency and Quality Improvement. Washington, DC: World Bank. ———. 2013a. Early Childhood Education and Development in Poor Villages of Indonesia: Strong Foundations, Later Success. Jakarta, Indonesia: World Bank. ———. 2013b. Indonesia: Spending More or Spending Better: Improving Education Financing in Indonesia. Jakarta, Indonesia: World Bank. ———. 2015. The Role of BOS in Improving Education Outcomes in Indonesia. Jakarta, Indonesia: World Bank. ———. 2016. Indonesia’s Rising Divide. Washington, D.C., USA. World Bank ———. 2018. World Development Report 2018: Learning to Realize Education’s Promise. Washington, DC: World Bank, doi:10.1596/978-1-4648-1096-1. ———. 2018b. Indonesia Economic Quarterly June 2018: Learning More, Growing Faster. Washington, DC: World Bank, https://openknowledge.worldbank.org/handle/10986/29921 License: CC BY 3.0 IGO. 87. 88. 89.