Report No: ACS6915 People's Republic of China China Early Child Development Early Childhood Education in Yunnan November 18, 2013 EASHE EAST ASIA AND PACIFIC Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper 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. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. 1 Contents Acknowledgements:.......................................................................................................................... 6 Abbreviations and Acronyms ............................................................................................................ 7 Overview ........................................................................................................................................... 8 Existing Policies on and Status of Early Childhood Education ....................................................... 8 Financing Early Childhood Education ............................................................................................ 9 Status of Children in Rural Yunnan: Results from a Rural Household Survey ............................. 10 Preschool Quality and Environment in Kunming ........................................................................ 13 Preschool Teacher Recruitment, Career Development, Salaries, and Shortages ........................ 14 Policy and Program Implications ................................................................................................. 15 Introduction .................................................................................................................................... 20 Part I. Existing Policies on and Status of Early Childhood Education............................................... 24 Legal and Institutional Framework.............................................................................................. 24 Administration and Service Delivery ........................................................................................... 27 Main Trends in Early Childhood Education in Yunnan................................................................. 30 Yunnan’s GER is low by national comparison, with major internal disparities ........................... 30 Share of private enrollment has been increasing ....................................................................... 32 One-year pre-primary classes are prevalent in rural areas of Yunnan ........................................ 33 Yunnan’s rural share of preschool enrollment is one of the highest in the country ................... 36 Declining pupil-teacher ratios and inequality in supply of trained teachers in rural areas ........ 39 Summary and Discussion ............................................................................................................ 42 Part II. Financing Early Childhood Education................................................................................... 44 Structure of Educational Finance ................................................................................................ 44 Financing Preschool Education ................................................................................................... 46 Total spending on preschool education ....................................................................................... 46 Summary and Discussions ........................................................................................................... 55 Part III. Status of Children in Rural Yunnan: Results from a Rural Household Survey ..................... 57 Rural Home Care Conditions, Childrearing Practices, and Childhood Developmental Status .... 58 Characteristics of Kindergarten Enrollment ................................................................................ 65 Correlates of Development Outcomes ........................................................................................ 70 Correlates of Kindergarten Enrollment ....................................................................................... 74 Summary and Discussion ............................................................................................................ 75 Part IV. Preschool Quality in Kunming ............................................................................................. 80 Assessment of Preschool Quality with ECERS-R .......................................................................... 81 Key Findings and Results ............................................................................................................. 82 Summary and Discussion ............................................................................................................ 87 Part V. Study of Preschool Teachers in Yunnan ............................................................................... 89 Description of the Teachers/Director Survey .............................................................................. 89 Key Findings and Results ............................................................................................................. 90 Demand and supply for preschool teachers in Yunnan............................................................... 99 International Comparison ......................................................................................................... 103 Summary and Discussion .......................................................................................................... 110 Part VI: Policy and Program Implications ...................................................................................... 112 Increasing public financing of preschool education and diversifying financing/delivery models .................................................................................................................................................. 112 Expanding service delivery models ........................................................................................... 115 Improving quality and atmosphere of existing preschools ....................................................... 116 Strengthening family and parental education........................................................................... 116 Increasing awareness of secondhand smoke’s harmful effects ................................................ 116 Developing a preschool teaching force ..................................................................................... 117 Moving toward integrated ECD provision for all including 0-3 Year Olds.................................. 118 Appendix: ...................................................................................................................................... 120 A. Kindergarten Gross Enrollment Rate (est.) of three Years before Primary School, 2004-2010, by Region................................................................................................................................... 120 B. Regression Results of Multivariate (Model 1-8) and Logistic (Model 9) Analysis of the Household Survey on Childhood Development in Yunnan, 2013 ............................................. 121 References..................................................................................................................................... 129 Boxes Box 3.1: The Effect of Early Reading on Children’s Cognitive Development ................................... 76 Box 3.2 The Effect of TV Watching on Childhood Development Outcomes .................................... 77 Box 6.1: Shanghai’s Voucher System to Improve Education Access and Quality for Migrant Children ......................................................................................................................................... 113 Box 6.2: Philippines Conditional Cash Transfers............................................................................ 115 Figures Figure 1.0: Research Design ............................................................................................................ 23 1 Figure 1.1 2011 Gross Enrollment Rate (GER) in selected Asia and Pacific Countries .................... 30 Figure 1.2 Gross Enrollment Rate (est.) of Preschools, by Region, 2004-2010 ............................... 31 Figure 1.3 Share of Enrollment in Private Preschools in China, 2005-2012 .................................... 33 Figure 1.4 Share of Enrollment in Private Preschools in Yunnan, 2003-2009 ................................. 33 Figure 1.5 All Preschool Classes in China, 2003-2010 ..................................................................... 34 Figure 1.6 One-year Pre-primary Classes in China, 2003-2010 ....................................................... 35 Figure 1.7 Enrollment in One-year Pre-primary Classes as a Percentage of Total Preschool Enrollment in Cities, Counties and Rural Areas, by Province, 2010 ................................................ 36 Figure 1.8 Enrollment in One-year Pre-primary Classes as a Percentage of Total Preschool Enrollment in Cities, Counties and Rural Areas, by Regions, 2010.................................................. 37 Figure 1.9 Number of Kindergartens in China, 2003-2010.............................................................. 37 Figure 1.10 Number of Students in Kindergartens and One-year Pre-primary Schools in China, 2003-2010 ....................................................................................................................................... 38 Figure 1.11 Number of Students in Kindergartens and One-year Pre-primary Schools in Yunnan, 2003-2010 ....................................................................................................................................... 38 Figure 1.12 Number of Students in Kindergartens and One-year Pre-primary Schools in Western Provinces, 2003-2010 ...................................................................................................................... 39 Figure 1.13 Pupil-Teacher Ratio, by Region, 2003-2010.................................................................. 40 Figure 1.14 Pupil-Teacher Ratios in Kindergartens, by Province, 2010 ........................................... 41 Figure 1.15 Share of Enrollment and Trained Teachers in Rural, County and Urban Areas by Region, 2010 ................................................................................................................................................ 42 Figure 2.1 Composition of National Educational Funds (all educational subsectors), 2011 ........... 44 Figure 2.2 Distribution of Students by Level of Education, 1980-2010 ........................................... 45 Figure 2.3 Composition of Preschool Education Funding, China and Yunnan, 2011 ...................... 47 Figure 2.4 Total Spending per Student on Preschool Education, by Type and Region, 2011 .......... 48 Figure 2.5 Budgetary Spending per Student on Preschool Education, by Type and Region, 2011 . 49 Figure 2.6 Budgetary and Total Spending per Student on Preschool Education in China, by Province, 2011................................................................................................................................. 50 Figure 2.7 Per-Student Kindergarten Personnel Expenditures in Western Provinces, in RMB, 2010 ........................................................................................................................................................ 51 Figure 2.8 Per-Student Kindergarten Operational Expenditures in Western Provinces, in RMB, 2010 ................................................................................................................................................ 51 Figure 2.9 Share of Rural Preschool Finance and Enrollment, by Region, 2011.............................. 52 Figure 2.10 Budgetary Spending per Student on Rural Preschool Education, by type, 2011 ......... 52 Figure 2.11 Total Spending per Student on Rural Preschool Education, by type, 2011 .................. 53 Figure 2.12 Percentage of Tuition and Fees out of Total Investment in Preschool Education in the Western Provinces, 2010 ................................................................................................................ 54 Figure 2.13 Proportion of Kindergarten Tuition and Fees of Average Per-Capita Income by Kindergarten Grade, Urban versus Rural residents, in RMB, 2010 ................................................. 55 Figure 3.1 Children's Primary Caregiver in Sample Counties in Yunnan, 2013 ............................... 58 Figure 3.2 Birth Weight of Children in Rural Yunnan, 2013............................................................. 59 Figure 3.3 Infant Breastfeeding in Yunnan, by County Type, 2013 ................................................. 59 Figure 3.4 Educational Levels of Mothers, Fathers, and Caregivers in Yunnan, 2013 ..................... 60 Figure 3.5 Annual Household Income in Yunnan, 2013 .................................................................. 60 2 Figure 3.6 Reported Time Spent Reading and Playing with Children and Watching TV in Yunnan, 2013 ................................................................................................................................................ 61 Figure 3.7 Reported Time Spent Reading and Playing with Children and Watching TV in Yunnan, 2013 ................................................................................................................................................ 62 Figure 3.8 Frequency of Various Types of Food Intake in Yunnan, 2013 ......................................... 63 Figure 3.9 Number of Books at Home across County Types in Yunnan, 2013................................. 64 Figure 3.10 Primary Communication Language(s) at Home in Yunnan, 2013 ................................. 65 Figure 3.11 Enrollment of Kindergarten/One-year Pre-primary Classes, by County Type in Yunnan, 2013 ................................................................................................................................................ 66 Figure 3.12 Enrollment Rate in Kindergartens or One-year Pre-primary Classes, by County Types, 2013 ................................................................................................................................................ 66 Figure 3.13 The Most Important Reason(s) for Selecting Kindergartens/Preschools in Yunnan, 2013 ................................................................................................................................................ 67 Figure 3.14 The Most Important Reason(s) for not Attending Kindergartens/Preschools in Sample Counties in Yunnan ......................................................................................................................... 68 Figure 3.15 Respondents' Estimates of Nearest Kindergartens and Associated Costs ................... 69 Figure 3.16 Percent as a Share of Households that Recognize a Closer Kindergarten than a Nearest Primary School in Yunnan, by County Type, 2013 ........................................................................... 70 Figure 3.17 Social and Cognitive Development Scores across Regions in Yunnan, 2013 ................ 72 Tables Table 1.1 Selected Laws and Regulation on Preschool Education ................................................... 24 Table 1.2 Selected Expansion of Preschool Education in Yunnan ................................................... 28 Table 1.3 Estimated Gross Enrollment Rate in Yunnan, by Prefecture, 2011 .................................. 31 Table 1.4 Share of Private and Public Preschools, Enrollment and Teachers, 2012 ........................ 32 Table 1.5 Annual Growth Rates of Teachers and Enrollment, 2003-2010 (%)................................. 40 Table 2.1 Distribution of Enrollment and Educational Funding, by Level of Education 2011 (in %)46 Table 2.3 Items and Standards for Kindergarten Fees in Kunming.................................................. 54 Table 3.1 Smoking Habits in Yunnan, by Region, 2013 (%).............................................................. 64 Table 3.2 Descriptive Statistics of Estimated Distances to Nearest Kindergartens, Estimated Monthly Fees, and Estimated Distances to Nearest Primary Schools ............................................. 69 Table 3.3 Average Monthly Fees as Perceived by Respondents ...................................................... 69 Table 3.4 Selected Index Scores across Main Explanatory Variables .............................................. 71 Table 4.1 Sample Size and Distribution ........................................................................................... 81 Table 4.2 Mean ECERS-R Scores and Ratings of Kunming Preschools Survey ................................. 82 Table 4.3 Comparison of Mean ECERS-R Scores between Kunming and Plan Bangladesh Preschools ....................................................................................................................................... 83 Table 4.4 Comparative Mean ECERS-R Scores between Kunming and Madrasa Project Preschools in East Africa.................................................................................................................................... 83 Table 4.5 Comparative Mean ECERS-R Scores between Kunming and Brazilian Preschools........... 84 Table 4.6 Mean ECERS-R Scores between Kunming and Beijing Preschools ................................... 85 Table 4.7 Mean ECERS-R Scores between Rated and Unrated Kindergartens in Kunming ............. 85 Table 4.8 Mean ECERS-R Scores in Kunming, by Area ..................................................................... 86 3 Table 4.9 Mean ECERS-R Scores in Kunming, by Ownership status ................................................ 86 Table 5.1 Sample Size and Distribution ........................................................................................... 90 Table 5.2 Mean and Standard Deviation of Age and Years of Preschool Work Experience of Kindergarten Teachers in the Sample.............................................................................................. 90 Table 5.3 Entry Educational Level of Preschool Teachers ................................................................ 91 Table 5.4 Current Educational Level of Preschool Teachers ............................................................ 93 Table 5.5 Pupil-Teacher Ratio across Kindergarten Types ............................................................... 95 Table 5.6 Salaries of Kindergarten Teachers across Kindergarten Types ......................................... 96 Table 5.7 Salary Level for Teachers within and outside and Staff Quota System ............................ 97 Table 5.8 Average Preschool Teacher Salary in Yunnan, 2003 -2010 .............................................. 98 Table 5.9 Average Salary (RMB 10,000), by Occupation in Yunnan, 2011....................................... 98 Table 5.10 Estimated Gross Enrollment Rate, 2011-2020（%） .................................................... 99 Table 5.11 Estimated Demand for and Shortages of Preschool Teachers in Yunnan, 2013-2020 . 100 Table 5.12 Estimated Pupil-Teacher Ratios, Demand for and Shortages of Preschool Teachers in Yunnan, 2013-2020 ....................................................................................................................... 100 Table 5.13 Estimated Demand for Preschool Teachers in Yunnan, 2013-2020 ............................. 101 Table 5.14 Estimated Shortage of Preschool Teachers in Yunnan, 2013-2020 .............................. 101 Table 5.15 Estimated Number of Teachers Required, by Level of Educational Background ......... 102 Table 5.16 Number of Enrollments, Admissions and Graduations at Teachers’ Colleges and Provincial Vocational Schools Majoring in Preschool Education in Yunnan, 2011-2012............... 103 Table 5.17 Kindergarten Teacher Salary Comparison across Countries ........................................ 106 Table 5.18 Salary Comparison across Industries in Austria ........................................................... 107 Table 5.19 Salary Comparison across Industries in Poland ........................................................... 107 Table 5.20 An International Comparison of Monthly Salary and Weekly Work Hours for Kindergarten Teachers................................................................................................................... 108 Table 5.21 In-Service Training for Kindergarten Teachers ............................................................. 109 Table 6.1: Examples of Extended Free Public Education in China ................................................. 112 4 5 Acknowledgements: The study was led by Xiaoyan Liang (Senior Education Specialist, Task Team Leader) under the guidance of Xiaoqing Yu (East Asia and Pacific region (EAP) Human Development Director) and Luis Benveniste (EAP Sector Manager). The report was co-authored by Xiaoyan Liang, Yu Fu (Consultant) and Yinan Zhang (Consultant), with significant contributions from a few researchers who partnered with us to carry out the background studies, including the following individuals:  Dr. Xin Zhou of East China Normal University and Dr. Zhenguo Zhao of Henan University who conducted the preschool teacher study. In particular, Dr. Zhou herself is a renowned scholar in early childhood education. Dr. Zhou also served as an advisor to the overall Yunnan Early Child Development (ECD) study and her contributions add tremendous rigor and substance.  Ms. Yunli Xie of Yunnan Normal University who led the Early Childhood Environment Rating Scale- Revised for Pre-Schools (ECERS-R) survey, and Ms. Yan who led the implementation of the rural household survey. With great enthusiasm and energy, Ms. Xie immersed herself into the implementation of ECERS-R in the 24 preschools in Kunming and also prepared a background report on the results.  Ms. Zhiying Yan of Yunnan Normal University who led a team including Ms. Yue Zhang, Ms. Yunli Xie, and Ms. Yi Hua. They successfully completed the adaptation of the household survey instrument as well as the implementation of the survey, entry and cleaning of data, and initial analysis of the data.  Ms. Yi Hua of Yunnan Normal University who provided a review of ECD policies and institutions and who became the main contact person for the studies.  Mr. Xingxu Li and Mr. Taozhen Bai of Yunnan University of Finance provided technical assistance on the sampling framework and trained the survey field workers. They also advised on the statistical modeling during the data analysis. We owe special thanks to Mr. Jingsong Li, Dean of the Education Management Institute in the Yunnan Normal University who bravely accepted the challenging task of contracting with the World Bank to conduct the various background studies. Mr. Li’s dedication to education and to supporting his young faculty and students will be a model for all. The study would not have been possible without the support and collaboration from the officials in the Yunnan Provincial Department of Education, including Ms. Liqong Zhen from the Basic Education Department, Mr. Jie Song and Mr. Tiezhen Fu from the Foreign Loan Department, Ms. Yaqin Li, and Mr. Zhijun Yang from the Education Research Institute. The authors extend a special thanks to Mr. Jie Song for always being supportive and professional and a great colleague and friend. 6 Abbreviations and Acronyms CDRF Child Development Research Foundation EAP East Asia and Pacific region ECD Early Child Development ECDE Early Childhood Development and Education ECERS-R Early Childhood Environment Rating Scale- Revised for Pre-Schools GER Gross enrollment ratio GDP Gross domestic product GPA Grade point average MOE Ministry of Education MRC Madrasa Resource Centers OECD Organization of Economic Co-operation and Development PPS Probability proportional to size RMB Renminbi SIDS Sudden infant death syndrome TVET Technical and vocational education and training 7 Overview Yunnan is a medium-sized and relatively poor Chinese province on the southwestern border of China. There is a long history of collaboration between Yunnan Province and the World Bank over the last three decades. In 2012, the Yunnan Department of Education formally requested Bank support in conducting a review of early childhood education policies and programs in order to gain an in-depth and evidence-based understanding of the challenges the province faces in expanding early childhood education— in particular to rural and mountainous regions. The Bank’s China Education Team embarked on raising funds, designing and implementing a rather elaborate research agenda around early childhood education. The goal was to investigate key challenges, and to propose policy interventions for expanding the Early Child Development (ECD) coverage in rural Yunnan. The research included five interrelated studies: (i) a survey of existing ECD policies and institutions in Yunnan; (ii) the financing of preschool education; (iii) a household survey of a representative sample of 3-6 year olds in rural Yunnan; (iv) a survey of preschool quality in Kunming using the internationally recognized Early Childhood Environment Rating Scale— Revised for Pre-Schools (ECERS-R) scale; and (iv) a comprehensive study of preschool teachers covering qualifications, career development, salaries, as well as demand and supply. This report presents the findings from the background studies, and draws potential policy implications for improving the access to and quality of preschool education in Yunnan province. Existing Policies on and Status of Early Childhood Education China has now almost achieved universal 9-year basic education. Over the last decade, the country has devoted increasing attention to policy development in early childhood education. Even though China does not yet have a specific early childhood education law, it has established a rather elaborate set of guidelines and regulations pertaining to early childhood education. The most prominent one is the Medium- and Long-Term Education Development Plan Outline, which sets the targets for achieving preschool enrollment for 95 percent of 5 year olds, 80 percent of all 4 year olds, and 70 percent of all 3 year olds by the year 2020. Most recently in 2012, the Ministry of Education issued Learning and Development Guidelines for 3-to-6 year old children which describe children’s development stages with measurable indicators and suggest appropriate education interventions in health, language, society, science, and art. Efforts are now underway to develop a specific early childhood education law in China. Based on a review of official statistics, there are a number of important trends in early childhood education in China, as well as in the Yunnan province. Early childhood education has expanded significantly within the last few years. There are two main types of preschool programs for 3-6 year olds including: a regular 3-year program which is called kindergarten, and a one-year program attached usually to primary schools (学前班). The gross enrollment ratio (GER) of early childhood education rose from 49 percent in 2004 to approximately 63 8 percent nationwide in 2010.1 Yunnan registered a similar level of growth in early childhood education access. However, the province lags behind the national average and other western provinces with regard to overall preschool education coverage. For example, the GER in Yunnan was 55 percent in 2010, which was lower than the national average of 63 percent, as well as lower than the average for the western provinces. The province also evidences substantial disparities, as GER reached 95 percent in the capital Kunming City, but was only 33 percent in Zhaotong County. The share of private preschool enrollment has been rising in both China and Yunnan. Approximately 42 percent of preschools in Yunnan are private, and absorb about 26 percent of preschool enrollment. This resonates with the national picture, as 41 percent of preschools are private, absorbing about 34 percent of total preschool enrollment. Enrollment in one-year pre-primary classes, as opposed to regular-three-year kindergarten programs, is more common in Yunnan than in other parts of the country. Almost one out of three preschool enrollments in Yunnan takes place in one-year, pre-primary classes, as compared to the national average of 15 percent and the average of 26 percent in the western part of the country. The proportion is even higher in rural Yunnan, where 56 percent of preschool enrollment is in one-year programs. Since one-year, pre-primary classes are often attached to primary schools and taught by primary school teachers, they are essentially an extension of the primary curriculum, that is, they are not considered as developmentally appropriate for the 3-6 year olds. This statistic underscores the lower development status of early childhood education in rural Yunnan. The rapid growth of preschool teacher supply has contributed to the drop in pupil-teacher ratios across the nation. 2 In Yunnan in particular, the ratio has decreased from approximately 30 to 20 in recent years. However, urban areas still enjoy a more favorable pupil-teacher ratio, as well as a higher proportion of qualified teachers compared to rural areas. Rural areas account for 50 percent of total preschool enrollment, but only 22 percent of all trained teachers serve in rural areas. Financing Early Childhood Education Governmental appropriation is the biggest contributor to educational funds in China. However, funding is skewed against early childhood education. Approximately 80 percent of all available funds in education are from government appropriations. As total government 1 It is noted that due to lack of statistics on the number of children between age 3-6, the estimates were different from publicized national GER, which states 56.6% in 2010, 62.3% in 2011 and 64.5% in 2012. The report however keeps the estimated numbers to maintain consistent comparisons for different years and provinces. Therefore it is believed that the trend of GER reflects a more accurate picture, though the absolute numbers might be slightly overestimated. Another explanation for the difference is that Three-year Gross Enrollment Rate to Kindergartens (verses one-year pre-school classes) was specifically used by official statistics after 2010, which does not count the one-year preschool enrollment. The estimated GER however, has included enrollment from both regular 3-year kindergarten and the one-year program for meaningful comparisons. 2 The pupil-teacher ratio includes full-time teachers, caregivers, healthcare and administrative personnel, and other supporting staffs – not including part-time teachers 9 allocations in education rose to 4 percent of national GDP in 2012, the educational budget at both the aggregate and per-student level experienced a boost. However, early childhood education still receives relatively little government support. Preschool enrollment accounts for 13 percent of total educational enrollment in China, but receives only 2 percent of the national educational budget allocation. In this context, preschool education demonstrates a heavier dependence on private investment, which contributes up to 59 percent of total preschool educational funds. Yunnan, in particular, has committed increasing financial support to preschool education, but the province still faces low per-student expenditures and a wide gap between urban and rural areas. Total educational funding for the preschool sector in Yunnan increased at 32 percent per year from 2005 to 2011. Despite increasing funding, total spending per-student and budgetary spending per-student is lower than the national average and other western provinces. For instance, total spending per-child in 2011 (renminbi (RMB) 2,750) is only half of the average for all 12 western provinces (RMB 5,319). On the one hand, personnel expenditures accounted for a higher proportion of per-student budgetary spending than operational expenditures in Yunnan— nearly three quarters of per-student budgetary spending was used primarily for teacher salary and benefits. On the other hand, both per-student personnel expenditures and per-student operational expenditures are still lower than the national average and other western provinces, indicating lower teacher salaries and daily operational costs in the province. The gap in funding between rural and non-rural areas remains an issue: rural preschools accounted for 50 percent of both enrollment and number of facilities in the province, but received only 28 percent of all preschool funding, and 13 percent of governmental budgetary funds. Preschool education tuition and fees remain a significant burden for families in Yunnan, especially for those in rural areas. Tuition and fees revenue account for 43.6 percent of total investment in preschool education in Yunnan, which is higher than the western region average. According to the China Statistics Yearbook 2011, the average per-capita annual income in Yunnan in 2010 was 16,065 RMB for urban residents, and 3,952 RMB for rural residents. Therefore, the annual fees for a first-grade kindergarten would account for 8.7 percent of urban residents’ per-capita annual income— but 35.2 percent of rural residents’ per-capita annual income, making quality preschool almost unaffordable for rural poor families. Status of Children in Rural Yunnan: Results from a Rural Household Survey Results from the Rural Household Survey 2013 in Yunnan reveal several key aspects of the status of 3-6 year-old children living in rural Yunnan including: their home care conditions, development status in both cognitive and social dimensions, preschool enrollment status, and caregiver childrearing practices. In addition, multivariate regression analysis results confirm several factors key to childhood development outcomes and enrollment status at preschools. 