Public Expenditure Tracking Survey in Education 2004 - Cambodia

Abstract
---------------------------

This study was designed to analyze the primary education Priority Action Program (PAP 2.1) in Cambodia.

Weaknesses in Cambodia's public expenditure management system - including the difficulty in channeling funds down to service providers - have resulted in costly inefficiencies in improving social welfare outcomes. In response to these problems, in 2000 the Royal Government of Cambodia started the Priority Action Program (PAP).

PAP education was launched in 10 provinces in 2000 and expanded to cover all provinces in 2001. The primary education Priority Action Program (PAP 2.1) provides for schools' operational budgets and accounts for over one quarter of the entire PAP budget in education. PAP 2.1 funds are to be used to support start-of-the-year operational costs as well as day-to-day school operations. PAP 2.1 funds are intended to cover small expenditures on basic teaching materials, sports, school environment improvements and other small spending items primary schools usually incur. The fixed grant component is expected to be disbursed first in any given budget year so as to allow schools to purchase teaching materials and carry out the school opening ceremony at the beginning of each school year (October).

The main objective of this study is to analyze the impact of PAP 2.1 in terms of providing resources to Cambodian schools in a timely and predictable manner, making sure those resources are used as intended. In addition to issues of efficiency (resource flow and use), Cambodia PETS in Primary Education examines equity in the allocation of PAP 2.1. resources and the impact on student outcomes.

Moreover, this study analyzed the flow and use of Chapter 11 funds in the regular budget system, and compared them with PAP 2.1 funds. Chapter 11 funds are intended for general operational and maintenance expenditure.

The Cambodia Public Expenditure Tracking Survey in Primary Education is a complex survey involving the Provincial Treasuries (PT), Provincial Education Offices (PEOs), District Education Offices (DEOs), school directors, parents and students. The survey covered a random sample of 200 primary schools in 21 districts and 7 provinces. Also, randomly selected 20 Grade 4 students and 20 Grade 6 students were tested in numeracy and literacy skills.

Geographic coverage
---------------------------

7 provinces (Banteay Meanchey, Kampong Cham, Kampong Chhnang, Kampot, Kratie, Prey Veng and Sihanoukville).

Analysis unit
---------------------------

- Students
- Teachers
- Parents
- School Principals
- Primary schools
- Provincial Treasuries
- Provincial Education Offices
- District Education Offices.

Kind of data
---------------------------

Sample survey data [ssd]

Sampling procedure
---------------------------

Initial sample of 220 primary schools was randomly drawn from 12 provinces and 34 districts. The final sample included 200 schools in 7 provinces since some schools could not be accessed and a replacement could not be found.

Stratified random sample based on the 2002-2003 school census was used. Each province was weighted according to size (number of schools). Five early (receiving funding from 2000 and onwards) and seven late (receiving funding from 2001 and onwards) provinces were randomly chosen from the population of 24 provinces. Each of the 12 selected provinces (Banteay Meanchey, Battambang, Kampong Cham, Kampong Chhnang, Kampong Speu, Kampong Thom, Kampot, Phnom Penh, Ratanak Kiri, Siemreap, Svay Rieng, and Takeo) was then allotted a number of schools based on the proportion of schools in the province to the total number of schools in the selected sample of 12 provinces. To make the survey effort feasible, it was decided that 2 or 3 districts would be picked from each province (the final number depending on the total number of districts in the province). Similar to the draw of provinces, each district was weighted according to size (number of schools). Thereafter, 2 or 3 districts were randomly chosen from the population of districts within each province, yielding a total of 34 districts to be included in the final sample.

The second stage of the sampling was based on the Cambodian 2003-2004 household surveys (HSES). The HSES sampling frame consisted of 900 villages and 15,000 households. The sampling design involved stratification of the country into five geographical regions (Phnom Penh, Plain, Tonle Sap, Coastal, and Plateau and Mountain), dividing up each region into separate urban and rural strata