10 In rural Yunnan, the proportion of children cared for by their grandparents (instead of parents) is 25 percent. Most of these are “left-behind children”.3 The parents surveyed have an average educational level of junior middle school, and their grandparents on average had only received a primary school education. Overall, the results confirm that children in rural Yunnan receive inadequate early stimulation in their households, and children from poor, border, or ethnic minority counties tend to fare even worse. Seventy-two percent of households in poor counties of Yunnan reported that they never played with their children; 47 percent reported that they never read to their children; 80 percent reported that they do not have any books at home. This finding merits particular attention, as there is much well-cited evidence in childhood development that playing with and reading books to children is associated with positive development outcomes. One positive childrearing practice – breastfeeding – is common in Yunnan: 92 percent of mothers breastfed their children. The negative practice of secondhand smoke is unfortunately prevalent in Yunnan. Approximately ninety percent of households have members smoke in the presence of their young children. A large number of research studies have confirmed that secondhand smoke has many harmful effects on infant and child health. Indeed, secondhand smoke during pregnancy is a direct cause of low infant birth weight. Infants who are exposed to secondhand smoke after birth face a bigger risk for sudden infant death syndrome (SIDS). Parental smoking leads to higher rates of both upper and lower respiratory tract illnesses such as asthma, bronchitis, middle ear infection (otitis media), and pneumonia in infants. Furthermore, secondhand smoke within 3 meters of the infant is associated with higher infant hospital admission rates for all illnesses in the first 18 months of life. For older children, those who breathe secondhand smoke experience less lung growth than those who do not. They are also more vulnerable to asthma attacks and ear infections. Rural children in Yunnan appear to watch TV frequently. Ninety-five percent of children watch TV every day, and 25 percent of them watch more than 2 hours per day. Survey data show that watching TV for more than 2 hours a day is positively associated with cognitive development.4 Further, research confirms that the effect of TV viewing is contingent upon the content and type of the TV programs watched. There is well-documented evidence that age-appropriate, curriculum-based educational television programs can have a positive effect on childhood cognitive development and school readiness. However, exposure to violence on TV during childhood is closely linked to more aggressive behavior. Close parental/adult monitoring of childhood TV viewing is therefore important. The multivariate analysis also reveals factors related to childhood physical and cognitive development. Childhood physical development as measured by weight is negatively correlated with being a female child in either poor or minority counties, but positively 3 Left-behind children refer to children whose parent(s) left the location of household registration to seek employment elsewhere. 4 One possible explanation is that TV viewing exposes children to more Mandarin language usage and helps children speak Mandarin with greater fluency. Another possible explanation is that the original regression analysis did not include all the possible variables that might affect children’s cognitive development. 11 associated with maternal education (tertiary education), family income, and regular health checkups. Similarly, family income, drinking milk frequently, urban versus rural residence, and size of family are all positively associated with a child’s height. In terms of cognitive and social development, family income and maternal education have a strong association with childhood cognitive development. Being a single child and being Han ethnicity (the biggest ethnic group in China) are also correlated with higher cognitive skills. Family income and Mandarin usage are positively correlated with social development. Coming from a poor or a border county tends to be negatively correlated with childhood social development. It is noteworthy that there is a clear positive correlation between preschool enrollment, frequency of health checkups, and positive childrearing practices and childhood development outcomes. Enrollment is both statistically correlated with children’s social and cognitive skills, and positively associated with children’s recent weight and height. In particular, there is a positive relationship between reading frequently to a child and the child’s social and cognitive development. Watching TV for 1-2 hours, or more than 2 hours a day, appears to be positively associated with cognitive skills— but negatively associated with social skills (not statistically significant). Several factors seem to contribute to childhood preschool enrollment: the family income, whether or not being “left behind” children, maternal education, distance to the nearest kindergarten, and the language spoken at home 5 . For example, the odds of children attending preschool with a family income between RMB 10,000 to RMB 20,000 are two times the odds for children with a family income lower than RMB 10,000. In addition, mothers with a high school are 1.5 times more likely to send their children to preschool than mothers who have no educational attainment at all. In light of these findings, the promotion of quality preschool education for rural children merits particular attention. Quality center-based programs (preschools) could be a suitable and effective environment to help improve childhood cognitive and social-emotional development and school readiness. Even though it might be difficult to influence some of the factors affecting the decision of preschool enrollment— such as the child’s gender, location, and family income— other factors such as maternal education, distance to nearest kindergartens, and affordability of preschool education can be more amenable to policy interventions. At the same time, families play an equal if not more important role in childhood development. A body of literature has shown that the lack of learning opportunities and 5 Yunnan’s cultural and ethnical diversity means that standard Chinese mandarin is not popularly used in most households. According to the household survey, only 2 percent speak standard mandarin as a primary means of communication at home, 74 percent speak a mandarin dialect that resembles the standard but with strong regional characteristics in pronunciation, while 24 percent use different ethnical minority languages that are often completely different in structure or pronunciation. These facts correlate with the difference in preschool enrollment, as almost all educational materials are prepared in standard mandarin. 12 low-quality caregiver-child interaction negatively affect childhood development. As the survey results suggest, an adequate level of early stimulation, playing, and reading to children in rural Yunnan is particularly important to reverse this trend. Various interventions that promote a positive home environment for young children can be devised. With proper guidance and support from health providers or group parent training, parents can play a crucial role in their children’s cognitive and social development by providing quality stimulation and interactions. Preschool Quality and Environment in Kunming In order to obtain a better understanding of preschool quality, a study of 24 kindergartens in the capital Kunming was conducted. The ECERS-R scale was used to assess the quality and atmosphere of those kindergartens. This is the first study in measuring quality (or quality proxy) of preschools using an international rating scale in Yunnan province, and it provides a comparative perspective on the topic. On the absolute scale, Kunming’s kindergartens scored low on all seven aspects of the ECERS-R assessment, receiving “inadequate” on “program structure, activities, personal care, language - reasoning, space and furnishing, and interaction”, and a “minimum” on “parents and staff”. Compared to selected preschool programs in Bangladesh, Kenya, Brazil, and Canada, Kunming also has a lower performance in all dimensions measured. Within the city, kindergartens that received a government-issued rating on average had a higher score than those that did not, while kindergartens in the urban area of the city performed better than those in the rural area. The low ECERS-R rating of Kunming’s kindergartens is mainly a reflection of the poor quality of preschools in Yunnan. To a lesser extent, some cultural habits might also have contributed to the findings. First, participating in collective activities and obedience are valued as good behaviors in China, although attention to developing children’s independence is often lacking. Therefore, children in observed kindergartens in China often subscribe to the activities that teachers have planned for them, as indicated in low scores in items such as: Nap/Rest; Encouraging Children to Communicate; Discipline; Staff-Children Interaction; and Children Interaction, and so on. Second, insufficient educational material was found in most kindergartens, which suggests that there might be a lack of awareness of or attention to children’s personal needs in development. Inadequate resources could also be one possible cause. This was evidenced in low scores for items including Blocks; Sand/Water (for playing purposes); Dramatic Play; Use of TV/Computer; Books and Pictures; Furnishing for Relaxing and Comfort; and Space for Privacy. Third, health and safety practices and construction standards need substantial improvement, as reflected by low scores for items such as Meals/Snacks; Toileting/Diapering; Health 13 Practices; Room for Play; and Indoor Space. Furthermore, kindergarten teachers also receive insufficient support, as evidenced by low scores in indicators such as: Provision of Professional Needs of Staff; Staff Interaction; and opportunities for Professional Growth. In the context of these findings, the following policy suggestions might be appropriate: There is a need for more comprehensive measures to assess preschool quality that go beyond the traditional focus on hardware and other inputs and that place greater emphasis on curriculum, activities, and teacher-child interactions. ECERS-R is one such example and has been increasingly used. The newly published Learning and Development Guidelines for the 3-6 Year Olds by the Ministry of Education could serve as important guides for such an evaluation. Further, a well-designed teacher training program (even a short-term one) can produce positive outcomes on the quality of teacher-child interaction as measured by a similar scale. Preschool Teacher Recruitment, Career Development, Salaries, and Shortages Using a combination of teacher and director surveys and reviews of existing policies, statistics and other related studies, we examine key aspects related to preschool teachers, including recruitment, career ladder, teacher training and support, pay and compensation. Future demand for preschool teachers is also projected based on a few plausible scenarios and the latest population estimates. In the area of qualifications, recruitment and career ladder, preschool teachers have relatively low educational qualifications upon entry into the teaching force (senior vocational or high school), but most are able to improve their educational qualifications through in-service education (to junior college level). Preschools share the problem of teacher shortages and difficulty in hiring qualified teachers, albeit at different levels. In response, preschools often hire temporary or contracted teachers. The current career ladder mechanism is closely tied to the professional title system. Teachers are granted different levels of professional titles based on factors such as their academic background and teaching experience. Professional titles can be a key determinant of salary increase. With respect to teacher support, Kunming preschools surveyed reached a more favorable ratio of 15 as compared to 20 for the entire province. However, this ratio is still higher than recommended by related guidelines. In some rural and private preschools, the ratio reached is as high as 33 per teacher. There are a variety of professional development options for teachers, both within and outside the kindergarten. The formation of teaching and research groups is one such example. Preschool teachers are generally dissatisfied with their salary and compensation. In Yunnan, average preschool teacher salary is only about 60 percent of that of a primary school teacher— whereas in many other parts of the country and internationally, there is a trend in particular in developed countries such as Australia, Belgium, Ireland, Portugal, Sweden, and the United States, for preschool and primary school teachers to be paid at similar levels. Within Yunnan, salaries at public preschools are significantly higher than those at private 14 preschools. Moreover, public kindergartens in China have to subscribe to the government-issued staff quota system. The system regulates the number of employees each kindergarten can hire to enjoy salary payment from the government and full benefits. Teachers hired within the quota system receive higher salaries than their counterparts hired outside the system (such as long-term temporary teachers and those in private preschools). Assuming the most progressive pupil-teacher ratio target of 15:1 is achieved by 2015 with the current population projection, Yunnan will face a shortage of almost 60,000 new preschool teachers in 2020. Zhaotong County will face the highest teacher shortage (11,887) in 2020. According to research, the demand for senior vocational or tertiary education teachers translates into a demand of 6,642 teachers every year from 2013 to 2020— much higher than the current number of enrollment or graduates in the field. There are marked disparities between different types of preschools based on survey results. Teachers in public preschools in urban areas fare the best in terms of entry qualification, access to professional development, and salaries and other benefits. Teachers in rural private kindergartens appear to have the lowest entry qualifications, the least access to professional development opportunities and other career opportunities, and the lowest pay and compensation. There are several reasons that make it a challenging task to introduce and maintain a qualified teaching force, particularly in rural areas and in private preschools, including insufficient funds, lack of career development, and weak incentives. Policy and Program Implications The research and findings presented here represent the most recent and comprehensive set of studies ever done on early childhood education in Yunnan. While achievements in preschool expansion are apparent in Yunnan, we were able to reveal some serious challenges facing the preschool education. In light of the aforementioned findings, seven main policy and programming interventions are relevant to enhancing early childhood education in China and Yunnan. Increasing public financing on preschool education, in particular for the disadvantaged The incorporation of preschool education into the free public education system, targeted funding for preschool provision in poor, border and minority counties, and a diverse and innovative set of demand-side, private-public partnership financing and delivery mechanisms are potentially effective measures to achieve this policy goal. Innovative ways include using the surplus primary classrooms, partnering with private entities such as “Publicly owned and privately managed” (公办民营), satellite centers (around the central preschool at the county level), and the purchasing of spaces in private preschools (vouchers). Public-private partnerships in particular are worth exploring to diversify sources of funding and models of delivery and create markets with new providers for Early Childhood Development and Education (ECDE). Shanghai’s use of vouchers for migrant children attending private primary schools could offer valuable lesson. 15 Expanding service delivery models The formal, but expensive, center-based full time model of care is not the only viable model. Other models employing flexible service delivery could be adopted in addition to formal center-based care. Home-based care, parenting education, communication and media campaigns have been used in other countries to improve access to ECDE services, improve parental knowledge and practices, and ultimately improve childhood development outcomes. International experience shows that most children will have cognitive gains even if they are exposed to learning activities for only 15 hours a week. Shortening the time of center-based care can also reduce costs and accommodate more children. Alternatively, parenting education can be used to deliver curriculum-based information for implementation at home, and could be complemented by organized peer interactions. Within Yunnan, there are already good experiences from rural preschool centers which are being piloted. For example, the Child Development Research Foundation (CDRF) uses a service delivery model that relies primarily on volunteer teachers and provision of teaching and learning materials, flexible hours of operation, as well as renovation or establishment of the physical site. Save the Children also has community based ECD provision in a number of villages through provision of teaching learning materials, use of parent volunteer teachers, and general community support. Improving quality and atmosphere of existing preschools More sophisticated instruments need to be developed for assessing the quality of preschools, going beyond the traditional focus on hardware and other inputs but with more attention to curriculum, activities, teacher-child interactions, and other process indicators. ECERS-R is only one example of such instruments, and is being adopted more and more widely. Regular assessment of the relatively “weak” dimensions can be used as a management strategy and tool to improve the classroom environment and interactions. Feedback from such evaluations can be used by teachers and preschool directors to improve teaching and learning practices in the classrooms. Results from continuous assessment can also be used to inform the development of preschool teacher training programs. A pilot study in Chongqing city demonstrates that well-designed, short-term teacher training programs can have an impact on observed classroom behavior. Strengthening Family and Parental Education Our study reveals a serious gap in young children’s home environments, especially in rural, poor, and border counties. It is important to remember that families play at least an equally important role, if not more important role, in childhood development. Vast research evidence confirms that the lack of learning opportunities and poor quality caregiver-child interaction pose major risks for poor development. In addition to basic health and nutrition needs, substantial gains in childhood development require improvements in parenting, stimulation and early education. Unfortunately, our survey 16 results indicate a general lack of awareness of the critical importance of early stimulation, playing, and learning in overall childhood development. This trend needs to be reversed as a matter of urgency via various interventions that promote a positive home environment for young children. Parents can provide more quality stimulation and interaction with their children by talking, reading, and playing with them. Parents’ ability for caregiving can be enhanced through home visits, guidance and support from health providers, and group parental training. Increasing awareness of secondhand smoke’s harmful effects Given that approximately ninety percent of households in the survey have members smoke in the presence of children, it is crucial to increase public awareness of the negative health impact of secondhand smoke. At the community level, an extensive information campaign, involving displaying graphic warnings in public space, distributing information brochures to households, and giving presentations in public gatherings, could be an effective strategy. Media could also be a powerful communication channel. At the household level, families with pregnant women or children merit special attention. For example, health facilities that offer prenatal and or pediatric care could give out health education packages with information on secondhand smoke. For households in which family members other than the parents are the primary caregivers, it is important to design targeted measures so that they ensure the children are in a smoke-free environment. Developing a preschool teaching force The adequate provision of qualified preschool teachers presents the biggest challenge for Yunnan. Preschool teachers as a group have become marginalized from the rest of the teaching force. There is a need to improve preschool teachers’ pedagogical competency, as well as awareness of and sensitivity to childhood development needs. The shortage of qualified preschool teachers is particularly acute in rural and poor counties. The standards for preschool teachers in particular need to be reviewed to emphasize more teacher professional and pedagogical competencies with regard to teaching young children. There is a need to enforce the requirement of teaching certificates at hiring in particular for rural and private preschools. In-service training and support of preschool teachers should be highly prioritized in the next decade, and more funds should be directed to the development and training of both new and current teachers in order to meet the growing demand. The focus of training should be devoted more to the preschool pedagogy along the lines of the Learning and Development Guidelines. At the same time, there is a need to devise more innovative policy measures to attract more capable high school graduates into pre-service programs in both senior vocational and tertiary preschool education programs. In particular, efforts should be made to recruit more male preschool teachers, promote a more active admissions management, and 17 develop career counseling. Tuition waivers, more access to scholarships, and targeted job placement should be made available. Colleges can also consider establishing preschool education preparation centers and encourage job replacement for former primary school teachers and other unemployed graduates who are interested in preschool education. There is a need to review the preschool teacher career ladder and associated salaries, with gradual improvement so that eventually there will not be any difference in career opportunities and salaries between preschool and primary school teachers. Particular emphasis should be paid to improve the working conditions and pay for rural preschool teachers. Special incentives should be developed to retain rural preschool teachers, including more training programs, higher compensation, better accommodations, and improved support network. In addition, exchanges of teachers between rural, county and urban areas should be encouraged, and exchange services should be incorporated into routine teacher evaluations. For private preschools, it is important to establish formal recruiting, training, contracting and evaluating mechanisms for private and rural preschool teachers, and incorporate them under the management of public educational authorities. There may be merit to setting minimum teacher salaries for rural as well as private preschools. Moving towards integrated ECD provision for all, including early childhood development (0-3 Year Olds) At the very beginning of the research design, we had hoped the study would be able to cover early childhood development as an integrated service. However, it soon became clear that early childhood development as a concept has yet to emerge in Yunnan or in China as a whole. The provision of services for young children continues to be organized along sectoral boundaries of education, health, and child protection. The research team found that it was impossible to collect data or to have any meaningful conversation on “integrated services.” Among the various sectors relevant to young children, education is the largest department, so the team narrowed its focus to only early childhood education. Increasingly, global trends move toward more integrated provision of services for young children— bringing together prenatal, health, early stimulation/education, nutrition and child protection services— often under an umbrella child development agency. Educators, psychologists, developmental specialists, and economists are converging on the view that human development is a lifecycle development process that starts before the child is born. In the long term, Yunnan could start developing the ECD concept and related policies and institutions, covering prenatal services to services for 0-3 year olds, 3-6 year olds, and even aligning such services with primary school education. While this may still be a long-term goal, there are some areas that already deserve more policy attention by the government, for example, the care and development of the 0-3 year olds. This is an area that has been largely neglected. There is currently no public service provision. Parents use their own means either within the family or in the private market to 18 care for their 0-3 year olds. The private market is often unregulated and poses a potential development threat to young children. At the same time, labor force participation including female labor force participation is very high in China. Most of the young children are now left in the care of grandparents or other family members who may have relatively less education. There is a great urgency to develop appropriate policy interventions for the 0-3 year olds, especially those who are at risk. For the 0-3 group, emphasis should be on improving prenatal care for pregnant mothers, improving the nutritional and health status of young children through the provision of nutritional packages, parental and caregiver education, and early social and cognitive stimulation to young children. International experience shows the home-based and/or community-based approaches that facilitate interaction and training of mothers and caregivers by persons knowledgeable about nutrition, health and stimulation can be highly cost effective. 19 Introduction Yunnan is a medium-sized and relatively poor province on the southwestern border of China. The province is known for its abundant natural resources, including minerals, forests, animals, and plants. Yunnan has a population of 46 million, 30 million of whom reside in rural areas and 16 million in urban areas. Nearly 84 percent of Yunnan’s land mass is rural and mountainous. Yunnan further distinguishes itself with its ethnic diversity: one-third of its population, or 15 million people, belongs to 26 different non-Han ethnic groups. Its gross domestic product (GDP) per capita in 2010 was RMB 15,752— only slightly more than half the national average of RMB 29,992. The annual net income per capita of Yunnan’s rural population was only RMB 3,952 (US$623) in 2010. About 22 percent of its population, or 10 million people, still live under the national poverty line of RMB 2,300, making Yunnan the third poorest province in China after Gansu and Guizhou. At the end of 2011, Yunnan had 73 national-level poverty counties, the highest number among all provinces. There is a long history of collaboration between Yunnan province and the World Bank. Yunnan has been a beneficiary province in almost all of the World Bank poverty alleviation projects in the 1990s and early 2000s. In 2012, the World Bank financed a Technical and Vocational Education and Training Project amounting to US$ 50 million in support of the improvement of quality in nine secondary and tertiary technical and vocational education and training (TVET) institutions. In 2002, the Ministry of Finance and the Department of Education in Yunnan formally requested Bank support in conducting a review of early childhood education policies and programs in order to gain an in-depth and evidence-based understanding of the challenges the province faces in expanding early childhood education, especially to its rural and mountainous regions. The World Bank recognizes that ECD has important benefits for individuals and the society (see Box 1.0). The focus of early childhood education is completely in line with the World Bank Strategy 2020 which calls for “investing early, investing smartly, and investing in all”. The proposal was also in line with the World Bank’s overall mission and strategy of poverty alleviation. Accordingly, the Bank’s China Education Team embarked on raising funds, and designing and implementing a rather elaborate research agenda around early childhood education. The aim was to investigate key challenges, and to propose policy interventions for expanding ECD coverage in rural Yunnan. 20 Box 1.0: The Benefits of ECD ECD has important implications for children’s physical, cognitive, as well as socio-emotional development. Attendance in ECD programs can also have a positive impact on individuals’ educational attainment and productivity later in life. Furthermore, ECD brings significant social benefits, contributing to lower crime rate and social inequality. Given its extensive individual and social benefits, ECD proves to be a highly cost-efficient educational investment. Ample evidence supports that prenatal care and experiences in the first six years have a significant impact on the physical and brain development of children, therefore affecting the cognitive and socio-emotional development in subsequent stages of their lives. Neurological studies have showed that synapses (connections or paths between neurons in the nervous system) go through rapid development during early childhood, laying the foundation for cognitive and emotional functioning for the rest of the child’s life. Therefore, stimulation in the first 5 years of a child’s life is crucial to the brain formation and development process. For example, Hart and Risley (1995) show that children accumulate words at a very early stage in life. The differences in vocabulary are already evident for 36-month-old children from different social groups (hence different amount of language interaction between parents and children) and persist when those children turn 9 years old. Through ECD, children have the opportunity to participate in learning activities, work with others, and develop skills that are crucial to formal learning and social interactions in later years. Evidence shows that children’s exposure to early interventions correlates highly with their educational performance in later years. According to results from the 2009 Program for International Student Assessment (PISA), 15-year-old students in Shanghai who had attended pre-primary school for more than a year scored over 60 points higher than those who had not. In the United States, the high-quality, active-learning preschool program High/Scope Perry Preschool also had a significant impact on the participants. Those who enrolled in the program had higher rates of high school completion, higher monthly earnings, and higher rate of homeownership at age 27 than their counterparts who did not. (Schweinhart et al. 2005) ECD also yields long-term economic benefits. According to Carneiro and Heckman (2003), investing in early childhood development and education (ECDE) has higher rate of return than interventions targeted at later stages of life. Results from evaluations of high-quality ECD programs suggest that the real internal rate of returns ranges from 7% to 18%. Not only does ECDE contribute to higher earnings, 80% of the benefits of ECD programs were for the general public, including less high-risk behavior and fewer crimes committed. For example, in Turkey, children whose families participated in a mother-child education program had lower rates of delinquency than their peers whose families did not participate (Kagitcibasi et al. 2009). ECD can also contribute to reducing inequity. If well designed, ECD programs can decrease the disadvantage of poor children and allow them to catch up at an early age with respect to 21 cognitive and social development, thus providing a more level playing field. Several long-term evaluations of ECD programs for disadvantaged children in the United States have found benefits for adult educational outcomes, thus leading to greater equality of opportunity. Source: Alderman, A. 2011. No Small Matter: The Impact of Poverty, Shocks, and Human Capital Investments in Early Childhood Development. Washington, DC: World Bank. Carneiro, P. M., and J. J. Heckman. 2003. Human Capital Policy. Institute for the Study of Labor (IZA) Discussion Paper No. 821; Hart, B., and T. R. Risley. 1995. Meaningful Differences in the Everyday Experiences of Young American Children. Baltimore, Md.: Paul H. Brookes Publishing Company, Inc. Naudeau, Sophie; Kataoka, Naoko; Valerio, Alexandria; Neuman, Michelle J.; Elder, Leslie Kennedy. 2011. Investing in young children : an early childhood development guide for policy dialogue and project preparation. Directions in development ; human development; Africa regional educational publications. Washington, DC: World Bank. Schweinhart, L. J., Montie, J., Xiang, Z., Barnett, W. S., Belfield, C. R., & Nores, M. (2005). Lifetime effects: The HighScope Perry Preschool study through age 40. (Monographs of the HighScope Educational Research Foundation, 14). Ypsilanti, MI: HighScope Press. Wu, Kin Bing. 2011. Early childhood development and education in China: breaking the cycle of poverty and improving future competitiveness. Washington, DC: World Bank. Young, M.E. 2002. Investing in Our Children’s Future. From Early Child Development to Human Development. Washington, DC: World Bank. The research produced five interrelated studies: (i) a survey of existing ECD policies and institutions in Yunnan; (ii) the financing of preschools; (iii) a household survey of a representative sample of 3-6 year olds in rural Yunnan; (iv) a survey of preschool quality in Kunming using the internationally-recognized ECERS-R scale; and (v) a study of preschool teachers in Yunnan. Figure 1 below summarizes the overall research design. 22 Figure 1.0: Research Design Accordingly, this report is structured into five distinctive substantive parts: Part I reviews the main relevant policies, regulations, as well as institutions relevant to early childhood education. At the same time, Part I summarizes the major achievements in ECDE over the last decade. In the process, it highlights some of the policy and service delivery gaps in particular as they relates to access to preschool for the rural and disadvantaged population. Part II reviews the education financing system, especially as it impacts the financing of ECDE and details the fact that decentralized education financing, coupled with a preoccupation with compulsory basic education, has resulted in relatively inadequate and inequitable financing on ECD. Part III presents the most recent findings from the rural household survey on the home care conditions, developmental status of children in rural Yunnan, and the determinants of preschool enrollment and development. Part IV provides a snapshot of the quality of Yunnan preschools based on the application of ECERS-R in a stratified random sample of 24 rural and urban preschools in Kunming area. Part V is an in-depth study of the preschool teacher demand and supply, as well as recruitment, professional development, salary and other working conditions. Finally, the report offers policy and program implications based on the findings from the study. This report does not reiterate the importance of early childhood education as it has been argued most convincingly already by many educators, sociologists, economists, psychologists, and development specialists from both global and country perspectives. The most relevant and recent argument can be found in the World Bank publication on “Early Child Development in China: Breaking the Cycle of Poverty and Improving Future Competitiveness” (Wu, Young, and Cai, 2012). 23 Part I. Existing Policies on and Status of Early Childhood Education Legal and Institutional Framework In the past decade, China has been working toward completing its target of achieving the “Two Basics,” that is, achieving universal access to 9-year compulsory education, and eliminating illiteracy among young and middle-aged people. With this, the attention to preschool education on the national agenda has been growing more than ever before. Since the new millennia, new regulations and guidelines specifically for Early Childhood Education have been increasing. China’s early childhood education policies were guided by laws approved by the People’s Congress and by guidelines and national regulations as set forth by the State Council, in particular, the Ministry of Education (MOE). The Provincial Congress and government will then, based on their regional situations, pass specific regulations to monitor and implement service delivery. Table 1.1 outlines the major laws and regulations relevant to the operation of preschool education at the national level, and at the provincial level in Yunnan. Table 1.1 Selected Laws and Regulation on Preschool Education Laws approved by the National Congress Date Passed or Promulgated Education Law March 1995 Teacher Law October 1993 Private Education Promotion Law December 2002 Guidelines and Regulations issued by the State Council and Ministry of Education on Kindergarten Management Decisions on National Primary Education Reform and 2001 Development Guidelines on National Early Childhood Education Reform and 2003 Development Private Education Promotion and Implementation Guidelines February 2004 Suggestions on Current Development of National Preschool November 2010 Education National Medium- and Long-term Educational Reform and May 2010 Development Plan (2011-2020) The Guidelines on Chinese Children Development (2011-2020) August 2011 MOE Regulation on Kindergarten Management September 1989 MOE Working Principles on Kindergarten Operations March 1996 MOE Guiding Principles on Kindergarten Education 2001 MOE Regulation on Kindergarten Fee Collection December 2011 MOE Learning and Development Guidelines of 3-to-6-year-old September 2012 Children MOE Guiding Principles on Preschool Education Monitoring and February 2012 Evaluation 24 MOE Kindergarten Staffing Standards January 2013 Guidelines and Regulations issued by State Council Teacher Training and Management Regulation on National Teacher Qualification 1995 Regulation on Qualifications and Responsibilities of January 1996 Kindergarten Directors MOE Regulation on Post-service Training of Primary and September 1999 Secondary School Teachers MOE Pre-service Curriculum Standards of National Teachers November 2011 MOE Qualification Standards for Kindergarten Teachers December 2012 Regulations and guidelines set forth by Yunnan Provincial Government Yunnan Kindergarten Registration Policy June 1993 Yunnan Provincial Exemplary Kindergarten Management Scales 20106 Yunnan Provincial Preschool Education 3-year Action Plan May 2011 (2011-2013) Yunnan Provincial Medium- and Long-term Educational Reform May 2011 and Development Plan (2011-2020) Yunnan Provincial Regulation on Preschool Education November 2012 Yunnan Evaluation Scale for Nonprofit Private Kindergarten January 2013 Yunnan Provincial Regulation on Kindergarten Fee Collection Under development Yunnan Provincial Standards for Kindergarten Staffing Under development The Education Law and the Teacher Law promulgated in the mid-90s laid the systematic foundation for the legal status of education governance and teachers. These laws also served as the basis for further implementation-oriented guidelines and regulations in broad terms. In 2002, to supplement the public education services, the 9th National People’s Congress passed the Private Education Promotion Law to legally outline the operations of private education institutions. Together with the 2004 Private Education Promotion and Implementation Guidelines set by the State Council, these documents paved the way for private-government cooperation to maximize educational effectiveness and efficiency. The 1989 MOE Regulation on Kindergarten Management and 1996 MOE Working Principles on Kindergarten Operations were the first national legal documents on the management and operations of kindergarten services. These regulations set specific norms and standards for kindergarten administration, basic activities, safety, as well as health and hygiene, organization of personnel, classes, family communities and finance. At the beginning of the 21st century, as 85 percent of the Chinese population reached the “Two Basics” requirements, more emphasis than ever was put on non-compulsory preschool education. Since 2001, an increasing number of national regulations on kindergarten management and teacher qualifications were issued to guide the implementation of Early 6 The 2010 revised version is currently under review. The original version was published in 1997. 25 Childhood Education services.7 However, it was not until the late 2000s when the whole nation accomplished the “Two Basics” (including Yunnan Province in 2009), that a blossoming of new national and local support for ECE was realized. A series of mutually reinforcing and crossed–referenced guidelines on childhood development and preschool education were introduced including:  China’s Medium- and Long-Term Educational Reform and Development Plan (2011-2020) sets targets for the expansion of preschool education and looks to incorporate preschool education into urban planning for the next decade.  Suggestions on the Current Development of National Preschool Education provide concrete suggestions about how to expand preschool education, including strengthening teacher quality, ensuring budgetary appropriations at different levels of government, and guiding coordination between government departments.  The Guidelines on Chinese Children Development (2011-2020) explain the interventions, goals, and targets to expand preschool education in further detail, including the government planning process, the role of the private sector in supplementing the public education system (especially in rural areas), the ideal state of coverage, the mandate to fund enrollment, and the flexible nature of development for different geographical areas. To translate these guidelines into actions, the MOE within the State Council, and the Yunnan provincial government proposed concrete standards, regional targets and implementation plans.  The MOE Guiding Principles on Kindergarten Education and MOE Learning and Development Guidelines of 3-to-6-year-old Children outline the specific targets of childhood development during the preschool education stage along a variety of social and cognitive dimensions, providing instructions for kindergarten education, and monitoring and evaluation.  The MOE Guiding Principles on Preschool Education Monitoring and Evaluation provide further guidance on the administration, methods, and accountability in monitoring and evaluating preschool education.  The MOE Kindergarten Staffing Standards, updated from the 1987 version, provide specific requirements on the number of teaching staff, caregivers, and directors for established and new kindergartens of different types.  The Yunnan Provincial Preschool Education 3-year Action Plan (2011-2013) and Yunnan Provincial Medium and Long-term Educational Reform and Development Plan 7 In 2001 and 2003 respectively, the National Decision on Primary School Reform and Development and the Guidelines on Early Childhood Education Reform and Development recommended that the poorer provinces (those covering 15 percent of the population not achieving the “two Basics”) should “work hard towards the fulfillment of ‘ Two Basics’, moderately develop Senior Middle education, and actively promote preschool education”, and “in 5 years (2003-2007) the gross enrollment of 3-6-year-olds at kindergarten should achieve 35 percent, while the 5-6-year-old gross enrollment rate should reach 60 percent ”. 26 (2011-2020), which set the regional education targets, acknowledge the challenges of current supply and demand for rural kindergartens, reaffirm the importance of preschool education as a priority for the provincial government in the next decade, and promulgate an “annual major increase” in budget appropriations to support the expansion.  The Yunnan Provincial Regulation on Preschool Education was the newest detailed provincial regulation on the administration of preschool education since 1993, which outlines the structure of responsibilities across government departments, location and management of facilities, principles of child care, education and teacher training, as well as support measures for private kindergartens and rural populations. These implementation-oriented measures and plans shine a light on the new era for the development of preschool education in China and Yunnan Province in particular. Yunnan has the most counties under the national poverty standards, 8 as well as the largest concentration of ethnic minorities. The goals are to break the poverty cycle, increase social harmony, and improve future competitiveness. In the last years of China’s 11th 5-year plan (2006-2010), a national fund of 2 billion RMB ($328 million) was allocated to support pilot preschool expansion projects 9 in central-western provinces (of which Yunnan Province received 300 million RMB). In 2011, an unprecedented 50 billion RMB ($8.2 billion) central budget was allocated to support the expansion of preschool education under the 12th 5-year plan (2011-2015). By 2012, in addition to 200 million annual special appropriations from the Yunnan provincial government, about 1.74 billion RMB was used in Yunnan to support kindergarten construction, private participation, training of teachers and poor family subsidies (Yunnan Provincial Government 2011, 2013). In 2012, the Ministry of Education issued Learning and Development Guidelines for 3-6 Year Olds which describe children’s development stages and suggest appropriate education interventions in health, language, society, science, and art. Efforts are now underway to develop a specific early childhood education law in China. Administration and Service Delivery To implement these policies, a tertiary level of administration and coordination currently exists in China. At the central level, the National Working Committee for Children and Women under the State Council is the highest body coordinating ministry responsibilities. It leads the promotion of women’s and children’s benefits by planning, coordinating and monitoring the National Plan of Action for Children, and the National Plan of Action for Women. Parallel line ministries are responsible for national policy-making and supervision. In particular, the Ministry of Education sets standards, drafts regulations and oversees the 8 According to the National Group Office of Poverty Alleviation and Development, 73 out of 592 National Poverty Alleviation and Development focal counties were in Yunnan as of March 2012. http://www.cpad.gov.cn/publicfiles/business/htmlfiles/FPB/fpyw/201203/175445.html 9 Interviews with staff of the Ministry of Finance, the Ministry of Education, and Development and Reform Commission on National Preschool Education Projects (教育部、发改委、财政部有关负责人就学前教育三年行 动计划和国家学前教育重大项目答问), 2011. http://www.mof.gov.cn/zhengwuxinxi/zhengcejiedu/2011zhengcejiedu/201109/t20110928_597049.html 27 implementation of preschool education for children aged 3-6. On the provincial level, the Bureau of Education sets local implementing policies affecting all preschool children, and usually operates and finances training sessions for local kindergarten teachers. Together with the Department of Children and Women, the Bureau of Health as well as other line bureaus, the Bureau of Education also oversees service delivery at the county level. Despite the large central and provincial financial transfers, service provision and financing are primarily the responsibilities of the county government. They oversee the operation of kindergartens, manage their own budgets and generate other extra-budgetary resources in supporting the mission. Table 1.2 shows the distribution of responsibilities toward the expansion of preschool education during the 12th 5-year plan at the Yunnan provincial level. Table 1.2 Selected Expansion of Preschool Education in Yunnan Bureau of Bureau Development Bureau Bureau Bureau Education of & Reform of of of Finance Committee Human Housing Health Resource Financing of new    Kindergartens Subsidy for poor families   Teacher (principal) training     Assessing kindergarten   quality Incorporate preschool   expenditures into routine budgets (Re)Construction of    kindergartens Incorporating preschool   education into urban planning Preschool teacher   retirement support Setting policy for    kindergarten fee collection Setting policy for    kindergarten staffing Childcare monitoring and   instruction Note: : Lead Authority; : Participating Authorities 28 Source: Yunnan Provincial Government (2011) 29 Main Trends in Early Childhood Education in Yunnan Across China over the last decade, there has been impressive expansion at the preschool education level. By 2011, official statistics indicate a GER of 54 percent in China, which places China at the middle of all East Asian countries. Within the region, Korea, Thailand, New Zealand, Australia and Malaysia all have a much higher GER at the preschool level. The latest estimated by the Organization for Economic Co-operation and Development (OECD) indicates that over 90 percent of children at the age of three or above attend some form of preschool. Figure 1.1 2011 Gross Enrollment Rate (GER) in selected Asia and Pacific Countries Source: Junko Miyahara, Regional ECD Profiles, Slide 12, 2013 Yunnan’s GER is low by national comparison, with major internal disparities An estimated regional GER in China indicated that western provinces have a lower enrollment rate as compared to developed eastern provinces, where Yunnan lagged even further behind. There was however a slow catching-up for poor provinces in recent years (Figure1.2)10. Within the province, there were also wide disparities in the estimated enrollment rates11, with the highest in the capital city (Kunming) at 95 percent to the lowest of 33 percent in Zhaotong, a mountainous prefecture bordering Sichuan and Guizhou. Eight out of sixteen prefectures were lower in GER than the provincial average in 2011; six of them were 10 The estimated GER accounted for enrollment to both kindergartens and one-year pre-school classes as shown in national statistics. The population of 3-6 year-old children was calculated based on annual birth rates and demographic surveys. Therefore it is not exactly equal to the Kindergarten GER of three years before primary schools, but slightly higher. For estimates based on only kindergarten enrollment, please refer to Appendix. 11 These rates also include one-year preschool enrollment. 30 primarily inhabited by ethnic minorities (Table 1.3). Figure 1.2 Gross Enrollment Rate (est.) of Preschools, by Region, 2004-2010 75 72 70 66 64 62 Gross Enrollment Rate (%) 65 59 57 57 63 60 54 51 53 55 49 51 50 55 50 49 50 55 45 45 47 43 44 49 40 44 35 40 30 37 33 34 25 20 2004 2005 2006 2007 2008 2009 2010 National Average Eastern Provinces Western Provinces Yunnan Sources: Author’s estimate based on NBSb 2000 – 2011, MOEa 2010 Note: MOE= Ministry of Education; NBS=National Bureau of Statistics Table 1.3 Estimated Gross Enrollment Rate in Yunnan, by Prefecture, 2011 Yunnan 62.1% 昭通 Zhaotong 33.0% 迪庆 Diqing* 35.7% 怒江 Nujiang* 37.7% 普洱 Pu'er* 41.3% 临沧 Lincang 47.4% 文山 Wenshan* 51.6% 楚雄 Chuxiong* 56.9% 丽江 Lijiang* 61.7% 红河 Honghe* 67.1% 西双版纳 Xishuangbanna* 67.6% 德宏 Dehong* 68.4% 保山 Baoshan 73.5% 大理 Dali* 74.4% 玉溪 Yuxi 78.4% 曲靖 Qujing 79.8% 昆明 Kunming 94.9% *Primarily populated by ethnic minorities Sources: Author’s estimate based on SBY 2006 – 2009, NBSb 2012 Note: NBS=National Bureau of Statistics; SBY=Statistical Bureau of Yunnan 31 Share of private enrollment has been increasing In direct contrast with the basic education sector where the majority enrollment is in the public sector, early childhood education in China is characterized by a relatively high share of private enrollment. Nationwide, in 2012, 41 percent of preschools were privately-owned, accounting for 34 percent of total enrollment and 40 percent of total teachers (Table 1.4). In general, private enrollment has been increasing slowly in both the share and number for the past decade (Figure 1.3). In Yunnan Province, private ownership accounted for 42 percent of preschools, 26 percent of preschool enrollment and 37 percent of total teachers. A similar increasing trend of private enrollment existed for the past decade in Yunnan as well (Figure 1.4). This indicates that in both Yunnan and China, private preschools on average tend to be smaller in size and have lower pupil-teacher ratios as compared to public preschools. Table 1.4 Share of Private and Public Preschools, Enrollment and Teachers, 2012 Share of Share of Share of Schools Enrollment Total Teachers National Private 40.7 33.5 39.6 National Public 59.3 66.5 60.4 Yunnan Private12 41.8 26.0 37.1 Yunnan Public 58.2 74.0 62.9 Sources: NBSa 2013, MOEb 2010. Note: NBS=National Bureau of Statistics; MOE= Ministry of Education. 12 Yunnan’s private and public preschool numbers and enrollments are derived from 2009 data. 32 Figure 1.3 Share of Enrollment in Private Preschools in China, 2005-2012 60 Number of Enrollments (millions) 50 40 30 20 10 33% 33% 30% 32% 23% 26% 27% 28% 0 2005 2006 2007 2008 2009 2010 2011 2012 National Private National Public a a Sources: NBS 2011-2013, MOE 2006-2010 Note: NBS= National Bureau of Statistics; MOE= Ministry of Education. Figure 1.4 Share of Enrollment in Private Preschools in Yunnan, 2003-2009 1400 Number of Enrollments (thousands) 1200 1000 800 600 400 200 26% 19% 21% 23% 12% 15% 16% 0 2003 2004 2005 2006 2007 2008 2009 Private Public a b Sources: MOE 2003-2009, MOE 2003-2009. Note: MOE= Ministry of Education. One-year pre-primary classes are prevalent in rural areas of Yunnan In China, early childhood education for 3-6-year-olds consists mainly of a three-year regular kindergarten program and a one-year pre-primary class. While the regular three-year kindergarten program is often financially independent, and features a development-oriented curriculum, the one-year pre-primary school class bears a close resemblance to grade one elementary school education. This type of provision started to appear in urban cities in the 33 1980s to alleviate the rapidly growing demand for preschool education. As a result, one-year pre-primary class has a more rigid focus on management and course structure. The curriculum often targets at pre-defined learning content such as Pinying (拼音), addition and subtraction, and English vocabulary. Some preschool children even need to submit homework for school. This learning model is widely believed by child development experts to disengage the learning interest and creativity of young children, and is therefore developmentally inappropriate for them. By 2010, after a continuous drop in the number of one-year pre-primary classes and the expansion of kindergartens nationwide,13 the share of one year pre-primary class enrollment had dropped to 14 percent (Figure 1.5), and among them 96 percent were primary school-attached (Figure 1.6). With the increasing number of mandates to forbid one-year pre-primary classes in developed cities and provinces in the late 2000s, it is likely that the vast increase of kindergarten classes was driven by the major decrease of independent pre-primary classes in 2010. Figure 1.5 All Preschool Classes in China, 2003-2010 1200 Number of classes (thousand) 1000 800 600 400 200 45% 42% 39% 37% 35% 32% 28% 14% 0 2003 2004 2005 2006 2007 2008 2009 2010 All kindergarten classes All pre-primary classes a Source: MOE 2003-2010. MOE= Ministry of Education. 13 As Zhang (2009) summarized, the decrease in the number of pre-primary classes was a result of the combined effect of increasing access to kindergartens (especially in cities), as well as the awareness of the negative impact of “Xiaoxue Hua”, an inappropriate practice to teach preschool children in the same way as primary school students are taught. 34 Figure 1.6 One-year Pre-primary Classes in China, 2003-2010 Number of classes (thousand) 350 300 250 200 150 100 50 0 2003 2004 2005 2006 2007 2008 2009 2010 Pre-primary classes (independent) Pre-primary classes (primary school-attached) a Sources: MOE 2003-2010. Note: MOE= Ministry of Education. Despite the recent improvements in access to kindergartens, different regions still have greatly distorted distributions. Figures 1.7 and 1.8 illustrate the share of one-year pre-primary classes in all provinces across rural, county and urban areas. In the developed eastern and coastal provinces14, the prevalence of one-year pre-primary classes as a share of total preschool enrollment is only one-third of the national average (5.4 percent), whereas in the western provinces15 it is almost twice as high (26 percent). In Yunnan province in particular, one out of three entrants was enrolled in one-year pre-primary classes. The imbalance is even greater when distinguishing among rural, county, and urban areas. Regarding the national average, more than one-fourth of the rural enrollment is through one-year pre-primary classes— 10 times higher than the average in urban area (2.6 percent) and 3 times higher than county areas (8.7 percent). In rural Yunnan however, this rate is as high as 56 percent, one of the highest in the country (but only below Tibet, Guizhou, Ningxia and Guangxi as illustrated in Figure 1.7) and almost 5 times as much in the rural areas of the eastern provinces. Given that almost all one-year pre-primary classes were attached to primary schools by 2010, this distribution implied rural children (especially in Yunnan) were much less likely to attend quality preschool services as compared to children living in county and urban areas. 14 Including Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan. 15 Including Inner Mongolia, Guangxi, Chongqin, Sichuan, Guizhou, Yunnan, Tibet, Shanxi, Gansu, Qinhai, Ningxia, and Xinjiang. 35 Figure 1.7 Enrollment in One-year Pre-primary Classes as a Percentage of Total Preschool Enrollment in Cities, Counties and Rural Areas, by Province, 2010 90 % of Pre-primary l classes in total Preschool enrollment Tibet Guizhou Ningxia 80 70 Guangxi 60 Yunnan 50 40 30 20 10 Beijing 0 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 GDP Per cappita (RMB) Rural County City a Source: MOE 2011. Note: MOE= Ministry of Education; RMB= renminbi. Yunnan’s rural share of preschool enrollment is one of the highest in the country There were over 150,000 kindergartens (including preschool classes) in China as of 2010, with half of them located in rural areas (Figure 1.9). While the number of kindergartens and enrollment have been growing in recent years, the share of children enrolled in rural areas had declined by about 6 percent since 2003 (Figure 1.10-1.11). This reflects the fact that most new enrollment was not realized in rural areas. Compared with the national average of 41 percent, the share of rural enrollment in Yunnan Province still held half of total enrollment by 2010, that is, the 5th highest in the country and second only to Xinjiang (58 percent) in all of the western provinces.16 Considering the much higher proportion of one-year primary-school-attached pre-primary classes rather than independent kindergartens in rural Yunnan, this could imply a much smaller population in rural Yunnan as compared to other provinces that had quality access to preschool services. 16 The provinces that have over 50 percent share of total preschool and kindergarten enrollment in rural areas are Hebei (62 percent), Xinjiang (58 percent), Jiangxi (55 percent), Henan (54 percent), Shandong (52 percent) a and Yunnan (50 percent).Source: MOE 2011. 36 Figure 1.8 Enrollment in One-year Pre-primary Classes as a Percentage of Total Preschool Enrollment in Cities, Counties and Rural Areas, by Regions, 2010 60 55.5 % of Pre-primary Classes in Total Preschool 50 43.9 40 32.8 30 26 27.1 20 14.7 15.5 11.2 11.6 8.7 10 7.1 6.9 5.4 2.6 3.4 Enrollment 0.4 0 Overall Cities Counties Rural Country Average Yunnan Province Western Provinces East Coastal Province a Source: MOE 2010. Note: MOE= Ministry of Education. Figure 1.9 Number of Kindergartens in China, 2003-2010 80 70 Number of Kindergartens (thousands) 60 50 40 30 20 10 0 2003 2004 2005 2006 2007 2008 2009 2010 Cities Counties Rural a Source: MOE 2003 – 2010. Note: MOE= Ministry of Education. 37 Figure 1.10 Number of Students in Kindergartens and One-year Pre-primary Schools in China, 2003-2010 35 30 Number of Students (millions) 25 41% 20 42% 44% 43% 47% 46% 15 47% 48% 10 5 0 2003 2004 2005 2006 2007 2008 2009 2010 Cities Counties Rural (Total enrollment %) a Source: MOE 2011. Note: MOE= Ministry of Education. Figure 1.11 Number of Students in Kindergartens and One-year Pre-primary Schools in Yunnan, 2003-2010 1.2 1 Number of Students (millions) 0.8 50% 53% 52% 53% 0.6 53% 59% 57% 60% 0.4 0.2 0 2003 2004 2005 2006 2007 2008 2009 2010 Cities Counties Rural (Total enrollment %) a Source: MOE 2011. Note: MOE= Ministry of Education. 38 Figure 1.12 Number of Students in Kindergartens and One-year Pre-primary Schools in Western Provinces, 2003-2010 9 8 Number of Students (millions) 7 43% 6 44% 45% 5 47% 46% 54% 52% 52% 4 3 2 1 0 2003 2004 2005 2006 2007 2008 2009 2010 Cities Counties Rural (Total enrollment %) a Source: MOE 2011. Note: MOE= Ministry of Education. Declining pupil-teacher ratios and inequality in supply of trained teachers in rural areas Another dimension that reveals inequity in access to preschool services was the distribution of teachers. From 2003 to 2010, the supply of teachers17 had been growing at a faster rate than preschool enrollment as a whole (Table 1.5). Most of them (over 95 percent in 2010) had a degree equivalent to junior vocational or higher. The number of teachers with proper training18 had therefore followed the same trend. It is noteworthy that the increase of trained teachers in rural areas had been growing at an even faster rate, especially in Yunnan and other western provinces. These factors had given rise to a steady decline of pupil-teacher ratios in most provinces according to recent statistics (Figure 1.13) - but there is still great space for improvement to reach the national standards.19 17 Including principals, head teachers and assistant teachers (caregivers), not including substitute or part-time teachers. 18 Defined as teachers with at least junior vocational or equivalent education. 19 The pupil-teacher ratio (教职工与幼儿比) is defined as the number of enrolled children divided by the number of kindergarten teaching staffs (including full-time teachers, caregivers, healthcare and administrative personnel, and other supporting staffs – not including part-time teachers). Based on the newly promulgated MOE Kindergarten Staffing Standards (KSS, 幼儿园教职员工配备标准) in 2013, this ratio for full-day operated kindergartens should be no more than 9:1, with each kindergarten class staffing two fulltime teachers and one caregiver. Another benchmark is the pupil-teaching staff ratio (保教人员与幼儿比), which only includes fulltime kindergarten teachers and caregivers. As was also regulated by the KSS, the ratio is expected to be no more than 7:1. 39 Table 1.5 Annual Growth Rates of Teachers and Enrollment, 2003-2010 (%) Average Growth rate Growth rate in Growth rate in growth rate in number of number of trained number of trained of enrollment teachers teachers rural teachers Yunnan 4.9 10.2 10.2 18.4 Western 5 9.4 9.5 13.6 Provinces Eastern 6 8.4 8.2 8.5 Provinces National 5.8 9.6 9.4 12.4 Average a Source: MOE 2004 – 2011 Note: MOE= Ministry of Education. Figure 1.13 Pupil-Teacher Ratio, by Region, 2003-2010 35 Number of Students per Teacher 30 25 20 15 10 2003 2004 2005 2006 2007 2008 2009 2010 Eastern Provinces National Average Western Provinces Yunnan a Source: MOE 2003 – 2010. Note: MOE= Ministry of Education. However, there was still a stark gap in pupil-teacher ratio between provinces, as well as among rural, county and urban areas. Figure 1.14 plots pupil-teacher ratios with provincial GDP-per capita in 2010, which shows a clear negative relationship. According to the newly-published MOE Kindergarten Staffing Standards, each full-time teacher or caregiver (保育员) is ideal for taking care of 7 to 13 children depending on the nature of kindergarten, because large classrooms are less efficient for child development. Under these new standards, it is clear that most provinces will need to improve the supply of teachers greatly in order to meet the requirements set for the next three years (2013-2015). For provinces 40 having high ratios at the time (such as Guizhou, Guangxi, Xinjiang, and Yunnan, and so on), this would require significant investment in teacher training and support. Figure 1.15 depicts the imbalance of enrollment and access of trained teachers in rural, county and urban areas. In Yunnan, with 50 percent of total preschool enrollment in rural areas, the percentage of trained rural teachers accounted for only 22 percent of all trained teachers in 2010. Meanwhile, 16 percent of enrollments in urban areas and 34 percent of enrollments in counties employed respectively 29 percent and 48 percent of all trained teachers. Despite the fact that the insufficient number of rural trained teachers was a national phenomenon, compared to other regions, the intensity of contrast in Yunnan was still among the greatest in all China. Figure 1.14 Pupil-Teacher Ratios in Kindergartens, by Province, 2010 35 Guizhou 30 Number of Students per teacher Xinjiang 25 Guangxi Anhui Hebei Yunnan 20 Fujian Gansu Jiangsu Shandong 15 Inner Mongolia Tianjin Shanxi Guangdong Hainan Jilin Zhejiang Shanghai 10 Beijing Liaoning 5 0 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 GDP percapita (RMB) Source: MOE 2011. Note; MOE= Ministry of Education. 41 Figure 1.15 Share of Enrollment and Trained Teachers in Rural, County and Urban Areas by Region, 2010 Enrollment (E.Provinces) 36% 34% 31% Trained Teachers (E.Provinces) 23% 33% 44% Enrollment (W.Provinces) 43% 36% 21% Trained Teachers (W.Provinces) 23% 40% 37% Enrollment (Yunnan) 50% 34% 16% Trained Teachers (Yunnan) 22% 48% 29% Enrollment (National) 47% 26% 27% Trained Teachers (National) 25% 35% 40% 0% 20% 40% 60% 80% 100% Rural County Urban a Source: MOE 2011. Summary and Discussion China does not yet have a specific early childhood education law. However, it has developed a rather comprehensive set of guidelines and regulations concerning early childhood education. Much of these efforts occurred during the last decade. It is apparent that after the achievement of compulsory basic education, the Chinese government, including the Yunnan Province, is spearheading policy development for early childhood education. A review of official statistics in the last decade reveals a few important trends in early childhood education in China as well as in Yunnan. Early childhood education experienced a rather impressive growth rate within just the last few years, with estimated GER for both kindergarten and one-year pre-primary classes increasing from 49 percent in 2004 to about 63 percent in 2010. While Yunnan experienced a similar rate of improvement in access to early childhood education, the province lags behind the national average as well as the other western provinces in the overall level of preschool coverage. Its GER of 55 percent in 2010 is lower than the national average of 63 percent as well as the average for western provinces. Disparities are also very significant within Yunnan, as GER ranges from 95 percent in the capital of Kunming city to only 33 percent in Zhaotong county. Both in China and in Yunnan, the share of private preschool enrollment has been increasing. About 42 percent of preschools in Yunnan are private, absorbing about 26 percent of 42 student enrollment. This picture is rather similar to the national trend where about 41 percent of preschools are private, and absorb about 34 percent of total enrollment. Another important feature is that a higher proportion of Yunnan’s preschool enrollment is actually in one-year pre-primary class as opposed to the regular three-year kindergarten programs. Almost one in every three children enrolled in preschool in Yunnan is in the one-year program as compared to the national average of 15 percent, and the western average of 26 percent. In rural Yunnan, the proportion of enrollment in the one year program reaches a high of 56 percent, further highlighting the relatively lower development status in rural Yunnan. Finally, the supply of preschool teachers has been growing at a faster rate than overall enrolment to result in the favorable effect that there has been a dramatic decrease in pupil-teacher ratio across the nation. In Yunnan, pupil-teacher ratio experienced a decline from about 30 in 2003 to 20 in 2010. However, the distribution of qualified teachers is skewed toward urban areas. Whereas 50 percent of total enrollment is in rural areas, only 22 percent of all trained teachers are serving in rural areas. 43 Part II. Financing Early Childhood Education The disparity in access to early childhood education services across different regions can be traced back to the very nature of China’s decentralized education financing system. Although the country is highly centralized in terms of policy and regulations, it is rather decentralized in local administration and finance. This section outlines the composition of educational finances in preschool services in China in general and the Yunnan province in particular. Structure of Educational Finance China’s total educational funds are categorized into governmental appropriations and non-governmental funds. Despite the differences among educational levels, about 80 percent of all educational funds available in China are channeled through the government. Private investment and public donations on the whole only account for less than 1 percent of total finances, while tuition and other school revenue account for about 19 percent (Figure 2.1). According to national statistics, total government allocations in education increased to 4 percent of national GDP in 2012, reaching the 4 percent threshold goal called for by the Educational Law. Given the country’s fast economic growth rate and equally fast-growing governmental revenues, this translates into a real-term increase in both aggregate and per student base in all subsectors of education. Figure 2.1 Composition of National Educational Funds (all educational subsectors), 2011 Other Budgetary appropriations educational 3% funds 90% Government Tuition and other educational school revenue appropriations 19% 78% Other extra- Taxes and fees 9% Private budgetary funds investment and 1% Public donation 1% c Sources: MOE 2011. Note: MOE= Ministry of Education. 44 There are two components under the government’s educational appropriations. The annual allocation of budgetary funds covers the fundamental needs for teacher compensation, new construction projects, research and other purposes. Meanwhile, local government also acquires extra-budgetary funds through educational taxes and fees. The average share of extra-budgetary allocations takes about 10 percent of the total of the government’s educational funds, but it is also unevenly distributed among different educational levels. Figure 2.2 illustrates the distribution of enrollment by levels of education. Table 2.1 summarizes the key statistics of the financing structure at the different levels of education. Figure 2.2 Distribution of Students by Level of Education, 1980-2010 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1980 1985 1990 1995 2000 2005 2010 Preschool Primary Junior Secondary Senior Secondary Vocational Higher Education a 20 Sources: MOE various years. MOE= Ministry of Education. 20 Special Education was not included in the graph due to its extremely small size in budget allocation share, it is however calculated in the national statistics. 45 Table 2.1 Distribution of Enrollment and Educational Funding, by Level of Education 2011 (in %) Enrollment Budgetary All Non-governmental (2010) educational governmental funds as a share of total funds appropriation available funds (e.g. tuition and private investment) Higher education 11.5 22.8 22.0 41.7 Senior vocational 8.3 6.2 6.8 23.2 Senior middle 9.4 9.2 9.7 27.8 Junior middle 19.4 21.0 21.0 6.3 Primary education 38.3 31.6 31.0 4.2 Special education 0.2 0.4 0.4 3.0 Preschool education 13.0 2.1 2.2 59.2 Others and -- 6.7 6.8 11.3 administration a c Source: MOE 2010, MOE 2011. By comparison, there is a strongly skewed distribution of budgetary educational funds toward higher education in China, where about 12 percent of total enrollment accounts for 22 percent of total government educational appropriations. In addition to government appropriations, the non-governmental share of total funds in higher education amounted to 42 percent in total. This means that aside from a larger share of budgetary support, higher education in China is also heavily financed through other means of social support (tuition, donations, and private sector investment, and so on). By contrast, preschool education accounts for 13 percent of total enrollment, but only commands 2 percent of the national educational budget. A staggering 59 percent of total educational funds come from non-governmental sources, which implies a much heavier financial reliance on private investment. In comparison to other countries, China’s educational budget for preschool education is also at the lower end of the spectrum. In Sweden, 10% of public expenditure on education went to pre-primary level in 2010. Similarly, pre-primary education enjoyed 7% and 6% of Thailand’s and Finland’s public expenditure, respectively.21 Given the importance of the long-term benefits of early childhood education, this lack of governmental support would probably result in access inequities and long-term development disparities between wealthy and under-developed areas. Financing Preschool Education Total spending on preschool education22 China has long placed its strategic educational emphasis on achieving universal 9-year compulsory schooling. In Yunnan, this objective was accomplished in late 2009. With 21 UNESCO Data Centre 22 The sources of total spending on preschool education include: government educational appropriations, private investment and public donation, tuition and other school revenue, and other appropriations. 46 numerous national guidelines on promoting preschool education already in place since the last decade, Yunnan is on the frontier of the national agenda to expand and catch up on preschool enrollment and facilities. In 2011, 6.4 percent of provincial GDP was invested in education compared to 4.9 percent of western provinces and 3.1 percent of eastern provinces. Yunnan’s total educational funding for the preschool sector has been increasing at an annual rate of 32 percent from 2005 to 2011, while total budgetary funds increased by 22 percent each year. From 2007 to 2011, total spending per student also increased from 1,732 RMB to 2,750 RMB (13 percent annually) per year. The following graph (Figure 2.3) and table (Table 2.2) compare governmental appropriations and budgetary spending in China in general and in Yunnan specifically. Figure 2.3 Composition of Preschool Education Funding, China and Yunnan, 2011 Funding Yunnan's Preschool Funding China's Preschool Education, Education, 2011 2011 Other Other appropriati appropriati on on 1% 2% Budgetary Tuition and Governmen educational Tuition and Budgetary other t other Governmen funds,96% educational school educational school t funds revenue appropriation revenue educational 85% 48% 46% appropriation 54% 41% Taxes &fees 12% Taxes &fees 2% Other extra- Other extra- Private Public budgetary Private budgetary Public investment donation funds investment funds donation 3% 0.1% 3% 5% 2% 0.5% c Source: MOE 2011. Table 2.2 Education Funding and Preschool Education in China, by Region, 2011 (%) National Yunnan Western Eastern Average Province Provinces Provinces GDP % in education 3.9 6.4 4.9 3.1 GDP % in preschool education 0.09 0.11 0.11 0.08 Government % in educational funding 77.9 85.9 82.4 76.0 Government % in preschool educational 40.8 46.0 50.8 40.1 funding Private investment % in educational funding 0.5 0.3 0.4 0.4 Private investment % in preschool 2.9 4.8 3.2 2.5 educational funding Tuition % in educational funding 13.9 9.2 10.7 14.7 Tuition % in preschool educational funding 48.4 34.5 36.4 50.1 Extra-budgetary funds as a share of 9.6 7.4 6.6 12.7 educational appropriations Extra-budgetary funds as a share of 15.4 3.6 11.9 17.1 47 preschool educational appropriations c Source: MOE 2011. Per-student expenditure on preschool education Yunnan fell short in terms of both total23 and budgetary preschool education spending per student. The overall spending per child (RMB 2,750) in 2011 is only half of the average for all 12 western provinces (RMB 5,319), and one-third of the average for coastal provinces (RMB 8,076), as shown in Figure 2.4. Budgetary spending per student follows the same pattern. Nearly three-quarters of budgetary spending in Yunnan was used on personnel expenditures (Figure 2.5), which includes primarily teacher salaries and benefits. Less than one out of five RMB was invested in daily operations of services, which spans from usage of supplies to land acquisition. Figure 2.4 Total Spending per Student on Preschool Education, by Type and Region, 2011 Eastern Provinces 55% 43% 2% Western Provinces 47% 38% 14% Regions Yunnan 66% 30% 4% National Average 54% 43% 3% 0 2,000 4,000 6,000 8,000 10,000 RMB Personnel Expenditure Operational Expenditure Construction Expenditure c Source: MOE 2011. 23 The sources of total spending per student include: government educational appropriations, private investment and public donation, tuition and other school revenue, and other appropriations. 48 Figure 2.5 Budgetary Spending per Student on Preschool Education, by Type and Region, 2011 Eastern Provinces 61% 37% 3% Western Provinces 51% 30% 19% Regions Yunnan 74% 19% 7% National Average 63% 32% 5% 0 1,000 2,000 3,000 4,000 5,000 6,000 RMB Personal Expenditure Operational Expenditure Construction Expenditure c Source: MOE 2011. It should also be noted that non-governmental funds for preschool education (tuition and fees) have been primarily directed toward the expansion of operations rather than personnel costs of preschool services. This is a nationwide phenomenon across regions, even though areas that have larger shares of tuition and smaller numbers of children (for example, in the eastern provinces) would benefit more than those that rely more heavily on governmental appropriations. Figure 2.6 illustrates the vast disparities of per student budgetary and total preschool funding across provinces. 49 Figure 2.6 Budgetary and Total Spending per Student on Preschool Education in China, by Province, 2011 20000 Beijing Shanghai 18000 Per student expenditure (RMB) 16000 Shanghai 14000 12000 Beijing 10000 Eastern Provinces 8000 Western Provinces 6000 National Average 4000 Yunnan 2000 0 -10000 10000 30000 50000 70000 90000 GDP per capita (RMB) Total Spending on Preschool Education Budgetary Spending on Preschool Education c Source: MOE 2011. In 2010, the per-student personnel expenditures24 in Yunnan reached 1,286 RMB, a growth rate of 4.6 percent from 2009 levels. 25 However, Yunnan’s per-student personnel expenditures still lag behind; at 886 RMB, they are lower than the national average. They are also 390 RMB lower than the average of the western regions. Among western provinces, per-student personnel expenditures in Yunnan are at the lower end of the spectrum: about ¼ of Inner Mongolia’s level, and 1/3 that of Qinghai. These differences reflect the relatively low income levels of kindergarten teachers in Yunnan. 24 Personnel expenditures include two parts: “wage and benefits expenditures” and “individual and family subsidies expenditures.” Wage and benefits expenditures reflect the various types of remuneration schools/firms pay to active and temporary employees, as well as the social security, insurance and other types of fees schools/firms pay for them. The individual and family subsidies expenditures include government subsidies to individuals and families. 25 “Analysis of Preschool Education Development in Yunnan,” Shanghai Academy of Educational Sciences Intelligence Development Agency, 2013. 50 Figure 2.7 Per-Student Kindergarten Personnel Expenditures in Western Provinces, in RMB, 2010 7000 5,723 6000 5000 4,247 3,733 4000 2,796 3000 2,172 2,307 1,676 1,931 1,889 2000 1,415 1,343 1,286 987 831 1000 0 Source: Shanghai Academy of Educational Sciences Intelligence Development Agency, 2013 In 2010, per-student operational expenditures26 amounted to 734 RMB in Yunnan. Even though expenditures grew by 48.8 percent from 2007, they are still 576 RMB lower than the national average, and 388 RMB lower than the average level of the western region. Compared to other western provinces, Yunnan’s per-student operational expenditures are very low, about 1/3 of Inner Mongolia’s level. This reflects the fact that the daily operation costs of kindergartens are relatively low in Yunnan. Figure 2.8 Per-Student Kindergarten Operational Expenditures in Western Provinces, in RMB, 2010 3000 2,525 2500 2,207 2000 1,850 1,842 1,310 1,423 1500 1,123 981 891 883 1000 810 808 734 545 500 0 Source: Shanghai Academy of Educational Sciences Intelligence Development Agency, 2013 26 Operational expenditures include “goods and services expenditures” and “other capital expenditures.” Goods and services expenditures reflect the expenditure on goods and services that schools/firms purchase (excluding expenditures on fixed assets). Other capital expenditures reflect expenditure on fixed assets, land, and intangible assets, as well as expenditures on infrastructure building and large-scale renovation that schools/firms incur outside the central arrangement by all levels of development and reform departments. 51 The largest funding disparity in the financing of preschool services, however, is not between the east and the west. It lies within provinces between rural and non-rural areas. In Yunnan, rural preschool institutions (including both kindergartens and one-year pre-primary classes) account for 50 percent of both enrollment and number of facilities in the province, to which only 28 percent of all preschool funding and 13 percent of government budgetary funds on preschool were dedicated (Figure 2.9). The contrast is even more polarized when the number of students is taken into account. Figures 2.9 and 2.10 summarize the total and budgetary expenditures per student in rural areas. 27 Figure 2.9 Share of Rural Preschool Finance and Enrollment, by Region, 2011 34 Natioanl Average 35 41 28 Yunnan 13 50 34 Eastern Provinces 38 43 33 Western Provinces 32 36 0 10 20 30 40 50 60 Rural Total Funding share Rural Budget Share Rural Enrollment Share a c Source: MOE 2010, MOE 2011. Figure 2.10 Budgetary Spending per Student on Rural Preschool Education, by type, 2011 41% 30% Rural Yunnan 29% National Rural 56% 36% 8% Regions Rural Western 34% 24% 42% Rural Eastern 61% 38% 2% 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 RMB Personnel Expenditure Operational Expenditure Construction Expenditure c Source: MOE 2011. 27 Enrollments are as of 2010 data. 52 Figure 2.11 Total Spending per Student on Rural Preschool Education, by type, 2011 33% 50% Rural Yunnan 17% National Rural 48% 48% 5% Regions Rural Western 34% 32% 35% 1% Rural Eastern 53% 46% 0 1,000 2,000 3,000 4,000 5,000 RMB Personnel Expenditure Operational Expenditure Construction Expenditure c Source: MOE 2011. Private Spending on Preschool Education The relatively high proportion of preschool education tuition and fees poses a significant burden on families. In 2010, tuition and fees revenue reached 650 million RMB in Yunnan, accounting for 43.6 percent of total investment in preschool education in the province. In the western region, tuition and fees on average accounted for 40.8 percent of total investment in preschool education. Among western provinces, tuition and fees make up a relatively high percentage of total investment in preschool education in Yunnan, higher than that of Inner Mongolia, Qinghai, and other provinces. 53 Figure 2.12 Percentage of Tuition and Fees out of Total Investment in Preschool Education in the Western Provinces, 2010 450 80.0 400 384.2 68 70.0 350 60.0 59 58 300 53 54 50.0 48 250 43 44 43 41 40.0 200 34 28 30.0 150 100 20.0 61.7 16 50 7.6 9.1 13.7 2.8 8.9 10 10.0 4.7 6.5 0.2 3.0 0.8 1.7 2.8 0 0.0 Tuition (100 million RMB) Share of Total Preschool Funding (%) Source: Shanghai Academy of Educational Sciences Intelligence Development Agency, 2013 Table 2.3 Items and Standards for Kindergarten Fees in Kunming Items Comments Fees 20 RMB per student per 1.Boarding kindergartens may charge （1）Miscellaneous semester extra 50 RMB/month boarding fee. 寄宿制幼儿园加收 50 元/月寄宿费 （2）Childcare 2．Day-care classes during summer Grade One 150 RMB/month and winter holidays may charge Kindergartens extra 50 RMB/month fee.寒暑假期 Grade Two 110 RMB/month Kindergartens 间的留园班加收 50 元/月留园费。 Grade Three 90 RMB/month Kindergartens Ungraded 70 RMB/month Kindergartens Source: Kunming Information Center, 2011 Table 2.3 presents the 2011 public announcement of kindergarten fee standards in Kunming, 54 Yunnan, which remain the same as the 2010 rates. A further calculation is needed to understand the burden of kindergarten tuition and fees on families in Yunnan. Assuming a non-boarding child takes summer and winter breaks off and attends kindergarten for 9 months, the annual childcare fee differs by the rated grade of the kindergarten: 1350 RMB/year for first-grade kindergartens; 990 RMB/year for second-grade kindergartens; 810 RMB/year for third-grade kindergartens; and 630 RMB/year for ungraded kindergartens. Families also need to pay an additional 40 RMB for miscellaneous fees. According to the China Statistics Yearbook 2011, the average per-capita annual income in Yunnan in 2010 was 16,065 RMB for urban residents and 3,952 RMB for rural residents. Therefore, the annual fees for a first-grade kindergarten child would account for 8.7 percent of urban residents’ per-capita annual income, and 35.2 percent of rural residents’ per-capita annual income. Figure 2.13 illustrates that kindergarten tuition and fees account for a higher percentage of average per-capita household income in Yunnan, particularly for rural residents. Figure 2.13 Proportion of Kindergarten Tuition and Fees of Average Per-Capita Income by Kindergarten Grade, Urban versus Rural residents, in RMB, 2010 40 35.2 35 30 26.1 25 21.5 20 17.0 15 8.7 10 6.4 5.3 4.2 5 0 Grade 1 Kindergarten Grade 2 Kindergarten Grade 3 Kindergarten Ungraded Kindergarten Urban residents Rural residents Source: Shanghai Academy of Educational Sciences Intelligence Development Agency, 2013 Summary and Discussions Governmental appropriations are the biggest funding source for education in China. However, allocations to preschool education remain small. Government appropriations comprise approximately 80 percent of all available educational funds. In 2012, total government allocations in education increased to 4 percent of national GDP, which resulted in an increase in the educational budget at both the aggregate and per-student levels. However, distribution of budgetary educational funds is still heavily skewed against preschool education: although preschool enrollment accounts for 13 percent of total enrollment, it only receives 2 percent of national educational budget allocations. Preschool education relies more heavily on private investment, as 59 percent of total educational funds 55 come from non-governmental sources. Yunnan has made a significant investment in preschool education, and yet low per-student expenditures and funding disparities still exist. Total educational funding for the preschool sector rose at an annual rate of 32 percent from 2005 to 2011 in the province. However, both total spending per-student and budgetary spending per-student in Yunnan lag behind the national average and other western provinces. Personnel expenditures accounted for nearly three-quarters of per-student budgetary spending. Both per-student personnel expenditures and per-student operational expenditures remain low as compared to the national average and other western provinces, reflecting lower teacher salaries and daily operational costs in the province. Considerable funding disparities persist between rural and non-rural areas. Even though rural preschool institutions accounted for 50 percent of both enrollment and number of facilities in the province, they only received 28 percent of all preschool funding, and 13 percent of government budgetary funds. Families, especially those in rural areas, bear a significant burden in paying for preschool education. In Yunnan, tuition and fee revenue comprise 43.6 percent of total investment in preschool education, a higher proportion than the western region average. Potentially, tuition and fees for a first-grade kindergartener could consume 8.7 percent of urban residents’ annual per-capita income and 35.2 percent of rural residents’ annual per-capita income. Given the significant social benefits of early childhood education, such as lower crime rate and inequality of opportunity, there is economic rationale for greater public investment to subsidize and lower the burden on individual households. 56 Part III. Status of Children in Rural Yunnan: Results from a Rural Household Survey The national and provincial data reveal major rural-urban disparities in Yunnan with regard to access to and quality of preschool education, as well as financial constraints in providing these services. In this section, we detail the status of 3-6 year-old children living in rural Yunnan based on an extensive household survey. About 1000 rural households with 3-6-year-old child (children) were interviewed for major physical, social and cognitive development indicators, home care conditions and childrearing practices, as well as enrollment in various early childhood education programs. The sampling of households followed a stratified random design. County units were selected and categorized into four subgroups: (i) counties on the border; (ii) non-border, ethnic minority counties; (iii) non-border, non-minority, poor counties; and (iv) non-border, non-minority, non-poverty counties. Within each group, all counties were sorted by a compound index comprised of GDP, annual budget expenses and income per capita. Ten county-level units were then randomly selected using the probability proportionate to group size. The total sample includes 10 counties, 30 villages, and 1,031 households.28 The survey instrument was initially designed by the World Bank in collaboration with the National Health and Family Planning Commission. It was first fielded in Hunan Province and most recently adopted by the CDRF (Child Development Research Foundation), and then applied in the Xundian County of Yunnan Province. The version used for the study has been slightly modified to suit the particular circumstances of Yunnan. It contains 461 multiple choices regarding demographic information on children and parents (or primary guardian), household characteristics, guardians’ opinion on gender, childrearing practices , attendance at parent-child classes and kindergartens, family behavioral conditions, nutrition and diet, details on educational status, and living conditions of households. In addition, a multiple-item test for direct observation of 3-6-year-olds was also used by survey workers to assess childhood social and cognitive development status. Two indexes were respectively calculated based on these test questions. The survey sought to provide empirical evidence to the following questions:  What is the development status of 3-6-year-olds living in rural Yunnan?  Who is taking care of rural 3-6-year-olds, and what are the prevalent home care conditions and childrearing practices?  How large is the disparity in social and cognitive development indicators between boys and girls, between Han and ethnical minorities, between national poverty and 28 During the implementation of the survey, 3 villages were not reached due to landfills or other road conditions; nearby villages were sampled instead. Three other villages were not sampled due to incorrect information about village size or villagers. 57 non-poverty counties, and between left-behind children and children cared for by their own parents?  Is there an association between parent-child classes and kindergartens, access to health checkups, and good parenting rearing practices and children’s development indicators, after controlling for gender, left-behind status, educational attainment of parents/guardians and family income?  What social, geographical or household factors correlate with kindergarten enrollment? Rural Home Care Conditions, Childrearing Practices, and Childhood Developmental Status Who is taking care of the children? Figure 3.1 Children's Primary Caregiver in Sample Counties in Yunnan, 2013 80 70 60 Father= 28% % of respondents 50 40 30 Mother= 48% 20 Grandparents: 10 23% 0 Parents Other relatives Source: Yunnan Household Survey, 2013. In total, 76 percent of the children in the survey were primarily taken care of by their own parents (N=778), 23 percent were cared for by their grandparents (N=253), and 2 percent by other relatives. Among those who were not cared for by their parents, 71 percent or 181 of them are considered “left-behind” children.29 Infant birth weight and prevalence of breastfeeding As Figure 3.2 shows, children’s birth weight is comparatively lower in minority and poverty counties versus non-minority and non-poverty ones. The difference in border/non-border counties is not significant. Breastfeeding appears to be the predominant form of infant feeding with 92 percent of all respondents reporting that their children were breastfed by their mothers (Figure 3.3). Different counties have similar results. 29 According to the survey, this was defined as dependent(s) whose parents work far away from home and are cared for by other relatives. 58 Figure 3.2 Birth Weight of Children in Rural Yunnan, 2013 3.25 3.23 3.22 Birth Weight in Kiilogram (kg) 3.2 3.17 3.16 3.15 3.14 3.10 3.1 3.05 3 Poverty County Minority County Border County Poverty, Minority or Border County Non-Poverty, non-Minority, non Border County Source: Yunnan Household Survey, 2013. Figure 3.3 Infant Breastfeeding in Yunnan, by County Type, 2013 100 93.5 95.0 93.3 90.1 89.6 91.0 Percentage of Breastfeeding 90 80 70 60 50 Poverty County Minority County Border County Non-Poverty, non-minority, non-border county Poverty, minority, border county Source: Yunnan Household Survey, 2013. Educational level of parents and caregivers Compared with fathers, the mothers’ educational level was more skewed toward the lower end of the educational spectrum (Figure 3.4). The average mother or father tends to have a junior secondary degree. The illiteracy rate among mothers was comparatively higher than that of fathers, while other shares of education attainments were lower. Caregivers had very similar education distribution to mothers, reflecting the fact that a majority of them were also mothers themselves. 59 Annual Household Income Overall, annual household income was unevenly distributed across minority autonomous counties and Han counties, as well as border and non-border counties (Figure 3.5). On average, households in minority and border counties had higher incomes respectively than households in non-minority and non-border counties. Figure 3.4 Educational Levels of Mothers, Fathers, and Caregivers in Yunnan, 2013 40 37 35 33 33 30 31 30 29 30 % of respondents 26 25 20 17 15 9 10 9 10 5 1 1 2 1 1 0 0 0 0 Mother Father Caregiver Source: Yunnan Household Survey, 2013. Figure 3.5 Annual Household Income in Yunnan, 2013 40 36 36 36 36 37 35 32 33 30 28 30 % of respondents 24 23 25 21 20 15 10 5 6 6 4 5 1 1 2 0 0 Between minority/non-minority counties Between border/non-border counties Non-minority/non-border counties Minority/border counties Source: Yunnan Household Survey, 2013. 60 Child rearing practices The time spent on reading and playing with children as well as spent on watching TV across minority and non-minority counties was quite similar, but not evenly distributed across poor counties and non-poverty counties (Figure 3.6 and 3.7). Disparities emerged as households in poverty counties spent considerably less time reading and playing with children, while the time spent on watching TV was similar. Almost half of the households living in poor counties never read to their children during the past year. An astonishingly high proportion of households in poverty counties never played with their children (72 percent) or read to them (47 percent) in the past year. On the other hand, TV watching is very prevalent among 3-6-year-olds. About 25 percent of children watch TV for more than 2 hours a day, and another 30 percent or more children watch TV for 1 to 2 hours per day. Figure 3.6 Reported Time Spent Reading and Playing with Children and Watching TV in Yunnan, 2013 70 60 59 58 % of respondents 50 40 40 36 37 32 36 31 35 35 38 34 31 30 29 26 23 20 10 4 5 6 3 2 2 0 Non-minority counties Minority counties Source: Yunnan Household Survey, 2013. 61 Figure 3.7 Reported Time Spent Reading and Playing with Children and Watching TV in Yunnan, 2013 80 72 70 60 % of respondents 50 47 49 44 40 40 36 39 38 33 31 30 25 28 25 22 25 24 20 10 5 5 4 4 2 2 0 Non-poverty counties Poor counties Source: Yunnan Household Survey, 2013. Dietary habits Dietary differences between poverty counties and non-poverty counties were striking. 26 percent of respondents in poverty counties reported that their children almost never drank milk, and 11 percent never ate meat/eggs, as compared with 7 percent in non-poverty counties. However, over 50 percent of households in non-poverty counties said that their children drank milk more than 2-3 times a week, as compared to 38 percent in poverty counties. Similar results were also observed for fruit intake (Figure 3.8). 56 percent of households in non-poor counties reported that their children ate fruits everyday as compared to only 37 percent in poor counties. 62 Figure 3.8 Frequency of Various Types of Food Intake in Yunnan, 2013 90 8386 80 70 60 56 51 50 46 37 40 34 32 33 25 26 28 30 22 25 2222 18 20 18 17 20 13 14 11 10 4 7 7 55 65 64 69 0 Milk Intake Meat/egg int Vegetable intake Fruit Intake Non-poverty County Poverty County Source: Yunnan Household Survey, 2013. Number of books at home Overall, rural Yunnan does not have a book/reading culture. Most households sampled do not own any books at home. There is a persistent difference in the number of books available at home between different county types. Minority, poor and border counties in general had fewer books than non-minority, non-poverty and non-border counties (Figure 3.9). Language spoken at home A majority of households used a Mandarin dialect (a local variant of the standard Mandarin) as the primary language of communication at home (Figure 3.10). However, the share of households using local minority dialects is considerably higher in minority counties than in non-minority counties. Since most border counties were administered as ethnic autonomous regions, this significance carried through border/non-border counties. Further, standard Mandarin, often the official language at schools, was used very little at home. Eighty percent of households in poor counties do not own any books at all. Overall, only 7 percent of households have more than 30 books at home. 63 Figure 3.9 Number of Books at Home across County Types in Yunnan, 2013 90 80 78 80 72 70 66 62 % of respondents 60 56 50 40 26 28 30 22 18 17 20 14 5 7 5 4 7 8 6 7 10 4 3 3 1 0 Non-minority Minority Non-poverty Poor counties Non-border Border counties counties counties counties counties None Less than 10 11-30 More than 30 Source: Yunnan Household Survey, 2013. Household smoking habits Tobacco is one of the main crops in rural Yunnan. Unfortunately the overall prevalence of smoking is high in rural Yunnan, as is the lack of awareness of avoiding the impact of second-hand smoking in front of children (Table 3.1). Households in poor/non-poverty households presented similar patterns of domestic smoking. Minority counties had on average a lower prevalence as compared to non-minority counties. Households in both minority and non-poverty counties were less likely to have members smoke in front of children. 93 percent of households in poor counties and 94 percent of all non-minority households have members smoke in front of their 3-6-year olds. Table 3.1 Smoking Habits in Yunnan, by Region, 2013 (%) Non-poverty Poor Non-minority Minority counties counties counties counties Someone in the household smokes 80 83 88 77 1 person smoke 55 57 63 51 2 person smoke 23 21 22 22 3 person smoke 1 4 2 3 More than 3 people smoke 1 1 0 1 Parent smoke 75 73 82 68 Smoke in front of child 87 93 94 88 Source: Yunnan Household Survey, 2013. 64 Language(s) spoken at home Only 2 percent of households surveyed speak standard Mandarin at home. Instead, the majority of households speak a Mandarin dialect (Yunnan style). A surprisingly high proportion (63 percent) of households in minority counties speak Mandarin dialect, and only 35 percent of the households in minority counties speak minority dialects. This is most likely due to the fact that there are also Han households (who primarily speak mandarin) located in minority counties. Figure 3.10 Primary Communication Language(s) at Home in Yunnan, 2013 Primary Communicatioan Language at home in Sample counties in Yunnan 100 89 80 % of respondents 63 60 40 35 20 9 2 2 0 Mandarin Madarin dialects Minority dialects Non-minority counties Minority counties Source: Yunnan Household Survey, 2013. Characteristics of Kindergarten Enrollment The average sampling enrollment rate is 59 percent (N=603). Twenty-five percent of children (N=151) were enrolled in one-year pre-primary classes. Across different counties, the enrollment differences were largest in minority/non-minority counties. In general, households in poor, border and minority counties have lower enrollment rates than in non-poverty, non-border, and non-minority counties, respectively. Figure 3.11 displays the regional differences. 65 Figure 3.11 Enrollment of Kindergarten/One-year Pre-primary Classes, by County Type in Yunnan, 2013 80 74 62 63 60 56 % of enrollment 49 50 48 48 41 41 41 40 30 25 20 14 15 16 8 11 0 Minority/non-minority Poverty/non-poverty Border/non-border counties counties counties Enrolled in preschools or kindergartens (non-minority, non-poverty and non-border counties) Enrolled in preschools or kindergartens (minority, poverty and border counties) Enrolled in Pre-primary Classes (non-minority, non-poverty and non-border counties) Enrolled in Pre-primary Classes (minority, poverty and border counties) Enrolled in kindergartens (non-minority, non-poverty and non-border counties) Enrolled in kindergartens (minority, poverty and border counties) Source: Yunnan Household Survey, 2013. Figure 3.12 Enrollment Rate in Kindergartens or One-year Pre-primary Classes, by County Types, 2013 From Minority County? 0.8 No Yes 74% 0.4 49% 0 62% 41% 63% From Poverty 56% From Border County? County? Source: Yunnan Household Survey, 2013. 66 Safety and distance were the first two considerations if choices of enrolling in preschools/kindergartens were available (Figure 3.13). For those children not enrolled in preschools/ kindergartens, distance was also a significant factor (Figure 3.14). It is not surprising to see that the cost was not high on the list — as in many cases, availability was the primary issue.30 Figure 3.13 The Most Important Reason(s) for Selecting Kindergartens/Preschools in Yunnan, 2013 35 33 30 % of respondents 25 22 20 20 14 15 10 3 3 4 5 1 0 All Households Enrolled in Kindergarten/Preschool Classes (N=602) Source: Yunnan Household Survey, 2013. 30 Other reasons for non-attendance include primarily: (i) children are considered too young to attend kindergartens (20 percent，家长或幼儿园认为孩子还太小); (ii) children are under-aged for pre-primary classes (14 percent，没有幼儿园, 学前班还太小); and (iii) parents believe that their children are unfit because of poor health or underdeveloped social or communication skills (7 percent 身体不好、不合群或沟通交流能力不佳). 67 Figure 3.14 The Most Important Reason(s) for not Attending Kindergartens/Preschools in Sample Counties in Yunnan 60 55 50 % of respondents 40 37 30 20 10 6 1 0 No Kindergartens Too costly Domestic Education Other Reasons Nearby is Better All Households not Enrolled in Kindergarten/Preschool Classes (N=402) Source: Yunnan Household Survey, 2013 Distance to the nearest kindergarten and monthly fees for kindergartens Respondents’ perceptions about the distance to the nearest kindergarten and the associated monthly fees varied significantly (Figure 3.15 and Tables 3.2, 3.3). As reported, average distance to the nearest kindergarten is rather long at 6 km, and more than twice the reported distance to the nearest primary school (2.6 km). This could reflect the relative lack of access to kindergartens as compared to primary schools in rural Yunnan. In fact, the standard deviation for the estimated distance to kindergartens is very large at 12.9 km, which could indicate that respondents may not be as accurate in estimating distance to kindergartens. In order to measure the relative distance of kindergartens, researchers created a new variable of whether there is a kindergarten within the distance of a nearest primary school. Indeed, the estimation of distance to primary schools was much more reliable (much smaller standard deviation). Figure 3.16 shows the statistics and the distribution of the new variable (named “closer”). The gap is widest among border/non-border counties followed by poor/non-poverty counties, which may imply that kindergartens were distributed even more thinly in those areas. 68 Figure 3.15 Respondents' Estimates of Nearest Kindergartens and Associated Costs 1,400 1,200 Monthly Fee (RMB) 1,000 800 600 400 200 0 0 10 20 30 40 50 60 70 80 Estimated Distance to Nearest Kindergarten (km) Source: Yunnan Household Survey, 2013 Table 3.2 Descriptive Statistics of Estimated Distances to Nearest Kindergartens, Estimated Monthly Fees, and Estimated Distances to Nearest Primary Schools Variable N Mean Std. Dev. Min Max Estimated distance to 990 6.1 km 12.9 km 0 km 74 km nearest kindergarten Estimated Monthly 794 229.4RMB 158.4RMB 0RMB 1225RMB fee Estimated distance to nearest primary 1013 2.6 km 5.7 km 0 km 74 km school Source: Yunnan Household Survey, 2013. Table 3.3 Average Monthly Fees as Perceived by Respondents Enrollment type Average Fees for Nearest Number of kindergarten (RMB) Observations Not enrolled 212 216 One-year, pre-primary class 164 138 Kindergarten 257 434 Total 229 788 Source: Yunnan Household Survey, 2013 69 Figure 3.16 Percent as a Share of Households that Recognize a Closer Kindergarten than a Nearest Primary School in Yunnan, by County Type, 2013 30 25 24 25 22 20 % of respondents 20 17 15 10 7 5 0 Minority/non-minority Poor/non-poverty counties Border/non-border counties counties Non-minority, non-poverty and non-border Counties Minority, Poor and Border Counties Source: Yunnan Household Survey, 2013. Correlates of Development Outcomes Based on the individual responses on the children’s social and cognitive questionnaire, we calculated the social development and cognitive development indexes for all 3-6-year-old children in the sample.31 The score is continuous, between 0 and 2 points, reflecting the average of all questions answered. Table 3.4 and Figure 3.17 depict the key findings when scores were tabulated with main explanatory variables. In this context, it is important to note the following:  Development scores do not differentiate between genders;  Left-behind children and children living in minority counties have significantly lower social development scores, but do not differentiate between cognitive development scores;  Minority children have significantly lower social and cognitive development scores than Han children;  Children living in poor or border counties have significantly lower social and cognitive 31 The indexes were calculated based on a separated children’s survey conducted with the household research. The survey was designed by the World Bank and China’s National Health and Family Planning Commission. Two interviewers measured the children’s social and cognitive development with a scaled instrument containing 25 – 37 items for different age levels. These covered basic social awareness (that is, names, body parts, emotions and social norms) as well as memorization and imitation. Six behavioral questions in the caregiver’s survey were also included in the social development index (question numbers 322, 323, 324, 327, 330, 331). 70 development scores than children who live elsewhere;  Children enrolled in kindergarten have significantly higher social and cognitive development scores than children who do not attend kindergarten. Table 3.4 Selected Index Scores across Main Explanatory Variables Social development Cognitive index development index Score N Score N Average 1.4 1019 0.84 1036 Boy 1.40 547 0.85 549 Girl 1.39 472 0.84 477 Han 1.42* 539 0.92*** 483 Minority 1.38 480 0.78 543 Enrolled in kindergarten 1.46*** 597 1.02*** 600 Not enrolled 1.31 414 0.60 418 Left-behind 1.35** 178 0.83 180 Live with parents 1.41 841 0.85 846 Live in poor county 1.34*** 517 0.80** 520 Do not live in poor county 1.46 502 0.89 506 Live in minority county 1.38* 609 0.83 612 Do not live in minority county 1.43 410 0.87 414 Live in border county 1.31*** 200 0.74*** 202 Do not live in border county 1.42 819 0.87 824 Is the single child 1.48 260 0.89* 258 Is not the single child 1.37*** 759 0.83 768 *indicates <0.05 significance **indicates <0.01 significance ***indicates <0.001 significance Source: Yunnan Household Survey, 2013. 71 Figure 3.17 Social and Cognitive Development Scores across Regions in Yunnan, 2013 1.6 1.46 1.34 1.381.43 1.31 1.42 1.4 Development Score 1.2 1 0.89 0.830.87 0.87 0.8 0.74 0.8 0.6 0.4 0.2 0 Poor county Minority Border Poor county Minority Border county county county county Social Development Score Cognitive Development Score Poor, Minority and Border Counties Non-poor, non-minority, and non-border Counties Source: Yunnan Household Survey, 2013. Multivariate analysis of the data was conducted to determine the effects of different variables on childhood development outcomes. Weight, height,32 social development and cognitive development were used as outcome measures. The mean and standard deviation of independent variables used can be found in the Appendix. Demographic Characteristics:  The weight of children is positively correlated with weight at birth, and the age at which the latest weight was measured. Being a girl and coming from a minority or poverty county was negatively correlated with weight. Compared with no educational attainment, mothers with a tertiary education were also positively correlated with the weight of the child. Breastfeeding, though not a significant predictor, has a positive association with the weight of the child.  The height of children is positively correlated with the age at which the latest measurement was taken. When controlling for other factors, a child from a high-income household (more than 100,000 RMB in annual income) was significantly taller than one from a low-income household (less than 10,000 RMB in annual income). Coming from a poverty county or living in a big family were also negatively correlated with height.  The social development index was positively correlated with children’s age and the wealth of the family. Compared with households with less than 10,000 RMB in annual income, households with 10,000-20,000, 20,000-50,000 and 50,000-100,000 RMB in 32 During the survey, no standard weight or height was measured. Instead, survey workers asked the parents/caregivers to give the latest weight/height measured. The regression models took this estimate as outcome variables while controlling for the age at which the measurement was taken. This result was still rough due to the quality of estimates as well as the limited number of respondents who provided complete information. 72 annual income had a positive and incremental correlation coefficient with children’s social development score. With all else being controlled, coming from a poverty or border county was negatively correlated with the social development score. Speaking Mandarin rather than minority dialects as the primary language was positively correlated with children’s social development.  Likewise, there is a strong association between family income levels and children’s cognitive skills. Higher annual incomes in general correlate with higher cognitive scores, except for the highest stratum (with over 100,000 RMB), which had a lower effect on children’s cognitive scores as compared to households with 50,000-100,000 RMB. Mothers’ education levels also highly correlate with children’s cognitive development outcomes. The regression coefficients for degrees higher than junior middle school showed an incremental positive effect on the outcome variable. Also, single child status and being a Han (other than ethnic minorities) also had positive correlations with cognitive scores. Other factors amenable to policy intervention New groups of variables were added by the researchers as predictors of child development outcomes because they are more likely to be influenced by policy interventions than family characteristics. They include the following:  Whether or not the child attends kindergarten (including one-year pre-primary classes) and parent-child classes;  Frequency of health check-ups and immunizations;  Parent or child-rearing practices (number of books at home, frequency of playing with child, reading to the child, watching TV, and means of punishment); and  Dietary habits and nutrition (calcium intake, frequency of eating meat, eggs and drinking milk). The main findings include the following:  Enrollment is both statistically correlated with children’s social and cognitive skills, and positively associated with children’s recent weight and height.  Parenting practices are highly correlated with social and cognitive development. Reading to children on a weekly basis, as compared to never reading to children, had a significantly positive correlation with both social and cognitive development. Watching TV for 1-2 hours a day and more than 2 hours a day also had a significantly positive effect on improving cognitive skills.  Age and residence in a poverty county remained significant factors of children’s weight and height. Health check-ups at the frequency of “Sometimes” were positively correlated with children’s recent weight as compared to “Never” responses. In addition, 73 taking milk once a week was significantly correlated with children’s height as compared to never taking any.  Compared with “no punishment” toward wrong-doings, children receiving “occasional” and “constant “punishments for being naughty or making mistakes have positive correlations with weight and height.  Compared with children who take free immunizations only, children who had taken both free and paid immunization have better cognitive development.  It is also noted when policy-friendly variables were added into the equation – such as whether to attend kindergartens, frequency of medical check-ups or immunizations, as well as child-rearing practices, the original correlation between geological location, household income and parents’ education and children’s physical, social and cognitive development become weaker. This suggests that some of the impact of location, income, and parents’ education on child development might be through the policy-friendly variables. The adjusted R-squared of these models were higher than the original four without variables amenable to policy interventions. The revised model 8 explained 51 percent of variance in cognitive scores, up from 43 percent in model 4. Correlates of Kindergarten Enrollment These survey analyses reveal that enrolling in kindergarten tends to correlate positively with children’s social and cognitive development. Logistic regressions also help us to understand the correlates of preschool enrollment in rural Yunnan. Based on the same survey, demographic, social and geographic variables were chosen. The full model results are contained in the Appendix. Key findings include:33  Age and household income were strongly correlated with the odds ratio of enrollment. Compared with households of less than 10,000 RMB of income, the rough likelihood of children attending kindergartens in households with 10,000-20,000 RMB of income almost doubled.  Left-behind children were almost twice as likely to attend kindergartens as compared with non-left-behind children, when other constraints were held constant.  Ethnicity plays an important role. Compared with ethnic minorities, Han children were more likely to attend kindergartens. This tendency was evident in minority and non-minority counties as well. Households in minority counties are 60 percent less likely to have enrolled their children in kindergartens. 33 All odds ratios are estimated as probabilities in the explanations. 74  The mothers’ educational background also played an important role in preschool enrollment. Those mothers with a high school or university education were more likely to have their children attend kindergartens as compared to mothers with no educational attainment.  There is also a strong association between distance to the nearest kindergartens and enrollments. In the survey, due to the roughness of personal estimates, a relative distance compared to the nearest township/village primary school was used. The result shows a strong likelihood of attending preschool when it was located within the distance of a nearest primary school.  Interestingly, language spoken in the family also had a strong correlation with kindergarten enrollment. Compared to Mandarin-speaking families, households using either a Mandarin dialect or minority dialects were much less likely to have children attending kindergartens. However, this result could be biased due to the very limited number of cases of Mandarin-speaking families. Summary and Discussion The analyses of rural household surveys provide a snapshot of the status of 3-to-6-year-olds living in rural Yunnan. Based on descriptive analysis, we were able to reveal rural children’s home care conditions, caregiver’s childrearing practices, and their developmental status for both cognitive and social dimensions, and status of enrollment in any form of preschool. Further, using multivariate regression analysis, we were also able to estimate what factors are strongly associated with childhood development outcomes, as well as their enrollment status at preschools. Both background and policy variables were used in the regression analysis. Contrary to expectations, only about 25 percent of children in rural Yunnan are not cared for by their own parents, but by grandparents and other relatives. Of those, most are indeed left-behind children. The proportion of left-behind children is smaller than we expected though it may differ from county to county, as we know some counties such as Zhaotong have a much higher proportion of left behind children, whereas other rural counties such as Xishuanbanna are known to have very few migrants out of the county. Instead, they tend to attract workers from outside the county or even province during busy rubber tree harvest seasons. The average education for the parents’ generation is junior middle school, which is higher than the average primary education for the grandparents. With regard to childrearing practices, although more than 90 percent of the households breastfed their children, there is much room for improvement in other important childrearing practices in rural Yunnan. In particular, households in poor counties are not providing sufficient early stimulation for their young children. Unfortunately, for households in poor counties, 72 percent reported that they never played with their children; 47 percent reported that they never read to their children; and 80 percent reported that they do not have any books in the home. Playing with and reading books to young children 75 are the most basic early stimulation measures that have been proven to be associated with positive child outcomes. Box 3.1: The Effect of Early Reading on Children’s Cognitive Development A body of literature attests to the positive effects of early reading on children’s cognitive development. Based on a meta-analytic review of 14 intervention studies (Sénéchal, 2006) of the relationship between early reading and the childhood development outcomes, parental involvement has a positive effect on children’s literacy skills from kindergarten to grade 3. The average effect size for parental involvement in children’s reading activities is a 10-point increase on a standard literacy test. With respect to the specific intervention types, the approach of parents teaching their children literacy skills is the most effective intervention. Parents in seven studies received training to teach their child specific reading skills, such as teaching them the alphabet or how to read words. These reading activities produce an average of a 17-point increase on the literacy test. Kalb and van Ours (2013) find evidence for a causal relationship between parental reading and children’s reading skills and cognitive development. Using longitudinal data of Australian children, the authors find that frequent parental reading to children at ages 4-to-5 has a significant and positive effect on children’s reading and cognitive skills, at least up to an age of 10-to-11. In terms of effect size, reading 6-7 days a week to boys has a similar effect on cognitive skills development as being almost one year older, and the effect is slightly larger for girls. Source: Sénéchal, M. (2006). The Effect of Family Literacy Interventions on Children’s Acquisition of Reading From Kindergarten to Grade 3. National Institute for Literacy, RMC Research Corporation. Kalb, G. & van Ours, J. (2013). Reading to Young Children: A Head-Start in Life? IZA Discussion Paper No. 7416. Some practices that are highly detrimental to children’s health and development are prevalent. For example, approximately 90 percent of households have members smoke in front of their young children, with absolutely no awareness that secondhand smoking affects their children’s health and development. A large number of research studies have confirmed that secondhand smoke has many harmful effects on infant and child health. Secondhand smoke during pregnancy is a direct cause of low infant birth weight. Infants who are exposed to secondhand smoke after birth face a bigger risk of SIDS. Parental smoking leads to higher rates of both upper and lower respiratory tract illnesses such as asthma, bronchitis, middle ear infection (otitis media), and pneumonia among infants. Furthermore, secondhand smoke within 3 meters of the infant is associated with higher hospital admission rates for all illnesses of infants in the first 18 months of life (Leung et al., 2004). For older children, those who breathe secondhand smoke experience less lung growth than those who do not. They are also more vulnerable to asthma attacks and ear infections. 76 Further, it appears that as living conditions improve, the majority of rural households own TV and their children spend substantial amounts of time watching TV every week. Multivariate analysis revealed a positive relationship between watching TV more than 2 hours a day and cognitive development. We speculate that this can be attributed to the fact that TV promotes Mandarin speaking and helps children master the Mandarin language better. Nevertheless, excessive TV viewing, tends to be associated with a lack of other positive interactions with parents and other caregivers, and can potentially affect the children in negative ways. Box 3.2 The Effect of TV Watching on Childhood Development Outcomes Research on the effect of TV watching on childhood development outcomes has produced mixed results. Focusing just on the number of hours of TV watching, Huang and Lee (2010) find that watching less than 2 hours of TV per day during ages 6-7 has a small but positive effect on reading scores at ages 8-9. However, watching TV during ages 6-7 and 8-9 has an overall negative effect on math scores at ages 8-9. A body of literature suggests that the effect of TV viewing is contingent upon the content and type of the TV programs. There is well-documented evidence that age-appropriate, curriculum-based educational television programs can have a positive effect on children’s cognitive development and school readiness. For example, according to Linebarger and Walker (2004), watching interactive TV programs such as “Dora the Explorer” and “Blue’s Clues” was associated with higher vocabularies and expressive language scores for 30-month-old American preschool children. Anderson et al. (2001) conducted a longitudinal study and found that watching educational TV programs was associated with higher grades and more books read in high school. A number of earlier studies have also suggested that exposure to violence on TV during childhood is closely linked to more aggressive behavior. Huesmann et al. (2003) investigates the long-term effect of viewing violent TV in a longitudinal study that interviewed 329 children in elementary school, and then again when they were in their early 20s. Results show that viewing more TV violence and identification with aggressive same-sex TV characters is associated with more adult aggression for both male and female participants, regardless of their initial aggression level or social status. Source: Anderson, D. R., Huston, A. C., Schmitt, K. L., Linebarger, D. L., & Wright, J. C. (2001). Early childhood television viewing and adolescent behavior: The recontact study. Monographs of the Society for Research in Child Development, 66(1), 1-147. Center on Media and Child Health, Children’s Hospital Boston. (2005). The Effects of Electronic Media on Children Ages Zero to Six: A History of Research. Prepared for the Kaiser Family Foundation Fali, H., and Myoung-Jae, L. (2010). Dynamic treatment effect analysis of TV effects on child cognitive development. Journal Of Applied Econometrics, 25(3), 392-419. doi:10.1002/jae.1165 Linebarger, D. L., and Walker, D. (2005). Infants' and Toddlers' Television Viewing and Language Outcomes. 77 American Behavioral Scientist, 48(5), 624-645. doi:10.1177/0002764204271505 Huesmann, L.R., Moise-Titus, J., Podolski, C.L., and Eron, L.D. (1993) "Longitudinal Relations Between Children's Exposure to TV Violence and Their Aggressive and Violent Behavior in Young Adulthood: 1977 - 1992," Developmental Psychology, Vol. 39, No. 2. Interesting results were also emerged from the multivariate analysis. Some patterns of childhood development are evident. Children’s physical development as measured by weight is negatively associated with being a female child either in poor or minority counties, but positively associated with maternal education (tertiary education), family income, and regular health checkups. Similarly, a child’s height also seems to be correlated with family income, as well as living in a poor county or coming from big family. Drinking milk frequently has a positive impact on child’s height as well. Family income and maternal (but not paternal) education are highly associated with child’s cognitive development. Being a single child and being a Han are also correlated with higher cognitive skills. Family income and speaking Mandarin are positively correlated with social development. Coming from a poor or border county tends to be negatively correlated with social development. Interestingly, enrollment in preschools, frequency of health checkups, and positive childrearing practices are clearly correlated with positive childhood development outcomes. In particular, reading frequently to a child is positively correlated with social and cognitive development. Watching TV 1-2 hours or more than 2 hours a day tends to be associated with higher cognitive skills, but lower social skills (not statistically significant). With regard to preschool enrollments, we find that family income, “left behind” children, maternal education, distance to the nearest kindergarten, and language spoken at home all have a high correlation with enrollment status. In particular, for households with income between 10,000 RMB to 20,000 RMB the odds of their children attending preschool are two times that of children from families with income below 10,000 RMB. Further, mothers with a high school education are 1.5 times more likely to send their children to preschool as compared to mothers who have no educational attainment. The study demonstrates that there is great merit in promoting more preschool education for rural children, in particular those who have not yet had access to such education. For children 3 years and older, global evidence suggests that center-based programs (preschools) are appropriate and effective in improving children's cognitive and social-emotional development and school readiness. The impact is greater with higher quality programs, although informal and community-based programs also have proven benefits. Although most of the factors impacting enrollment such as gender, location, and family income may not be immediately amenable to policy interventions, some factors can be more easily influenced. These include maternal education, distance to nearest kindergartens, and affordability of preschool education. 78 At the same time, it is important to remember that families play at least an equally important role, if not more important role, in their children’s development. Vast research evidence confirms that a lack of learning opportunities and poor quality caregiver-child interaction pose major risk for childhood development. In addition to basic health and nutrition needs, substantial gains in childhood development require improvements in parenting, stimulation and early education. Unfortunately our survey results indicate a general lack of awareness of the critical importance of early stimulation, playing, and learning play in overall childhood development. Such a trend needs to be reversed as a matter of urgency, and can be done via various interventions that promote a more positive home environment for young children. Parents can provide more quality stimulation and interactions for their children by talking to them, and reading and playing with them. Parental caregiving can be enhanced through home visits, guidance and support from health providers, and group parent training. 79 Part IV. Preschool Quality in Kunming In this section, we provide a measure of the overall quality of preschools in Yunnan Province. This is indeed a very challenging task with multiple layers of complication. One ambitious measure of outcome of course would be childhood level developmental outcomes. For the purpose of this report, however, we adopt more realistic and measurable indicators of preschools through the use of a quality proxy. At the policy level, there are official quality standards and rankings of various preschools. In Yunnan, new public and private preschools have to subscribe to a series of standards. New public preschools need to meet the requirements under the Yunnan Province Preschool Education Guidelines ( 云 南 省 学 前 教 育 条 例 ) on facilities, location, qualifications of administrators, quality of teachers and healthcare professionals, size of classroom and class, and sanitation. Specifically, the Guidelines state that each kindergarten class needs to have at least two teachers and one caregiver (保育员). New private schools have to fulfill standards stipulated by the Yunnan Province Private Preschool Education Institutions Establishment Guidelines and Standards (云南省民办学前教育机构设置指导标准) on the aforementioned areas, plus provide sustainable funding sources. The local department of education is responsible for approving the creation of both public and private preschools. Currently, kindergartens in Yunnan are evaluated based on the Yunnan Exemplary Kindergarten Performance Comprehensive Evaluation Plan (云南省示范幼儿园办园水平综 合评价法案). Of the standard of three classes, kindergartens may receive a rating title of Class 1, 2, or 3, with Class 1 being the best rating. Within each class, there is a further distinction of Grade 1, 2, and 3. Class 1, Grade 1 is the highest rating title under the evaluation plan. In the evaluation process, kindergartens first conduct a self-evaluation on a set of 36 criteria, and then receive an external evaluation by an expert team commissioned by the relevant department of education. The current rating system in Yunnan covers four major areas of kindergarten performance: kindergarten conditions, kindergarten administration, quality of the system, and school impact. It is noteworthy that the kindergartens have to self-evaluate on a wide range of aspects in kindergarten performance, including the following: planning and development, school facilities, human resources, management and incentive mechanism, meals, sanitation, pedagogy and research, engagement with parents and the community, evaluation and feedback, the child’s all-around development, teacher quality and improvement, and the development of special kindergarten features. The external evaluation team collects information that ranges from parental satisfaction, teacher quality, physical resources, to administrative documents. However, the current evaluation system in Yunnan, particularly the external evaluation, still lacks appropriate emphasis on student learning and development experience within the institution. Child-centered aspects such as curriculum, student-teacher interactions within the classrooms, and personal care received by the children are areas that still need greater 80 attention and focus in evaluations. The research team applied the ECERS-R scale to assess the quality and atmosphere of Yunnan preschools. Since the ECERS-R scale is internationally benchmarked, the assessment results can also be used for comparative analysis. Assessment of Preschool Quality with ECERS-R A study on the quality of 24 kindergartens in Kunming was conducted to reflect the status of preschool education services in the capital city of Yunnan using the internationally- recognized ECERS-R rating scale. A sample of 24 kindergartens was drawn according to the overall preschool education quality at the county/district level, as well as type of ownership (public or private). A PPS (probability proportional to size) random sampling was used in two sub-groups (counties with high and low preschool education quality) and further stratified by ownership types. Table 4.1 summarizes the distribution of the sample result. Table 4.1 Sample Size and Distribution Sample size Regions Rural 3 Rural-urban continuum 7 Urban 14 Geographic Plain area 21 Mountain area 3 Ownership Private 18 Public 6 Government rating Class 1 Grade 1 (Best 1 available) Class 1 Grade 2 3 Class 1 Grade 3 5 Class 2 Grade 1 4 Class 2 Grade 2 1 Class 3 Grade 1 1 Others and Unrated 9 Source: Kunming Preschool Survey 2013. The instrument for assessment is the Early Childhood Environment Rating Scale-Revisited (ECERS), translated by the Chinese scholars Zhenguo Zhao and Xin Zhou. The scale has 43 items and 470 indicators in the systemic evaluation of 7 subscales (space and furnishings, personal care routines, language-reasoning, activities, interactions, program structure, and 81 parents and staff), and is scored on a 7-point scale ranging from ”Inadequate” to ”Excellent.” The study aims to answer the following questions of interest:  What is the overall quality of kindergartens in Kunming?  What is the quality of Kunming kindergartens as compared with other domestic or international benchmarks?  Are there disparities of quality between rural and urban areas, between rated and unrated kindergartens, and between public and privately-owned schools? Key Findings and Results ECERS-R scores for the Kunming sample are low overall. The total average score for all 24 kindergartens is 90.5, or 2.17 for each item, which are both categorized as “Inadequate” according to the instrument. Among all schools, 5 kindergartens had scores rated “Minimum,” while the remaining 19 were categorized as “Inadequate.” There were also significant internal differences within the sample. Among the 7 subscales (as shown in Table 4.2), parents and staff had the highest average score of 3, which was the only scale categorized as “Minimum,”, while the remaining measures were classified as “Inadequate.” The program structure was rated the lowest of all (with only 1.69 out of 7). Table 4.2 Mean ECERS-R Scores and Ratings of Kunming Preschools Survey Mean Category Program Structure 1.69 Inadequate Activities 1.88 Inadequate Personal Care 1.88 Inadequate Language –Reasoning 1.91 Inadequate Space and Furnishing 2.26 Inadequate Interaction 2.36 Inadequate Parents and Staff 3.00 Minimum Source: Kunming Preschool Survey 2013. For each subscale and associated items, the “Inadequate” rating was the largest share of all sample kindergartens. The only exception was related to parents and staff, which included slightly more kindergartens rated as “Minimum.” Only two subscales had one kindergarten rated at the level of “Good.” There were significant gaps in ECERS-R scores as compared with available international research. Aboud (2006) evaluated Plan Bangladesh, a non-governmental half-day preschool program in rural Bangladesh, where in 2001, ECERS-R scores were taken as a quality assessment measure after five years of operations. Three rural sites were chosen and 22 preschools were assessed. The average sample score was 2.89. 82 Table 4.3 Comparison of Mean ECERS-R Scores between Kunming and Plan Bangladesh Preschools Kunming Plan Bangladesh Program Structure 1.69 2.64 Activities 1.88 2.47 Personal Care 1.88 2.21 Language–Reasoning 1.91 3.40 Space and Furnishings 2.26 2.23 Interactions 2.36 4.06 Parents and Staff 3.00 3.25 Total 2.17 2.89 Sources: Aboud (2006) and Kunming Preschool Survey 2013. In the World Bank’s 2008 publication Africa’s Future, Africa’s Challenge: Early Childhood Care and Development in Sub-Saharan Africa, Mwaura and Mohamed (2008) describe the impact study of a Madrasa preschool program in East Africa. A regional initiative involving Kenya, Uganda and Zanzibar, it sought to “create an effective, community-based and low-cost approach” to early childhood education among children with low social-economic background since the 1980s. The study compared ECERS-R scores of 32 Madrasa preschools with 21 non-program preschools in the region as part of the impact evaluation, and found that Madrasa Resource Centers (MRCs) were rated better on most subscales than non-program schools. Table 4.4 Comparative Mean ECERS-R Scores between Kunming and Madrasa Project Preschools in East Africa MRC Preschools Non-MRC Kunming Preschools Program Structure 5 3 1.69 Activities 3 2 1.88 Personal Care 4 3 1.88 Language – 5 3 1.91 Reasoning Space and 4 2 2.26 Furnishings Interactions 6 4 2.36 Parents and Staff 4 4 3.00 Total 4.43 3 2.17 Sources: Mwaura and Mohamed (2008) and Kunming Preschool Survey 2013. 83 Goelman (2006) reported the results of a quality study of 326 classrooms in 239 participating Canadian child care centers, the largest and most extensive study ever undertaken in Canada. ECERS-R scores were used to measure the quality of classroom learning environments. The total mean score was 4.83 for non-profit centers, and 4.44 for commercial centers. Across 7 provinces participating in the research, average scores ranged from 4.09 in Saskatchewan to 5.56 in British Columbia. The total sample mean was 4.71. The most recent and comparable research was the Brazil National Survey of 138 preschools conducted in 2010, and supported by the Brazilian Ministry of Education and the Inter-American Development Bank. In general, the comparison showed that Brazil had higher scores in 6 out of 7 subscales. Kunming’s kindergartens had scores only slightly higher in parents and staff, although lagging behind mostly in the subscales of interaction, personal care and language–reasoning. The following table (Table 4.5) summarizes the average itemized score of the comparison. Table 4.5 Comparative Mean ECERS-R Scores between Kunming and Brazilian Preschools Brazil Kunming Program Structure 2 1.69 Activities 1.9 1.88 Personal Care 3.1 1.88 Language –Reasoning 2.8 1.91 Space and Furnishing 2.4 2.26 Interaction 4.1 2.36 Parents and Staff 2.7 3.00 Total 2.6 2.17 Sources: Kunming Preschool Survey 2013 and Brazilian National Survey, 2010. In China, very little research on preschool quality has been systematically conducted using the ECERS-R. In 2008, Biyin and Zongshun (2008) conducted 40 classroom observations at 7 Exemplified Class 1, Grade 1 kindergartens in Beijing. Despite a non-randomized sample and the fact that research didn’t contain the last sub-scale (parents and staff), there are still some similarities between kindergartens in Beijing and Kunming in terms of high and low subscale score distribution (such as interaction, program structure and activities), which could possibly trace back to similar cultural roots in China. Still, the Beijing sample scored significantly higher than Kunming on all subscales. Table 4.6 depicts the comparison. 84 Table 4.6 Mean ECERS-R Scores between Kunming and Beijing Preschools Beijing Kunming Program Structure 3.55 1.69 Activities 3.29 1.88 Personal Care 5.10 1.88 Language –Reasoning 4.66 1.91 Space and Furnishings 3.75 2.26 Interactions 5.60 2.36 Parents and Staff - 3.00 Total 4.32 2.17 Sources: Hu and Zhu (2009) and Kunming Preschool Survey 2013. There are significant internal disparities within Kunming’s kindergartens. Comparisons between rated and unrated kindergartens were also drawn to reveal any internal differences. Rated kindergartens have higher and significant scores in all subscales. However, except for parents and staff being categorized as “Minimum,” all other subscales in both rated and unrated kindergartens were labeled as “Inadequate.” Of all 43 items, 12 of the rated kindergartens were scored at “Minimum” (with a score higher than 3), while only 3 of the unrated kindergartens had the same level. Due to sample sizes, there were no significant differences for different rating status among all rated kindergartens ( for example, Class 1, Grade 1 rating versus Class 1, Grade 2 rating). Table 4.7 Mean ECERS-R Scores between Rated and Unrated Kindergartens in Kunming Rated Kindergartens Unrated Kindergartens Program Structure 1.08 1.94 Activities 1.23 2.14 Personal Care 1.22 2.16 Language –Reasoning 1.36 2.13 Space and Furnishings 1.48 2.58 Interactions 1.57 2.69 Parents and Staff 2.07 3.38 Total 2.46 1.45 Source: Kunming Preschool Survey 2013. Kindergartens located in urban, urban-rural continuum, and rural areas followed a similar pattern. The urban and urban-rural continuum kindergartens scored the highest. The 85 parents and staff subscale was rated as “Minimum,” with all of the remaining subscales classified as “Inadequate.” Further, among rural kindergartens, 36 items out of 43 (74 percent) had the very lowest score of 1, reflecting a very poor quality in rural Kunming by international standards. Table 4.8 Mean ECERS-R Scores in Kunming, by Area Rural-Urban Urban Area Rural Area Continuum Program Structure 1.86 1.63 1 Activities 2.05 1.88 1.07 Personal Care 2.18 1.64 1.06 Language -Reasoning 1.88 2.18 1.42 Space and Furnishing 2.55 2.14 1.17 Interaction 2.61 2.34 1.17 Parents and Staff 3.34 3.05 1.28 Total 2.38 2.16 1.17 Source: Kunming Preschool Survey 2013. Public kindergartens on average had higher scores than private kindergartens. The difference was largest again in for the subscale parents and staff, and relatively small for space and furnishings, interactions, language-reasoning, program structure and personal care. Due to the sample size, these differences were not statistically significant. Table 4.9 Mean ECERS-R Scores in Kunming, by Ownership status Public Kindergartens Private Kindergartens Program Structure 2.12 1.61 Activities 2.23 1.79 Personal Care 2.28 1.81 Language –Reasoning 2.29 1.83 Space and Furnishings 2.56 2.14 Interactions 2.82 2.30 Parents and Staff 3.7 2.74 Total 2.61 2.05 Source: Kunming Preschool Survey 2013 86 Summary and Discussion The research and evaluation instrument presented here represent the first ever effort in Yunnan province to measure quality (or quality proxy) of preschools using an international rating scale. Using the ECERS-R as the assessment instrument, the kindergarten survey conducted in Kunming provides an account of the quality and atmosphere of preschool education services in the capital of Yunnan. Importantly, the research team was able to provide a comparative perspective. On an absolute scale, Kunming’s kindergartens scored low on all seven aspects of the ECERS-R assessment, receiving “inadequate” on “program structure, activities, personal care, language - reasoning, space and furnishings, and interactions,” and a “minimum” score on “parents and staff.” Compared to select preschool programs in Bangladesh, Kenya, and Canada, Kunming also lags behind in its preschool services. Within the city, rated kindergartens on average scored higher than unrated ones, and urban kindergartens performed better than rural kindergartens. The research team believes that the very low scores of Kunming’s kindergartens according to the standard ECERS-R reflect both deficiencies in quality of service provision in Yunnan and Kunming, as well as some cultural habits that are unique to China. First, it is widely believed that joining collective activities and obedience are symbols of good behavior in China. Little attention has been given to promoting children’s own independence. This resulted in a situation in most observed kindergartens whereby children were asked to do only what the teachers had planned for them. This was in turn reflected in low scores in items such as nap/rest, encouraging children to communicate, discipline, staff-children interaction and children interaction, and so on. Second, most kindergartens were not equipped with sufficient material to support a diverse and innovative. It was reflected in low item scores in many areas including blocks, sand/water, dramatic play, use of TV/computer, books and pictures, furnishings for relaxing and comfort, and space for privacy. This suggests that there might be a lack of awareness of or attention to children’s personal needs in development. Inadequate resources could also be one possible cause. Third, health/safety practices and construction standards in Kunming’s kindergartens need great improvement. This was primarily reflected in low scores on meals/snacks, toileting/diapering, health practices, room for play, and indoor space. There was also a lack of ongoing support to kindergarten teachers, which relates closely to the above deficiencies. Low scores on the provision of professional needs of staff, staff interaction, opportunities for professional growth were also key indicators. Based on the aforementioned research findings, some important policy lessons may be drawn. Overall, it may be advisable to develop a more sophisticated instrument for assessing the quality of preschools, beyond the traditional focus on hardware and other inputs, and with more attention to curriculum, activities, and teacher-child interactions. The ECERS-R is only one example of such an instrument, which is being adopted more widely. Regular 87 assessment of the relatively “weak” dimensions can be used as a management strategy and tool to improve the classroom environment and interactions. Feedback from such evaluations can be used by teachers and preschool directors to improve teaching and learning practices in the classrooms. Results from continuous assessment can also be used to inform the development of preschool teacher training programs. A pilot study in Chongqing, using a similar scale, demonstrates that well-designed teacher training programs can have an impact on observed classroom behavior (Huang and Song, 2012). A group of teachers were randomly selected to attend a two-weekend (24 hours) training program during which they received various forms of training on preschool pedagogy, including lectures, school visit, and group presentation, using children-friendly materials developed by UNICEF and Ministry of Education. The teachers that received the training exhibit on average an improvement in the Wilson Scale score of 0.30, as compared to only 0.16 of the control group. The newly-published MOE Development Guidelines for Children 3-to-6–year-old (3-6 岁儿童 学习与发展指南) represents the first effort by the government to provide standards for childhood development and learning, not only for inputs, but also for process as well as outcomes. These guidelines can be used to steer the development of preschool teacher training modules that can in turn be implemented to improve the overall quality of preschool teachers in Yunnan. 88 Part V. Study of Preschool Teachers in Yunnan Similar to most other countries, a key constraint to the provision of early childhood education is the availability of qualified personnel who are motivated and well supported to teach and care for the children. Yunnan is no exception. The research team used a combination of teacher and director surveys, as well as a review of existing policies, statistics and other related studies to examine key aspects related to preschool teachers, including recruitment, career ladder, teacher training and support, and pay and compensation. Future demand for preschool teachers is further projected based on a few plausible scenarios and the latest population estimates. To the extent possible, relevant international comparisons are also provided. Finally, we also provide a brief summary of international comparisons of preschool teachers based on qualifications, salaries, and teacher support. Description of the Teachers/Director Survey Specific preschool teacher and director questionnaires were developed by preschool education experts focusing on five main aspects of basic characteristics: recruitment requirements and practices, teacher career ladder, teacher support, and pay and compensation. The Questionnaire is attached in the Annex. Through structured interviews with the preschool principal and the teachers, the study intended to answer the following questions: 1) What are the qualifications, recruitment process, and career development of preschool teachers in Yunnan? 2) What is the support mechanism for these teachers? 3) What is the status of teacher payment and compensation? This study sample includes the same 24 kindergartens selected for the preschool quality study, plus two urban kindergartens drawn through convenience sampling, and encompassing public, private, urban, and rural schools in Kunming. For each kindergarten, the researchers conducted a structured interview with the principal and two teachers recommended by the principal. A total of 52 kindergarten teachers and 26 principals were interviewed. Table 5.1 presents the distribution of the kindergartens. Table 5.2 presents the descriptive statistics of the teachers with regard to their age and years of work experience. 89 Table 5.1 Sample Size and Distribution Sample size Regions Rural 11 Urban 15 Ownership Private 20 Public 6 Government rating Grade 1 (Best available) 10 Grade 2 7 Grade 3 2 Unrated 7 Source: Kunming Preschool Survey 2013. Table 5.2 Mean and Standard Deviation of Age and Years of Preschool Work Experience of Kindergarten Teachers in the Sample Age Preschool Work Experience (year) Mean N SD Mean N SD Public Urban 32.75 8 4.16 12.75 8 4.56 Rural 36.25 4 7.41 12.00 4 5.52 Total 33.91 12 5.38 12.50 12 4.70 Private Urban 30.22 22 6.81 10.04 22 6.67 Rural 29.94 18 5.99 5.00 18 4.32 Total 30.10 40 6.38 7.77 40 6.21 Source: Kunming Preschool Survey 2013. Note: N= number of teachers; SD=standard deviation. . Key Findings and Results Preschool Teacher Qualifications The entry requirements for preschool teachers in Yunnan are based on laws and regulations promulgated by the People’s Congress and the State Council. In particular, the Ministry of Education sets forth and supervises the overall operation of teacher training, certification, and qualifications management. A rigorous application, educational requirement, Mandarin test, class teaching, recognition and certification process jointly compose the entry system of preschool teachers in China. However, as the research indicated, compared with the 90 specific and strongly-enforced requirements in other developed nations, the qualification requirements in China are sometimes compromised because of the over-emphasis on academic certificates and test scores. While the Teachers Law of the People's Republic of China stipulates that all preschool teachers need to have a vocational senior secondary degree or above, Yunnan released a provincial Preschool Education Regulations in 2013 which requires preschool teachers to have at least a junior college degree or pass relevant government tests. Under the new Regulations, the entry educational level for preschool teachers in the sample is on average relatively low. Upon entering the teaching force, 51.9 percent of preschool teachers have a senior vocational /high school degree, and 42.3 percent have a junior college (Dazhuan) degree or higher. Teachers in rural kindergartens have lower qualifications than their urban counterparts. Indeed, 46.7 percent of urban preschool teachers have a junior college degree or higher, whereas the percentage of rural teachers with the same educational level is only 36.4 percent. In addition, 9.1 percent of rural teachers have only a lower secondary degree (see Table 5.3). Table 5.3 Entry Educational Level of Preschool Teachers Entry Educational Level Vocational Senior Secondary/High Junior Lower School College College secondary Total Public/Private Public f 8 3 1 0 12 % 66.7 25.0 8.3 .0 100.0 Private f 19 15 3 3 40 % 47.5 37.5 7.5 7.5 100.0 Total f 27 18 4 3 52 % 51.9 34.6 7.7 5.8 100.0 Urban/Rural Urban f 15 12 2 1 30 % 50.0 40.0 6.7 3.3 100.0 Rural f 12 6 2 2 22 % 54.5 27.3 9.1 9.1 100.0 Total f 27 18 4 3 52 % 51.9 34.6 7.7 5.8 100.0 Source: Kunming Preschool Survey 2013. In-service teachers seek ways to improve their educational level through various channels. In the sample, one hundred percent of in-service public kindergarten teachers and 77.5 percent of private kindergarten teachers have met the requirement of obtaining at least a junior 91 college degree. In rural kindergartens, however, 31.8 percent of teachers still have not obtained the required educational level (see Table 5.4). 92 Table 5.4 Current Educational Level of Preschool Teachers Current Educational Level Vocational Senior Junior Lower Secondary College College Secondary Total Public/Private Public f 0 2 10 0 12 % 0.0 16.7 83.3 .0 100.0 Private f 6 22 9 3 40 % 15.0 55.0 22.5 7.5 100.0 Total 6 24 19 3 52 11.5 46.2 36.5 5.8 100.0 Urban/Rural Urban f 1 14 14 1 30 % 3.3 46.7 46.7 3.3 100.0 Rural f 5 10 5 2 22 % 22.7 45.5 22.7 9.1 100.0 Source: Kunming Preschool Survey 2013. However, a significant proportion of private and rural kindergarten teachers do not have a teacher licensure. According to the Yunnan Preschool Education Guidelines, preschool administrators, teachers, and caregivers need to have appropriate teacher licensure or certification. This requirement has not been completely fulfilled based on survey results. On average, 67.3 percent of kindergarten teachers have appropriate teacher licensure. There is also some disparity across school types: whereas 100 percent of public kindergarten teachers have teacher licensure, only 57.5 percent of private kindergarten teachers and 37.8 percent of rural kindergarten teachers have met this requirement. Recruitment Public kindergartens face a higher level of teacher shortages than private kindergartens. However, the teacher recruitment rate is lower in public kindergartens. Of the public kindergarten principals interviewed, 82.3 percent indicate that they face teacher shortages, which is higher than the 50 percent rate for private kindergartens. However, only 16.7 percent of public kindergartens recruit new teachers every year, as compared to 77.7 percent of private kindergartens that conduct annual recruitment. This suggests a higher teacher mobility in the private kindergarten setting, as well as potentially more constraints in the hiring process for public kindergartens. Principals express a general sense of difficulty in recruiting qualified preschool teachers. Seventy-two percent of kindergarten principals said that the recruitment process of new 93 teachers is unsatisfactory. Furthermore, 73.9 percent of principals think that the newly-recruited teachers cannot meet the expected standards. During the interviews, principals pointed out that a general shortage of preschool teacher candidates and the inadequate institutional training of those candidates are two major challenges in the recruitment process. In addition, high teacher mobility poses a challenge to sustaining a qualified teaching force. In response, most kindergartens use contract teachers and/or substitute teachers to fill the gap in teacher demand. Some kindergartens have also tried to decrease enrollment or increase class size. Preschool Teacher Career Ladder Mechanism The professional title system is essential to the career ladder mechanism. Yunnan currently adopts the 1986 evaluation system which assigns kindergarten teachers different levels of professional titles based on their educational background, teaching experience, attainment of various certificates, and so on. Teachers’ titles reflect their work performance and can be directly tied to salaries. Therefore, having an effective career ladder mechanism through professional titles can help provide incentives to teachers to pursue career development, as well as promote teacher retention. Based on sample results, 83.3 percent of rural private teachers do not have a professional title, whereas 100 percent of public kindergarten teachers and 50 percent of private kindergarten teachers do have various levels of professional titles. Teacher Support The pupil-teacher ratio is, on average, suboptimal across all kindergarten types. According to the Kindergarten Teachers and Administrators Provision Standards (Tentative), the recommended pupil-teacher ratio should be below 9-to-1.34 Nevertheless, pupil-teacher ratios are still high. Table 5.5 shows the average pupil-teacher ratio across kindergarten types.35 Although the overall average of 15:1 may sound reasonable, the disparity can be very large with the maximum ratio reaching 33 students for one teacher/caregiver. Rural private schools tend to have the highest pupil-teacher ratios. 34 including full-time teachers, caregivers, healthcare and administrative personnel, and other supporting staffs – not including part-time teachers 35 Teachers, caregivers, and other administrators are all included to calculate the ratio. 94 Table 5.5 Pupil-Teacher Ratio across Kindergarten Types Mean N SD Minimum Maximum Public Urban 12.31 12 1.85 9.82 14.74 Rural 18.51 6 10.09 9.31 27.72 Total 14.38 18 6.419 9.31 27.72 Private Urban 11.46 33 7.45 5.44 33.20 Rural 19.68 26 8.08 7.50 32.00 Total 15.08 59 8.70 5.44 33.20 Total Urban 11.69 45 6.43 5.44 33.20 Rural 19.46 32 8.32 7.50 32.00 Total 14.92 77 8.19 5.44 33.20 Source: Kunming Preschool Survey 2013. Note: N= number of schools; SD= standard deviation. Professional Development In 2003, China’s State Council issued the MOE Guideline on Preschool Education Reform and Development, which placed an unprecedented heavy emphasis on post-service training of teachers. Accordingly, since 2010, annual national training supported by the central government was designed and put forward. It is composed of tiered sessions for experienced head teachers, principals and new entrants. In 2012, 4,403 teachers in rural Yunnan participated in the program. In addition, Yunnan has been receiving continuous support from foreign non-profit organizations. Save the Children UK, for example, has provided funding for Yunnan to establish early childhood centers as a platform for ECD-related activities and resources. The organization has also provided ECD training for parents and teachers. Kindergartens place considerable emphasis on professional development. Preschool teachers have access to various types of training. Based on sample results, 92.8 percent of kindergartens provide in-service teacher training. The training takes on different forms, ranging from pedagogy to student safety. Teachers can receive in-school training or seek training opportunities outside of their kindergartens. Kindergarten principals indicated a supportive attitude toward their teachers’ professional development. The creation of teaching and research groups is an example of administrative support for teacher training. Ninety-three percent of urban kindergartens and 77.4 percent 95 of rural kindergartens have established teaching and research groups to facilitate communication among teachers. Teachers belonging to the same group gather together regularly to discuss pedagogy and evaluate lessons. Teacher Salaries and Compensation Teacher salary is an important factor for teacher recruitment and retention. In China, the payment of preschool teachers has been closely related to geographic location, academic credentials and qualification levels. These, in turn, depend heavily on the ownership and staffing status of preschools. In general, public preschool teachers have more years of education. Indeed, a larger proportion of them have higher qualifications, and consequently a higher pay as compared to teachers in private preschools. In addition, both the qualifications and salary of preschool teachers in the rural area are lower than those of their urban counterparts. Disparities exist in teacher salaries across kindergarten types. The average salary for kindergarten teachers in the sample is approximately 2,200 RMB. However, urban public kindergarten teachers enjoy the highest salary level of 2,900 RMB, whereas rural private teachers receive the lowest average salary of approximately 1,600 RMB. Table 5.6 compares salary levels across kindergarten types. Table 5.6 Salaries of Kindergarten Teachers across Kindergarten Types Mean N SD Minimum Maximum Public Urban 2901.67 12 575.91 2000.00 4000.00 Rural 2263.64 33 621.40 1100.00 4200.00 Total 2433.78 45 667.20 1100.00 4200.00 Private Urban 2583.33 6 466.55 2000.00 3000.00 Rural 1605.56 27 741.92 800.00 4000.00 Total 1783.33 33 792.41 800.00 4000.00 Total Urban 2795.56 18 549.98 2000.00 4000.00 Rural 1967.50 60 749.00 800.00 4200.00 Total 2158.59 78 787.36 800.00 4200.00 Source: Kunming Preschool Survey 2013. Note: N= number of teachers; SD= standard deviation.  Teachers within the staff quota system have much higher salaries than those outside of the staff quota system. Public kindergartens in China have to subscribe to the staff quota system regulated by 96 governmental authorities. The staff quota system defines the number of employees each kindergarten can hire. Not only does the government directly pay for those teachers’ salaries, they also enjoy full benefits of medical insurance, social security, and other subsidies. However, the quota is often insufficient for teacher needs in Yunnan, resulting in a severe teacher shortage. Within this context, kindergartens have to resort to hiring temporary teachers or contracted teachers outside of the staff quota system. However, teachers hired within the quota system have significantly higher salaries than those hired outside of the system in both urban and rural settings. Table 5.7 shows the comparison of salaries for teachers within and outside of the quota system. Table 5.7 Salary Level for Teachers within and outside and Staff Quota System Quota Mean N SD System Urban Within 2680.00 9 446.21 Outside 2238.10 21 670.43 Total 2370.67 30 638.22 Rural Within 2471.43 7 543.80 Outside 1350.00 15 376.54 Total 1706.82 22 681.79 Total Within 2588.75 16 485.69 Outside 1868.06 36 714.66 Total 2089.81 52 729.87 Source: Kunming Preschool Survey 2013. Note: N= number of teachers; SD= standard deviation. On an aggregate scale, preschool teachers do not match primary school teachers in annual compensation, according to the 2010 China Labor Statistical Year Book. From the highest to the lowest, high school teachers received an average salary of 47,693 RMB ($7,790), middle school teachers 34,169 RMB ($5,583), primary school teachers 31,037 RMB ($5,071), and preschool teachers only 18,533 RMB ($3,028). In Yunnan, other sources of national statistics also indicated that the average income for preschool teachers in 2010 is 82 percent lower than the national average salary. and 50 percent lower than the average teachers’ income at all levels. 97 Table 5.8 Average Preschool Teacher Salary in Yunnan, 2003 -2010 2010 2009 2008 2007 2006 2005 2004 2003 Average Budgetary 91.0 45.6 40.2 37.1 32.3 29.0 26.5 21.7 Expenditure (Personal) in millions RMB Number of Teachers 47,001 40,322 37,011 33,826 30,944 28,284 26,410 23,857 Average Preschool Teacher 19,357 11,306 10,872 10,961 10,461 10,231 10,027 9,104 Salary (RMB) Provincial Average Salary 30,000 26,992 24,030 20,481 18,711 16,980 —— —— (RMB) National Average 23,183 9,930 9,007 8,018 6,619 6,022 5,623 5,140 Preschool Teacher Salary (RMB) National Average Salary 36,539 32,244 28,898 24,721 20,856 18,200 15,920 13,969 (RMB) Sources: MOE- various years, Yunnan Bureau of Education- various years. Within the province and compared with other occupations, education in Yunnan ranked 6 out of 11 industries in average salary in the province. Table 5.9 Average Salary (RMB 10,000), by Occupation in Yunnan, 2011 National Yunnan Average 4.18 3.54 Finance 8.11 8.56 Water, Electricity, Gas 5.27 5.15 Information System 7.09 4.87 Transportation 4.71 4.63 Research 6.43 4.4 Education 4.32 3.71 Public Management 4.21 3.57 Mining 5.22 3.44 Manufacturing 3.67 3.4 Construction 3.21 2.58 Restaurant 2.75 1.96 Sources: MOE- various years, Yunnan Bureau of Education- various years. 98 Teachers revealed dissatisfaction with their salary levels. Almost two-thirds of preschool teachers surveyed in this study are either dissatisfied or very dissatisfied with their salaries. Many kindergarten teachers think their salaries do not correctly reflect their workload or work intensity. Demand and supply for preschool teachers in Yunnan In the past ten years, the number of kindergartens in Yunnan has tripled from 1,530 (2001) to 4,768 (2012). In the meantime, the number of teachers had increased almost twofold from 13,100 to 35,600. However, the enrollment also doubled from 627,000 to 1.1 million students. Accordingly, pupil-teacher ratios dropped from 47.86 to 31.53. However, as compared with the national average of 26, it is still significantly higher. According to the latest MOE Kindergarten Staffing Standards (2013), pupil-teacher ratios at kindergartens should ideally be below 15. Several estimates were made based on different ratio assumptions for Yunnan’s preschool teacher demand in the next decade while taking into consideration estimated changes in student enrollment. Estimated gross enrollment rates from 2011 to 2020: Table 5.10 Estimated Gross Enrollment Rate, 2011-2020（%） Gross Enrollment 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Rate 1-yr Preschool 77 79 81 83 85 87 89 91 93 95 2-yr Preschool 66 67 68 69 70 72 74 76 78 80 3-yr Preschool 56 57 58 59 60 62 64 66 68 70 Source: Zhao, Zhou (2013). 99 Scenario 1: Pupil-teacher ratio maintained at 2012 level: Table 5.11 Estimated Demand for and Shortages of Preschool Teachers in Yunnan, 2013-2020 2013 2014 2015 2016 2017 2018 2019 2020 Estimated Demand 38,053 39,149 40,394 41,129 42,237 43,300 44,366 45,426 for Preschool Teachers Estimated 2,472 3,568 4,813 5,548 6,656 7,719 8,785 9,845 Shortage of Preschool Teachers Source: Zhao, Zhou (2013). At the current pupil-teacher ratio, the demand for preschool teachers in Yunnan will increase by about 27.7 percent from 2012 to 2020, simply due to the overall expansion of enrolment. Different prefectures will face varying degrees of shortages: for example, Zhaotong, Puer and Wenshang will experience the highest demand for preschool teachers. Scenario 2: pupil-teacher ratios drop to 15 by 2020: Table 5.12 Estimated Pupil-Teacher Ratios, Demand for and Shortages of Preschool Teachers in Yunnan, 2013-2020 2013 2014 2015 2016 2017 2018 2019 2020 Pupil-Teacher ratios 29.7 27.86 26 23.38 21.6 19.94 17.2 15 Estimated Demand 40,410 44,321 49,001 55,484 61,673 68,489 81,355 95,516 for Preschool Teachers Estimated Shortage 4,829 8,740 13,420 19,903 26,092 32,908 45,774 59,935 of Preschool Teachers Source: Zhao, Zhou (2013). If Yunnan is to decrease the pupil-teacher ratio gradually to 15 (a ratio stipulated by the government) by 2020 while expanding preschool coverage, the shortage of teachers will reach almost 60,000. This indicates that an average of 7,492 new preschools teachers per year will be needed. 36 36 This estimate assumes that no teacher leaves the teaching force. 100 Scenario 3: Pupil-teacher ratios drop to 15 by 2015: In order to reflect the actual differences in pupil-teacher ratios among prefectures, a further assumption was made in the third scenario to estimate the demand of teachers by 2020. For those prefectures that already reached the national standards of 15 in 2012, the ratios would remain the same; for those that had yet to reach the target, there will be a 3-year window before the gaps are closed. The roadmap thus requires all prefectures to reach or maintain the pupil-teacher ratio above 15 by 2015. As a result of the estimate, Yunnan will face a general shortage of 60,000 new preschool teachers in 2020. Kunming, Qujing, Zhaotong, Honghe, Wenshang, and Dali will have the largest demand for new teachers, whereas Zhaotong, Honghe, Qujing, Wenshan, Lincang, Kunming, and Baoshan will have the largest shortage. Zhaotong will face the highest shortage of over 11,000 teachers in 2020. Table 5.13 Estimated Demand for Preschool Teachers in Yunnan, 2013-2020 2013 2014 2015 2016 2017 2018 2019 2020 Yunnan 云南 47,690 61,377 85,208 86,874 89,189 91,423 93,667 95,895 Kunming 昆明 10,842 12,344 14,037 13,952 13,926 13,897 13,871 13,845 Qujing 曲靖 6,062 7,749 11,381 11,465 11,630 11,793 11,953 12,115 Yuxi 玉溪 2,815 3,431 4,650 4,613 4,602 4,591 4,581 4,571 Baoshan 保山 1,570 2,457 5,019 5,060 5,131 5,192 5,254 5,314 Zhaotong 昭通 2,841 4,932 11,789 11,969 12,313 12,637 12,962 13,286 Lijiang 丽江 1,222 1,518 2,008 2,050 2,107 2,157 2,208 2,258 Pu’er 普洱 2,714 3,413 4,521 4,605 4,735 4,852 4,970 5,087 Lingcang 临沧 1,659 2,504 4,579 4,694 4,827 4,949 5,071 5,193 Chuxiong 楚雄 2,346 3,134 4,481 4,542 4,667 4,780 4,893 5,005 Honghe 红河 4,215 5,837 9,019 9,173 9,432 9,673 9,915 10,156 Wenshan 文山 3,263 4,486 6,862 6,956 7,148 7,330 7,512 7,693 Panna 版纳 1,151 1,456 2,086 2,094 2,127 2,155 2,184 2,211 Dali 大理 3,526 4,399 6,366 6,261 6,250 6,235 6,221 6,208 Dehong 德宏 1,216 1,704 2,723 2,755 2,801 2,841 2,881 2,920 Nujiang 怒江 566 795 1,106 1,115 1,147 1,177 1,207 1,236 Diqing 迪庆 316 439 635 610 627 642 657 671 Source: Zhao, Zhou (2013). Table 5.14 Estimated Shortage of Preschool Teachers in Yunnan, 2013-2020 2013 2014 2015 2016 2017 2018 2019 2020 Yunnan 云南 12,109 25,796 49,627 51,293 53,608 55,842 58,086 60,314 Kunming 昆明 1,189 2,691 4,384 4,299 4,273 4,244 4,218 4,192 Qujing 曲靖 608 2,295 5,927 6,011 6,176 6,339 6,499 6,661 Yuxi 玉溪 852 1,468 2,687 2,650 2,639 2,628 2,618 2,608 Baoshan 保山 446 1,333 3,895 3,936 4,007 4,068 4,130 4,190 Zhaotong 昭 1,442 3,533 10,390 10,570 10,914 11,238 11,563 11,887 101 通 Lijiang 丽江 298 594 1,084 1,126 1,183 1,233 1,284 1,334 Pu’er 普洱 1,090 1,789 2,897 2,981 3,111 3,228 3,346 3,463 Lingcang 临沧 661 1,506 3,581 3,696 3,829 3,951 4,073 4,195 Chuxiong 楚 691 1,479 2,826 2,887 3,012 3,125 3,238 3,350 雄 Honghe 红河 1,014 2,636 5,818 5,972 6,231 6,472 6,714 6,955 Wenshan 文 703 1,926 4,302 4,396 4,588 4,770 4,952 5,133 山 Panna 版纳 261 566 1,196 1,204 1,237 1,265 1,294 1,321 Dali 大理 850 1,723 3,690 3,585 3,574 3,559 3,545 3,532 Dehong 德宏 281 769 1,788 1,820 1,866 1,906 1,946 1,985 Nujiang 怒江 194 423 734 743 775 805 835 864 Diqing 迪庆 163 286 482 457 474 489 504 518 Source: Zhao, Zhou (2013). Based on the results of the third scenario, the research team further estimated the number of teachers required with different qualifications by 2020. Table 5.15 Estimated Number of Teachers Required, by Level of Educational Background Year Total Junior Vocational Senior Vocational Tertiary and Below and above Number % Number % Number % 2012 23,178 12,001 33.73 23,580 66.27 6,462 18.16 2013 47,690 15,261 32.00 32,429 68.00 9,052 18.98 2014 61,377 18,610 30.32 42,767 69.68 12,177 19.84 2015 85,208 24,369 28.60 60,839 71.40 17,638 20.70 2016 86,874 23,352 26.88 63,522 73.12 18,730 21.56 2017 89,189 22,440 25.16 66,749 74.84 19,996 22.42 2018 91,423 21,430 23.44 69,993 76.56 21,283 23.28 2019 93,667 20,344 21.72 73,323 78.28 22,611 24.14 2020 95,895 19,179 20.00 76,716 80.00 23,974 25.00 Source: Zhao, Zhou (2013). Although the supply of preschool teachers from teachers’ colleges and relevant vocational programs is increasing substantially, it is still insufficient to meet the surging need. The majority of shortages (53,136) will be for teachers with senior vocational or tertiary education according to the research. This translates into about 6,642 teachers each year from 2013 to 2020 (taking 2012 as baseline). This figure is much higher than the number of enrollments or graduates in 2012 (Table 5.16). As the number of students in primary school decreases, transferring the surplus of primary 102 school teachers to preschool education could be one feasible solution to the shortage. However, this carries the risk of making preschool education too “primary-education-focused.” If this transfer is adopted, practical training is crucial to transitioning these teachers to the preschool setting. Some parts of China have been exploring new ways to increase the supply of rural preschool teachers. One of the experiments supported by the China Development Research Foundation was conducted in Xundian, Yunnan based on the “walk-teaching (走教)” model of volunteer teachers. The pilot utilized empty school houses as teaching sites in separated villages. The pilot program recruited and subsidized preschool teaching volunteers, and sent them between sites to offer preschool education that lasted more than 3 times as long per session. The volunteer teachers needed to be able to give instructions in Mandarin. The pilot model considerably improved upon the children’s daily habits, language expression, observation abilities, and so on. Given the success and low cost of this operation, the Ministry of Education expanded the program to all volunteering counties in central-western provinces in 2011. Six countries in Yunnan participated, with 80 teachers in 2013. There are, however, still discussions about the program’s financial sustainability, as well as recognition and qualifications of volunteers. Table 5.16 Number of Enrollments, Admissions and Graduations at Teachers’ Colleges and Provincial Vocational Schools Majoring in Preschool Education in Yunnan, 2011-2012 Preschool 2011 2012 Education Admitted Enrolled Graduated Admitted Enrolled Graduated Junior 5,993 11,619 1,949 6,796 15,423 2,677 Vocational Tertiary 1,080 2,631 310 1,184 3,365 435 Senior 2,061 3,951 857 1,757 5,079 609 Vocational Source: Zhao, Zhou (2013). International Comparison This section presents a comparison of the status of preschool teachers in other countries such as United States, Japan, and Korea, focusing on three areas: qualifications, salary, and support. 103 Qualifications In the United States, the Early Childhood Teacher Education Guidelines for Four- and Five-Year Programs released by the National Association for the Education of Young Children frames preschool teacher preparation in the country. The guidelines are centered on six core standards: 1) Promoting childhood development and learning; 2) Building family and community relationships; 3) Observing, documenting, and assessing support for young children and families; 4) Using developmentally-effective approaches to connect with children and families; 5) Using content knowledge to build a meaningful curriculum; and 6) Becoming a professional. The guidelines are applicable to preschool teacher preparation programs across degree levels at tertiary institutions, and are used in a number of nationwide or statewide preschool teacher evaluations. They are also relevant to assessing any professional working in the preschool setting, including caregivers, public school teacher aides, child protection service workers, curriculum developers, and so on. Obtaining a teacher certification is the prerequisite for being able to teach in the United States. The specifications for earning a preschool teacher certification differ across states, but the general criteria overlap. For example, in Pennsylvania, teacher candidates have to meet the following requirements to receive a teacher certification: (i) hold a bachelor’s degree; (ii) complete an approved program of study leading to certification in the area of certificate requested; (iii) hold a 3.0 Grade Point Average (GPA); 4) successfully complete Pennsylvania state-required tests; and (v) meet all Pennsylvania requirements related to citizenship, moral, ethical and physical/mental fitness. In most states, teacher candidates must also pass the Praxis test, a teacher certification exam on reading, writing, mathematics, preschool content knowledge, curriculum, and instruction. Japan has very high preschool teacher standards. Regardless of kindergarten type, all teachers must obtain teacher certification, either through relevant college programs or through in-service training (for educators who have already worked in the field for a number of years). Released in 1949, the Teacher Certification and Implementation Law in Japan established the country’s teacher certification policy. There are two types of preschool teacher certifications: regular and temporary. The regular certifications require at least two years of tertiary-level early childhood education study and has lifelong validity. The temporary certification is granted to high school graduates who successfully complete the local preschool personnel examination. However, the temporary certification is valid for three years, and only for becoming teacher aides or caregiver aids. 104 Korea has a clear advancement system for preschool teacher qualifications from prospective teacher to teacher certification level 1. According to legal regulations, candidates who successfully pass the preschool certification exam can receive the title of prospective teacher. Only after finishing further relevant academic studies can the prospective teacher receive the teacher certification level 2. After 3 or more years of work experience or obtaining a Master’s degree, preschool teachers are then eligible for receiving a teacher certification level 1. Salary The salary level of kindergarten teachers differs across countries. According to the U.S. Department of Labor Statistics, in 2004, the salary for kindergarten teachers in the United States was significantly lower than for grade 1-3 elementary school teachers. By contrast, public kindergarten teachers in Korea and Japan are recognized as civil servants. Therefore, their salary level and social status are on par with elementary and secondary teachers. In Japan, preschool teacher salaries are 20 percent higher than that of an average civil servant. In Korea, graduates of a four-year preschool university program have an average salary of approximately 1,600 dollars, with subsequent annual increases. Table 5.17 compares the salary level of preschool teachers in a number of countries. Table 5.18 and Table 5.19 present a salary comparison across industries in Austria and Poland, respectively. Table 5.20 compares the monthly salary and weekly work hours of preschool teachers across countries. 105 Table 5.17 Kindergarten Teacher Salary Comparison across Countries Country Education Personnel Type Student Age Kindergarten Group Teacher Salary/Elementary School Teacher Salary (%) Australia Teacher 0-8 100 Belgium Preschool teacher 2.5-6 100 Czechoslovakia Preschool teacher 3-6 75 Teachers at daycare centers need to 0-7 81 have a Bachelor’s or Master’s Degree in Education. Finland Social studies teachers need to have a 0-6 related associate degree. Practical caregiving 0-7 Ireland Teacher 4-12 100 Caregiver/ aide 0-6 60 Italy Preschool teacher 0-3 or 3-6 100 The Preschool teacher 4-12 or 0-4 100 Netherlands Norway Preschool teacher 1-6 88-96 Portugal Preschool teacher 3-6 100 Teacher, caregiver 0-3 Sweden Preschool teachers, 闲暇 教 师 , and 1-7 100 elementary school teachers comprise the teaching force. United States Public school teacher 4-8 or 0-8 100 Pilot program teacher 0-5 Sources：Jiang, Y., Yan, J., and Xu L., 2013. 106 Table 5.18 Salary Comparison across Industries in Austria Industry Monthly Salary Weekly Work (US$) Hours Pilot 2565 38.5 Accountant 2179 38.5 Nurse 1684 40.0 Engineer 1499 38.0 Programmer 1427 38.5 Office Clerk 1338 38.5 Nursing Aid 1279 40.0 Psychologist 1271 40.0 Car Mechanic 1238 40.0 Teacher 1231 40.0 Driver 1151 40.0 Bread Baker 1188 38.5 Mail Carrier 1103 40.0 Miner 1147 38.5 Carpenter 1143 39.0 Firefighter 1134 40.0 Furniture Manufacturer 1027 40.0 Salesperson 925 38.5 Tailor 830 40.0 Hotel Receptionist 816 40.0 Sources：Jiang, Y., Yan, J., and Xu L., 2013. Table 5.19 Salary Comparison across Industries in Poland Industry Monthly Salary Weekly Work (US$) Hours Programmer 892 41.0 Miner 694 43.0 Engineer 664 41.0 Accountant 617 40.0 Nurse 563 22.0 Teacher 543 25.0 Firefighter 500 40.0 Driver 408 40.0 Office Clerk 398 41.0 Tailor 389 41.0 Mail Carrier 371 40.0 Car Mechanic 365 41.0 107 Bread Baker 363 42.0 Carpenter 324 41.0 Hotel Receptionist 318 42.0 Psychologist 296 35.0 Nursing Aid 296 35.0 Furniture Maker 281 41.0 Salesperson 264 41.0 Sources：Jiang, Y., Yan, J., and Xu L., 2013. Table 5.20 An International Comparison of Monthly Salary and Weekly Work Hours for Kindergarten Teachers Country Net Monthly Income (US$) Weekly Work Hours United States 4,055 36.6 Germany 3,309 40.0 Britain 3,568 32.5 Korea 2,096 39.7 Australia 2,742 39.1 Canada 2,236 31.1 Finland 2,311 36.4 Portugal 1,441 24.4 Austria 1,231 40.0 Thailand 388 38.0 Kuwait 1,506 36.0 Czechoslovakia 681 37.9 Mexico 651 38.0 Poland 543 25.0 Lithuania 300 29.4 Brazil 299 22.3 Slovakia 363 31.1 Romania 302 38.8 Sources：Jiang, Y., Yan, J., and Xu L., 2013. Teacher support Most countries in the world have realized the importance of improving preschool teacher quality as a means of advancing early childhood education. Many countries have implemented new policies to provide incentives for in-service training. Eliminating lifelong tenure for preschool teachers and regular teacher certification approval (United States and Japan), and linking teacher training with promotion (United States, Japan, and Britain) are examples of such incentives. In addition, many countries’ governments have made 108 substantial investments in preschool teacher training (Untied States, Japan, Britain, France, Sweden, and so on) to ensure sufficient funding. United States, Japan, and Korea all have a comprehensive set of standards for in-service training. To improve preschool teacher quality, the United States has promoted in-service training to preschool teachers. In addition, relevant regulations stipulate that teachers participate in regular in-service training and accrue university credits; otherwise their contract will not be renewed. Similarly, legal documents require kindergarten teachers to devote time to further studies in Japan. They can access a variety of training options ranging from university courses to workshops provided by ECD organizations. In Korea, various types of in-service training are designed to train new teachers, improve teachers’ pedagogy and practice, and so on. In particular, many countries have initiated specific policies to improve teacher quality in rural and poor regions. Examples include: the Education Renewal Zones initiative in Arkansas and the Rural Education Achievement Program in the United States, the Program to Reduce Educational Backwardness (Program para Abatir el Rezago Educativo) in Mexico, and the Country Areas Program in Australia. Table 5.21 presents the in-service training schemes across countries. Table 5.21 In-Service Training for Kindergarten Teachers Country Student Training Content Age Group Australia 0-8 Teachers receive subsidized training for a few days every year. In 2001, the Australian Ministry of Education launched the “Teachers for the 21st Century: Making the Difference” initiative. The government allocated 80 million Australian dollars to teacher training, administrator training, management investment, and incentive schemes. In particular, teacher training funding accounts for 90 percent of total funding. A variety of training programs and initiatives are provided to enhance teachers' basic qualifications, to improve the professional status of teachers, and to promote overall teacher quality. Austria 3-6 Subsidized training 3-5 days every year. Finland 0-7 Teachers must devote three days per year to planning and professional development. The government allocates approximately $30 million to teacher professional development. Korea 3-6 Training is provided by local teacher centers to all kindergarten teachers and caregivers. Mexico 3-6 Teachers receive subsidized training for a few days every year. They must also receive three days of professional development training every year. Portugal 3-6 Teachers have to receive at least 56 hours of training, provided by local teacher centers and universities. 109 Sources: Jiang, Y., Yan, J., and Xu L., 2013. Sahlberg, P. (2010). The Secret to Finland’s Success: Educating Teachers. Standard Center for Opportunity Policy in Education. Summary and Discussion The preschool teacher study conducted in Kunming reflects the current status of the preschool teaching force in the city with regard to qualifications, recruitment, the career ladder system, teacher support, and payment and compensation. It should be noted that as the capital of Yunnan, Kunming is one of the most developed cities in the province. Therefore, the results are likely to be more favorable than those of other parts of the province. This survey provides a glimpse of the existing problems in the preschool teaching force in the province as a whole, reflecting provincial disparities. With respect to qualifications, recruitment and the career ladder, preschool teachers have relatively low educational qualifications when first entering the teaching force. However, they are able to improve their educational qualifications through in-service education. Kindergartens in general face various levels of teacher shortages, as well as difficulties in hiring qualified teachers. The common strategy kindergartens adopt is to hire temporary or contractual teachers. The current career ladder mechanism in place is the multi-level professional title evaluation. In the area of teacher support, the pupil-teacher ratio is still much higher than recommended by related guidelines. Teachers do have access to various types of professional development opportunities, both within their kindergartens and outside of the school setting. Teaching and research groups are examples of administrative support for teachers’ professional development. The topic of teacher salaries and compensation garnered much dissatisfaction from preschool teachers. There is a marked difference in teacher salaries between public and private kindergartens. In addition, teachers hired within the quota system receive higher salaries than their counterparts hired outside of the system. Most teachers are dissatisfied with their salary levels. Projections based on the most progressive scenario of reaching a pupil-teacher ratio of 15:1 by 2015 with the current population projections demonstrate that in 2020, Yunnan will face a general shortage of 60,000 new preschool teachers. Zhaotong will face the highest shortage of over 11,000 teachers. Kunming, Qujing, Zhaotong, Honghe, Wenshang, and Dali will have the largest demand for new teachers, whereas Zhaotong, Honghe, Qujing, Wenshan, Lincang, Kunming, and Baoshan will have the largest shortage. The majority of shortages (53,136) will be for teachers with senior vocational or tertiary education according to the research. This translates into about 6,642 teachers each year from 2013 to 2020. This is much higher than the number of enrollment or graduates in 2012. 110 The study highlights the disparities between different types of preschools. Public preschools in urban areas fare the best of all categories of entry qualification, access to professional development, higher salaries and other benefits. At the other end of the spectrum are the teachers serving in rural private kindergartens that tend to have the lowest entry qualification, the least access to professional development and other career opportunities, and the lowest pay and compensation. Insufficient funds, lack of career development, and weak incentives make it very hard to maintain a qualified teaching force, especially in rural areas and for the private preschools. 111 Part VI: Policy and Program Implications Based on the in-depth analyses presented here, we propose policy and programming interventions in the following six main ways: (i) increasing public financing on preschool education, especially for the disadvantaged; (ii) expanding service delivery models; (iii) improving the quality of preschools; (iv) strengthening family and parental education; (v) Increasing awareness of secondhand smoke’s harmful effects; (vi) developing a preschool teaching force; and (vii) moving toward integrated ECD provision. Increasing public financing of preschool education and diversifying financing/delivery models Preprimary education is the most seriously underfunded sector in China, accounting for 13 percent of the total number of students in the system, but receiving only 2 percent of the budget. Internationally, for some advanced countries preprimary education claims 6-8 percent of the total education budget. In Yunnan, only 0.1 percent of GDP is spent on preschool education, as compared to the 0.2 percent national average. To improve the coverage and efficacy of its ECDE programs, particularly for the rural poor segments of the population, Yunnan and/or some cities/counties within Yunnan may consider further increasing their level of financial commitment. This can be done through a variety of approaches. The most progressive approach which has already been piloted by some local governments is the incorporation of preschool education into the free public education system. Table 6.1 provides a list of local governments implementing such policies. Table 6.1: Examples of Extended Free Public Education in China Type of Length of Extension Extended Free Basic Duration Implemented Areas Free Basic Education (year) Education (in years) 1 10 Xiamen (Fujian), Yan’an Zhidan (Shaanxi) Pre-School 3 12 Dongying Hekou (Shandong) Zhuhai (Guandong), Wuqing (Tianjing), Ningbo Yinzhou(Zhejiang), Shanshan (Xinjiang), Haixi (Qinghai), Fuzhou Mawei (Fujian), Ankang Zhenping (Shaanxi), Senior High 3 12 Xiangxi (Hunan), Yuxi Hongta (Yunnan), Zoucheng School (Shandong), Nilka (Xinjiang), Inner Mongolia, Nansha (Guangzhou), Menghai (Yunnan), Linfen Gu (Shanxi), Linfen Pu(Shanxi), Linfen Ning (Shanxi), Shuozhou Pinglu (Shanxi) Pre-School + Ankang Ningshan (Shaanxi), Wuqi (Shaanxi), Shenmu Senior High 3+3 15 (Shaanxi), Fugu (Shaanxi), Yanzhou (Shandong), Changzhi School (Shanxi), Tibet Note: The name of the province/city is indicated inside the parentheses. 112 The cost of such provisions can be easily projected based on current per student expenditures and the total number of preschool children in Yunnan. Understandably, such a policy may be too expensive for a province-wide rollout. Another approach is to set up targeted funding for preschool provision in poor, border, and minority counties. Our research clearly shows that children living in these counties tend to have worse home care conditions, lower developmental status. In addition, they are less likely to be enrolled in preschool education due to lack of availability. It would be important to map out the gaps in service provision from village to village, identify priority villages, and establish provincial fiscal transfer mechanisms, as such villages are most likely to be poor and unable to afford new provisions themselves. A diverse and innovative set of financing and delivery mechanisms can be considered. The formal, but expensive, center-based full-time model of care is not the only viable model. Other models employing flexible service delivery could be adopted in addition to formal center-based care. Home-based care, parenting education, and communication and media campaigns, have been used in other countries to improve access to ECDE services, parental knowledge and practices, and ultimately childhood development outcomes. Innovative ways include using the surplus primary classrooms, partnering with private entities such as “公办民营”, satellite centers (around the central preschool at the county level), and the purchasing of spaces in private preschools (vouchers). Public-private partnerships in particular are worth exploring to diversify sources of funding, models of delivery, and the creation of markets with new providers for ECDE. Shanghai’s use of vouchers for migrant children attending private primary schools is instructive in this context. Box 6.1: Shanghai’s Voucher System to Improve Education Access and Quality for Migrant Children As one of the largest migrant-receiving cities, Shanghai hosts 410,000 migrant children as of 2010. The large number of migrant children in need of quality education poses considerable challenges to the city. In response, Shanghai has devoted significant attention to resolving the issue through the building of new private migrant schools and the adoption of a voucher system to encourage educational access for migrant children. Shanghai has invested significant resources in building quality private migrant schools to meet the increasing educational demand. Between 2008 and 2010, the government approved 162 private elementary schools specifically designed to educate migrant children. The government allocated 500,000 RMB as an incentive to improve school facilities and resources. In addition, the government invested over 20 million RMB in 2010 to improve the library and gym facilities at these private migrant schools. 113 Most notably, the Shanghai government has created a voucher system to encourage both public and private migrant schools to admit migrant children. It should be noted that Shanghai became the first city in China to eliminate basic education tuition and miscellaneous fees for migrant children, thereby facilitating migrant children enrollment at public schools. In addition, since 2008, the government has allocated a per-capita subsidy to both public and private schools to defray costs associated with enrolling migrant children. For each admitted migrant student, the subsidy increased from 1,000 RMB per student in 2008 to 2,000 RMB per student in 2011. This voucher system not only lifts the financial barrier for migrant families to enroll their children in either type of school, but also demonstrates the government’s continuous commitment to improving access to basic education for migrant children. Shanghai has achieved impressive results in improving educational access for migrant children. By 2012, 73 percent of migrant children were enrolled in public schools in Shanghai. However, as the migrant children population continues to grow, Shanghai will face increasing pressure on educational resources and teacher capacity. It is estimated that by 2015, the student population for junior high school in Shanghai will surpass 610,000— creating an additional need of approximately 3000 class settings. With this projected population growth, Shanghai will need to strengthen its initiatives to balance educational access and quality for migrant children. In addition, the provision of public high school education also merits attention. Migrant parents in Shanghai must meet a set of requirements so that their children can take the entrance exam for regular high school. How to adjust policy so that migrant children can enjoy regular public high school education —such as their urban peers have— remains an open policy and implementation question. Source: Xinhua News 2011/08/30. It is also important to introduce demand-side financing mechanisms. Even if preschool education were to be made free, there would still be families that do not send their children to preschool. Thus, scholarships for the poor children, or even conditional cash transfers, might also be considered. The Philippines conditional cash transfer program, which was recently evaluated by the World Bank, confirms a positive impact on boosting the demand for education. 114 Box 6.2: Philippines Conditional Cash Transfers The Pantawid Pamilyang Program provides cash transfers to poor households, conditional upon investments in child education and health as well as the use of maternal health services. The objective of the program is to promote investments in the education and health of children. Poor households are identified based on a transparent poverty targeting mechanism. Households with an estimated income below the poverty line are classified as poor. From the resulting database of poor households, Pantawid Pamilya identifies and selects eligible households who have children 0-14 years of age and/or a pregnant woman. These households then receive cash grants every two months ranging from PhP 500 to PhP 1,400 per household per month, depending on the number of eligible children. Findings of the study indicate that, overall, the program is meeting its objective of helping to keep poor children in school by increasing enrollment among younger children (3-11 years old) as well as by increasing attendance among 6-17 year olds. The study found higher rates of school enrollment among children 3-11 years of age in the beneficiary households (by 10 percentage points for 3-5 year olds, and by 4.5 percentage points for 6-11 year olds), as compared to poor households who were not participants in the program. Source: Chaudhury, Nazmul, Jed Friedman, and Junko Onishi. 2013. Philippines Conditional Cash Transfer Program Impact Evaluation 2012. World Bank. Expanding service delivery models The formal, but expensive, center-based, full time model of care is not the only viable model. Other models employing flexible service delivery could be adopted in addition to the formal center-based care. Home-based care, parenting education, communication and media campaigns have been used in other countries to improve access to ECDE services, improve parental knowledge and practices, and ultimately improve childhood development outcomes. International experience shows that most children will have cognitive gains even if they are exposed to learning activities for only 15 hours a week. Shortening the time of center-based care can also reduce costs and accommodate more children. Alternatively, parenting education can be used to deliver curriculum-based information for implementation at home that can also be complemented by organized peer interactions. Within Yunnan, there are already good experiences from pilot programs at rural preschool centers, for example by the Child Development Research Foundation (CDRF), using a service delivery model that relies primarily on volunteer teachers and provision of teaching learning materials, flexible hours of operation, as well as renovation or establishment of the physical site. Save the Children also has community based ECD provision in a number of villages through provision of teaching learning materials, use of parent volunteer teachers, and general community support. 115 Improving quality and atmosphere of existing preschools More sophisticated instruments need to be developed for assessing the quality of preschools— beyond the traditional focus on hardware and other inputs, but with more attention to curriculum, activities, teacher-child interactions, and other process indicators. ECERS-R is only one example of such an instrument that is being adopted more widely. Regular assessment of the relatively “weak” dimensions can be used as a management strategy and tool to improve the classroom environment and interactions. Feedback from such evaluations can be used by teachers and preschool directors to improve teaching and learning practices in the classrooms. Results from continuous assessment can also be used to inform the development of preschool teacher training programs. A pilot study in Chongqing demonstrates that well-designed, short term teacher training programs can have a positive impact on observed classroom behavior. Strengthening family and parental education Our study reveals a serious gap in young children’s home environment, especially in rural, poor, and border counties. It is important to note that families play at least an equally important role— if not more important role— in childhood development. Vast research evidence confirms that a lack of learning opportunities and poor quality caregiver-child interactions pose major risks for poor development outcomes. In addition to basic health and nutrition needs, substantial gains in children’s development require improvements in parenting, stimulation and early education. Unfortunately, our survey results indicate a general lack of awareness of the critical importance of early stimulation, playing, and learning plays in overall childhood development by parents. Such a trend needs to be reversed as a matter of urgency via various interventions that promote a positive home environment for young children. Parents can provide more quality stimulation and quality interactions to their children by talking to them, and reading and playing with the child. The parents’ ability for caregiving can be enhanced through home visits, guidance and support from health providers, and group parent training. Increasing awareness of secondhand smoke’s harmful effects Given that ninety percent of households in the survey have members smoke in the presence of children, it is crucial to increase public awareness of the negative health impact of secondhand smoke. At the community level, an extensive information campaign, involving displaying graphic warnings in public space, distributing information brochures to households, and giving presentations in public gatherings, could be an effective strategy. Media could also be a powerful communication channel. Well-designed TV and radio commercials that emphasize secondhand smoke’s negative impact might reach a wider population. At the household level, families with pregnant women or children merit special attention. For example, health facilities that offer prenatal care and or pediatric could give out health education packages with information on secondhand smoke. Training of nurses and doctors to regularly ask pregnant women and parents whether they smoke in the 116 household and inform them of the need to stop smoking when the children are around is another important measure. Providing counseling to parents against smoking could be included as part of the diagnostic and treatment protocols for asthma, ear infections and respiratory tract infections in infants and children. For households in which family members other than the parents are the primary caregivers, it is important to design targeted measures so that they ensure the children are in a smoke-free environment. If resources permit, the local government could also consider interventions that reduce smoking in general, such as offering assistance to adults who wish to quit smoking and enhancing education effort to prevent children and youth from starting to smoke. Developing a preschool teaching force Adequate provision of qualified preschool teachers presents the biggest challenge for Yunnan. Preschool teachers as a group have become marginalized from the rest of the teaching force. There is a need to improve preschool teachers’ pedagogical competency as well as awareness and sensitivity to children’s development needs. The shortage of qualified preschool teachers is particularly acute in rural ，poor， and minority counties. In particular, standards for preschool teachers need to be reviewed to emphasize more teachers’ professional and pedagogical competencies with regard to teaching young children. There is a need to enforce the requirement of teaching certificates at hiring, especially for rural and private preschools. In-service training and support for preschool teachers should be a high priority in the next decade. Indeed, more funds should be directed to the development and training of both new and current teachers in order to meet the growing demand. The focus of training should be devoted more to preschool pedagogy as per the Learning and Development Guidelines. At the same time, there is a need to devise more innovative policy measures to attract more capable high school graduates into pre-service programs in both senior vocational and tertiary preschool education programs. In particular, efforts should be made to recruit more male preschool teachers, and provide more active admission management, as well as career counseling. Tuition waivers, more access to scholarships, and targeted job placement should be made available. Colleges can also consider establishing preschool education preparation centers and encouraging job re-placement for former primary school teachers and other unemployed graduates who are interested in preschool education. There is a need to review the preschool teacher career ladder and associated salaries. Gradual improvements should be made so that eventually there will not be any difference in career opportunities and salaries between preschool and primary school teachers. Particular emphasis should be paid to improving the working conditions and pay for rural preschool teachers. Special Incentives should also be developed to retain rural preschool teachers, including more training programs, higher compensation, better accommodations, and improved support networks. Encouraging exchanges of teachers between rural, county and urban areas would also be beneficial, as would incorporate exchange services into 117 routine teacher evaluations. It is also important to establish formal recruiting, training, contracting and evaluating mechanisms for private and rural preschools teachers, and incorporate them under the management of public educational authorities. There may be merit to setting minimum teacher salaries for rural as well as for private preschools. Moving toward integrated ECD provision for all including 0-3 Year Olds With this research design, the team had hoped that the study would be able to cover early childhood development as an integrated service. However, it soon became clear that early childhood development as a concept has yet to emerge in Yunnan or in China as a whole. The provision of services for young children continues to be organized along sectoral boundaries of education, health, and child protection. It was impossible to collect data or to have any meaningful conversation about “integrated services.” Among the various sectors relevant to young children, education is the most organized and mature department, so the team narrowed its focus to only early childhood education. By contrast, increasingly, global trends move toward more integrated provision of services for young children— bringing together prenatal, health, early stimulation/education, nutrition and child protection services— usually under an umbrella childhood development agency. Educators, psychologists, developmental specialists, and economists are converging on the view that human development is a lifecycle development process that starts before the child is born. In the long term, Yunnan could start developing the ECD concept and related policies and institutions, covering prenatal services, to services for 0-3 year olds, 3-6 year olds, and even aligning with primary school education. Although this may still be a long-term goal, there are some areas that deserve more policy attention now by the government, for example, the care and development of the 0-3 year olds. This is an area that has been largely neglected. As of yet, there is no public service provision for this age group. Parents use their own means either within the family or in the private market to care for their 0-3 year olds. The private market is often unregulated and poses a potential development threat to young children. At the same time, labor force participation, including female labor force participation, is very high in China. Most of the young children are now left in the care of grandparents who have relatively less education than their children. There is a great urgency to develop appropriate policy interventions for the 0-3 year olds, and in particular for those who are at risk. For the 0-3 group, emphasis should be placed on improving prenatal care to pregnant mothers, improving the nutritional and health status of young children through the provision of nutritional packages, enhancing parental and caregiver education, and promoting early social and cognitive stimulation to young children. International experience shows that home-based and/or community-based approaches that facilitate interaction and training of mothers and caregivers by persons knowledgeable about nutrition, health and stimulation can be highly cost effective. 118 119 Appendix: A. Kindergarten Gross Enrollment Rate (est.) of three Years before Primary School, 2004-2010, by Region 70 62 60 54 Gross Enrollment Rate (%) 48 50 43 40 48 40 40 36 36 33 32 32 30 28 30 30 32 37 28 23 24 20 20 21 26 22 19 10 15 17 14 0 2004 2005 2006 2007 2008 2009 2010 National Average Eastern Provinces Western Provinces Yunnan b a Source: Sources: Author’s estimate based on NBS 2000 – 2011, MOE 2010 Note: MOE= Ministry of Education; NBS=National Bureau of Statistics This estimated Gross Enrollment Rate (GER) accounts only for three-year kindergarten enrollment (not counting one-year pre-primary classes) three years before primary school, as compared to counting both kinds of enrollment in the report. Because of the large share of one-year pre-primary classes as a mean of pre-school enrollment in western provinces including Yunnan, the enrollment rate contrasts even more between developed eastern provinces and relatively underdeveloped western provinces, even though the latter is experiencing some catching up. The researchers were not able to collect complete regional data beyond 2010 due to data deficiencies. But the estimate resembles publicized statistics on national level well in 2010. Before 2010, there has been a mixed use of both terms on national and provincial level in terms of GER for kindergartens enrollment alone and GER for both kindergartens and one-year pre-primary class enrollment, thus creating some confusions. 120 B. Regression Results of Multivariate (Model 1-8) and Logistic (Model 9) Analysis of the Household Survey on Childhood Development in Yunnan, 2013 Model (1) (2) (3) (4) Variable Weight Weight Height Height 1.13 1.03 -0.25 -0.69 Weight at birth (0.28) *** (0.27)*** (1.02) (1.03) Age when recent weight (or 0.12 0.12 0.54 0.55 height) is measured (0.01)*** (0.01)*** (0.05)*** (0.05)*** 0.77 0.63 1.87 1.06 Boy=1 (0.30)** (0.31)* (1.12) (1.16) 0.34 0.09 -0.54 0.40 Breastfed=1 (0.47) (0.12) (1.78) (1.88) 0.10 0.35 -0.78 -0.90 Family size (0.11) (0.49) (0.39)* (0.40)* -0.75 -0.71 -4.83 -4.42 Poverty county=1 (0.32)* (0.31)* (1.41)*** (1.32)*** -0.48 0.08 -2.54 -3.42 Border county=1 (0.45) (0.52) (2.44) (2.63) -0.82 -0.65 -0.86 -1.41 Minority county=1 (0.33)** (0.34) (1.35) (1.46) -0.65 -0.60 -1.78 -0.36 Left-behind children=1 (0.48) (0.51) (1.96) (2.03) Household income=more than 0.02 -0.05 -1.86 0.35 10,000, less than 20,000RMB) (0.40) (0.43) (2.05) (2.22) Household income=more than 0.56 0.60 -2.20 0.02 20,000, less than 50,000RMB) (0.42) (0.44) (2.13) (2.17) Household income=more than 0.70 0.44 0.31 3.19 50,000, less than 100,000RMB) (0.53) (0.62) (2.63) (2.65) Household income=more than 0.98 1.40 6.34 12.57 100,000RMB (1.31) (1.36) (3.03)* (3.17)*** Mother’s education= Primary -0.32 -0.23 3.30 3.22 School (0.57) (0.60) (3.12) (3.18) Mother’s education= Junior 0.41 0.17 0.58 0.68 middle school (0.58) (0.57) (2.91) (3.12) Mother’s education= Senior 0.26 0.37 1.59 4.08 high school/Junior vocational (0.85) (0.82) (3.37) (3.14) school Mother’s education= Senior 2.94 2.80 3.61 6.33 vocational school (1.70) (1.80) (3.54) (3.99) Mother’s education= Tertiary 3.03 2.76 6.33 9.23 and above (0.78)*** (0.89)** (4.21) (4.88) 121 Caregiver’s education= Primary -0.35 -0.71 -3.52 -5.23 School (0.53) (0.57) (2.82) (2.91) Caregiver’s education= Junior 0.01 0.16 0.75 0.03 middle school (0.56) (0.58) (2.55) (2.82) Caregiver’s education= senior -0.69 -1.08 -3.26 -6.08 high school/Junior vocational (0.86) (0.85) (3.21) (3.00) school Caregiver’s education= senior -2.46 -3.00 -1.66 -7.56 vocational school (1.42) (1.42) (3.34) (4.37)* Caregiver’s education= Tertiary -2.38 -2.28 1.64 -3.13 and above (1.08)* (1.21)* (6.32) (6.60) 0.27 0.32 -1.02 -0.87 Single child=1 (0.33) (0.34) (1.29) (1.50) 0.05 -0.13 1.76 1.66 Han ethnicity=1 (0.28) (0.32) (1.17) (1.37) Enrolled in kindergarten or 0.35 0.91 one-year pre-primary classes (0.34) (1.60) 0.52 4.14 Attended parent-child classes (0.77) (3.25) 1.04 5.50 Health check=Occasional (0.52)* (3.53) 0.49 4.40 Health check=once a year (0.46) (3.50) Immunization = both free and 0.35 -1.61 paid (0.39) (1.80) Read to children frequency= 1-2 -0.57 0.91 times per week (0.34) (1.61) Read to children frequency= >2 0.12 2.00 times per week (0.39) (1.78) Play with children= <1 hour per 0.39 -2.63 day (0.34) (1.41) Play with children= 1-2 hours -0.27 -2.17 per day (0.51) (2.12) Play with children= >2 hours per 0.13 -3.77 day (0.62) (4.37) -0.25 1.12 Milk intake=Everyday (0.71) (1.90) -0.88 1.02 Milk intake=once every 2-3 days (0.67) (2.03) 0.24 8.53 Milk intake=once every week (0.84) (2.60)*** 122 -0.10 1.91 Milk intake=Seldom (0.61) (2.22) 0.63 -1.32 Meat intake=every day (0.40) (3.20) Meat intake=once every 2-3 -0.02 -4.26 days (0.44) (3.24) 0.33 -2.17 Meat intake= Occasionally (0.47) (3.30) 0.48 3.00 Vegetable intake=every day (0.67) (2.97) Vegetable intake=once every 0.45 1.38 2-3 days (0.80) (3.47) 0.83 2.39 Vegetable intake=Occasionally (0.87) (3.76) -0.49 -2.56 Fruit intake=every day (0.84) (3.55) -0.79 -0.33 Fruit intake=once every 2-3 days (0.85) (3.83) -0.48 -0.84 Fruit intake=Occasionally (0.85) (3.64) -0.03 -2.03 Calcium supplement=1 (0.39) (1.45) 0.98 6.30 Discipline=sometimes (0.41)* (1.74)*** 1.62 8.65 Discipline=always (0.63)** (3.01)** 5.89 4.52 77.88 70.09 Constant (1.39)*** (1.79)* (5.84)*** (9.20)*** Number of Observations 554 531 313 301 R–squared 0.28 0.34 0.45 0.53 Adjusted R–squared 0.25 0.27 0.40 0.44 Source: Authors. Note: Standard errors in parentheses. * significant at 5 percent; ** significant at 1 percent; *** significant at 0.1 percent. 123 Model (5) (6) (7) (8) Social Social Cognitive Cognitive Variable development development development development index index index index 0.003 0.001 0.026 0.022 Age in months (0.001)** (0.001) (0.001)*** (0.001)*** 0.012 0.010 -0.009 -0.011 Boy=1 (0.019) (0.019) (0.023) (0.023) 0.000 -0.001 0.007 0.009 Family size (0.006) (0.005) (0.008) (0.007) -0.076 -0.067 -0.063 -0.048 Poverty county=1 (0.020)*** (0.021)** (0.025)* (0.026) -0.077 -0.084 -0.025 -0.012 Border county=1 (0.033)* (0.036)* (0.040)* (0.044) -0.003 0.017 0.068 0.102 Minority county=1 (0.022) (0.023) (0.028)* (0.028)***37 -0.018 -0.020 -0.022 -0.060 Left-behind children=1 (0.027) (0.028) (0.033) (0.033) Household income=more than 0.054 0.044 0.089 0.032 10,000, less than 20,000RMB) (0.024)* (0.025) (0.030)** (0.029) Household income=more than 0.086 0.071 0.099 0.024 20,000, less than 50,000RMB) (0.025)*** (0.025)** (0.032)** (0.032) Household income=more than 0.132 0.099 0.268 0.123 50,000, less than (0.043)** (0.045)* (0.051)*** (0.050)* 100,000RMB) Household income=more than 0.113 0.088 0.200 0.086 100,000RMB (0.069) (0.074) (0.078)** (0.089) Mother’s education= Primary 0.000 -0.014 0.016 -0.001 School (0.033) (0.034) (0.042) (0.040) Mother’s education= Junior -0.005 -0.010 0.141 0.095 middle school (0.036) (0.035) (0.042)*** (0.040)* Mother’s education= Senior -0.002 -0.023 0.154 0.082 high school/Junior vocational (0.062) (0.062) (0.060)* (0.063) school Mother’s education= Senior -0.101 -0.072 0.188 0.175 vocational school (0.138) (0.137) (0.082)* (0.075)* 37 The significance for Minority County having effect on children’s cognitive development is somewhat misleading. The researcher later found this could because the survey design had categorized a seemingly non-minority county (with less than 20% minority population) into the sampling pool of minority counties (usually above 50% of population), and that county was selected. No changes were made as to respect the original design and a note was added. 124 Mother’s education= Tertiary 0.048 0.025 0.271 0.080 and above (0.045) (0.055) (0.057)*** (0.068) Caregiver’s education= -0.005 -0.029 -0.058 -0.074 Primary School (0.034) (0.035) (0.042) (0.039) Caregiver’s education= Junior 0.056 0.010 -0.033 -0.057 middle school (0.035) (0.035) (0.041) (0.037) Caregiver’s education= senior 0.062 0.013 -0.096 -0.138 high school/Junior vocational (0.057) (0.055) (0.059) (0.057)* school Caregiver’s education= senior 0.124 0.049 -0.160 -0.264 vocational school (0.134) (0.126) (0.100) (0.090)* Caregiver’s education= 0.027 -0.014 0.007 0.041 Tertiary and above (0.080) (0.083) (0.113) (0.143) 0.041 0.032 0.059 0.019 Single child=1 (0.023) (0.024) (0.028)* (0.027) 0.013 -0.007 0.102 0.070 Han ethnicity=1 (0.024) (0.025) (0.027)*** (0.026)* Primary language -0.068 -0.012 0.012 0.101 spoken=Mandarin dialect (0.047) (0.047) (0.070) (0.066) Primary language -0.113 -0.069 -0.026 0.074 spoken=minority dialect (0.054)* (0.055) (0.074) (0.073) Enrolled in kindergarten or 0.098 0.226 one-year pre-primary classes (0.023)*** (0.029)*** 0.005 0.084 Attended parent-child classes (0.038) (0.057) -0.010 0.048 Health check=Occasional (0.028) (0.035) -0.003 -0.004 Health check=once a year (0.028) (0.036) Number of books at 0.010 0.005 home=1-10 (0.026) (0.031) Number of books at 0.020 -0.006 home=11-30 (0.035) (0.047) Number of books at home= -0.017 0.085 more than 30 (0.039) (0.052) Time watching TV= less than 1 -0.005 0.084 hour per day (0.049) (0.053) Time watching TV= 1-2 hours 0.012 0.128 per day (0.048) (0.055)* 125 Time watching TV= more than -0.008 0.142 2 hours per day (0.050) (0.057)* Immunization = both free and -0.041 0.070 paid (0.021) (0.027)** Read to children frequency= 0.084 0.065 1-2 times per week (0.024)*** (0.030)* Read to children 0.089 0.097 frequency= >2 times per week (0.026)*** (0.031)** Play with children= <1 hour 0.015 -0.015 per day (0.021) (0.025) Play with children= 1-2 hours 0.006 0.000 per day (0.038) (0.058) Play with children= >2 hours 0.075 -0.018 per day (0.061) (0.064) -0.010 0.020 Punish=sometimes (0.024) (0.028) -0.053 -0.041 Punish=always (0.050) (0.057) 1.297 1.289 -0.783 -0.907 Constant (0.084)*** (0.096)*** (0.104)*** (0.111)*** Number of Observations 983 938 990 945 R–squared 0.13 0.18 0.45 0.53 Adjusted R–squared 0.11 0.14 0.43 0.51 Source: Authors. Note: Standard errors in parentheses. * significant at 5 percent; ** significant at 1 percent; *** significant at 0.1 percent. Model (9) Variables Enrollment status (0 or 1) 0.946 Boy=1 (0.162) 1.094 Age in months (0.009)*** 2.366 Household income=more than 10,000, less than 20,000RmB) (0.530)*** 2.739 Household income=more than 20,000, less than 50,000RmB) (0.614)*** 12.402 Household income=more than 50,000, less than 100,000RmB) (6.470)*** 126 4.096 Household income=more than 100,000 (3.424) 1.920 Left-behind children=1 (0.526)** 0.971 Mother’s education= Primary School (0.280) 1.686 Mother’s education= Junior middle school (0.531) Mother’s education= Senior high school/Junior vocational 2.539 school (1.185)* 2.547 Mother’s education= Senior vocational school (2.618) 4.3e7 Mother’s education= Tertiary and above (5.7e7)* 1.366 Caregiver’s education= Primary School (0.377) 1.085 Caregiver’s education= Junior middle school (0.328) Caregiver’s education= senior high school/Junior vocational 1.530 school (0.652) 1.701 Caregiver’s education= senior vocational school (1.779) 0.000 Caregiver’s education= Tertiary and above (.) 4.620 Kindergarten closer to primary school=1 (1.149)*** 1.987 Single child=1 (0.424)*** 0.366 Minority county=1 (0.074)*** 0.867 Poverty county=1 (0.153) 1.138 Border county=1 (0.347) 0.117 Primary language spoken=mandarin dialect (0.098)** 0.106 Primary language spoken=minority dialect (0.093)* 2.126 Han ethnicity=1 (0.430)*** Number of Observations 989 Pseudo R-Squared 0.29 127 Source: Authors. Note: Displays Odds Ratio, standard errors in parentheses. * significant at 5 percent; ** significant at 1 percent; *** significant at 0.1 percent. 128 References Aboud, F.E. 2006. “Evaluation of an early childhood preschool program in rural Bangladesh.” Early Childhood Research Quarterly 21[1]: 46-60. Alderman, A. 2011. No Small Matter: The Impact of Poverty, Shocks, and Human Capital Investments in Early Childhood Development. Washington, DC: World Bank. Anderson, D. R., Huston, A. C., Schmitt, K. L., Linebarger, D. L., & Wright, J. C. (2001). Early childhood television viewing and adolescent behavior: The recontact study. Monographs of the Society for Research in Child Development, 66(1), 1-147. Carneiro, P. M., and J. J. Heckman. 2003. Human Capital Policy. Institute for the Study of Labor (IZA) Discussion Paper No. 821. [Accessed March 24, 2007, from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=434544] Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Health Effects of Secondhand Smoke. http://www.cdc.gov/tobacco/data_statistics/fact_sheets/secondhand_smoke/health_effects / Center on Media and Child Health, Children’s Hospital Boston. (2005). The Effects of Electronic Media on Children Ages Zero to Six: A History of Research. Prepared for the Kaiser Family Foundation Chaudhury, Nazmul, Jed Friedman, and Junko Onishi. 2013. Philippines Conditional Cash Transfer Program Impact Evaluation 2012. World Bank. China Education and Research Network, Xinhua News 2011/08/30. http://www.nmg.xinhuanet.com/xwzx/2011-08/30/content_23581726.htm Chagas, F.C. (2010). An overview of the study: Early Childhood Education in Brazil – A qualitative and quantitative assessment. Inter-American Development Bank. Fali, H., and Myoung-Jae, L. (2010). Dynamic treatment effect analysis of TV effects on child cognitive development. Journal Of Applied Econometrics, 25(3), 392-419. doi:10.1002/jae.1165 Garcia, M., Pence, A., & Evans J.L. (2008). Africa’s Future, Africa’s Challenge: Early Childhood Care and Development in Sub-Saharan Africa. World Bank Madrasas Early Childhood Development Program: Making a Difference, pp. 389. Goelman, H. (2006). Towards a predictive model of quality in Canadian child care centers. Early Childhood Research Quarterly. 21: 280-295. Kunming Information Center. (2011). Kunming Government School Fees Standards for the New School Year. http://www.km.gov.cn/structure/sylm/kmxwxx_169866_1.htm 129 Hart, B., and T. R. Risley. 1995. Meaningful Differences in the Everyday Experiences of Young American Children. Baltimore, Md.: Paul H. Brookes Publishing Company, Inc. Hu, B. (2009). Exploring the Quality of Early Childhood Education in China: Implications for Early Childhood Teacher Education. Journal of Early Childhood Teacher Education. 30 (3): 1– 28. Hu, B., and Zhu, Z. (2009). The U.S. ECERS-R and Its Preliminary Applications in China (美国 《幼儿学习环境评量表》及其在中国的初步应用). Early Childhood Education (《幼儿教育 ∙教育科学》), 11:47-51 Huang, X. & Song, Y. (2012). Developing New Evaluation Tools: A performance evaluation of kindergarten procedure quality (开发新的评价工具： 对幼儿园过程性质量进行表现性评 价). Huesmann, L.R., Moise-Titus, J., Podolski, C.L., and Eron, L.D. (1993). Longitudinal Relations Between Children's Exposure to TV Violence and Their Aggressive and Violent Behavior in Young Adulthood: 1977 – 1992. Developmental Psychology, Vol. 39, No. 2. Jiang, Y. Yan, J. and Xu, L. (2012). Analysis of Preschool Teachers Policy in the International Context (国际学前教师教育政策研究). Shanghai: East Normal University. Kalb, G. & van Ours, J. (2013). Reading to Young Children: A Head-Start in Life? IZA Discussion Paper No. 7416. Miyahara, J. (2013). Regional ECD Profiles. World Bank Linebarger, D. L., and Walker, D. (2005). Infants' and Toddlers' Television Viewing and Language Outcomes. American Behavioral Scientist, 48(5), 624-645. doi:10.1177/0002764204271505 Leung GM, Ho L., Lam T. (2004). Secondhand Smoke Exposure, Smoking Hygiene, and Hospitalization in the First 18 Months of Life. Arch Pediatr Adolesc Med. 158(7), 687-693. doi:10.1001/archpedi.158.7.687. Ministry of Education (MOEa). Educational Statistical Year Book of China, 1985, 1990, 1995, 2000, 2004-2010. Ministry of Education (MOEb). China Education Year Book, 2004-2010. Ministry of Education (MOEc). Educational Finance Statistical Year Book, 2004-2011. Miyaha, J. (2013). Regional ECD Profiles Presentation. ESDP Global Session I, Session 4: Trends in the Region. The World Bank Mwaura, Peter; Mohamed, Bishara. 2008. Madrasa Early Childhood Development Program: Making a difference. Africa’s Future, Africa’s Challenge: Early Childhood Care and 130 Development in Sub-Saharan Africa. Washington, DC: World Bank. National Bureau of Statistics of China (NBSa). China Statistical Abstract, 2007-2013. National Bureau of Statistics of China (NBSb). China Statistical Year Book for Regional Economy, 2005-2011. Naudeau, Sophie; Kataoka, Naoko; Valerio, Alexandria; Neuman, Michelle J.; Elder, Leslie Kennedy. 2011. Investing in young children : an early childhood development guide for policy dialogue and project preparation. Directions in development ; human development; Africa regional educational publications. Washington, DC: World Bank. Office on Smoking and Health (2006). The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. Atlanta (GA): Centers for Disease Control and Prevention (US); Reproductive and Developmental Effects from Exposure to Secondhand Smoke. Available from: http://www.ncbi.nlm.nih.gov/books/NBK44317/ Sahlberg, P. (2010). The Secret to Finland’s Success: Educating Teachers. Standard Center for Opportunity Policy in Education. Sénéchal, M. (2006). The Effect of Family Literacy Interventions on Children’s Acquisition of Reading From Kindergarten to Grade 3. National Institute for Literacy, RMC Research Corporation. Schweinhart, L. J., Montie, J., Xiang, Z., Barnett, W. S., Belfield, C. R., & Nores, M. (2005). Lifetime effects: The HighScope Perry Preschool study through age 40. (Monographs of the HighScope Educational Research Foundation, 14). Ypsilanti, MI: HighScope Press. Shanghai Academy of Educational Sciences Intelligence Development Agency. (2013). Analysis of Yunnan Early Childhood Development Status (云南学前教育发展情况分析). Statistical Bureau of Yunnan (SBY). Yunnan Statistical Year Book, 2006-2009. Wu, Kin Bing. 2011. Early childhood development and education in China: breaking the cycle of poverty and improving future competitiveness. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/2011/01/13829205/early-childhood-developm ent-education-china-breaking-cycle-poverty-improving-future-competitiveness Young, M.E. 2002. Investing in Our Children’s Future. From Early Child Development to Human Development. Washington, DC: World Bank. Yunnan Normal University Education Management Institute. (2013). Evaluation Report on the Quality of Kindergarten Educational Environment (昆明市幼儿园教育环境质量评估子项目 研究报告). World Bank Yunnan Early Childhood Development Project. Yunnan Provincial Government. (2011). Detailed Working Plan for Yunnan Preschool 131 Education Reform ( 云 南 省 学 前 教 育 改 革 发 展 重 点 工 作 任 务 分 解 方 案 ) http://xqjy.ynjy.cn/index.php?case=archive&act=show&aid=184 Yunnan Provincial Government. (2013). On Implementing Provincial Preschool Education (云 南省教育厅关于认真做好学前教育有关工作的通知) http://xqjy.ynjy.cn/index.php?case=archive&act=show&aid=188 Zhang Y. (2009). On the Policy of Abolish the Pre-school Classes (对取消学前班政策的思考) Studies in Preschool Education (学前教育研究), 11, Serial No.179. Zhou X. & Zhao Z. (2013). Study of Preschool Teachers in Yunnan Province: Evidence from preschool survey and review of existing policies (云南幼儿教师研究——对政策文献的回顾 和实地调研结果的分析). Report for World Bank Project in Yunnan, China 132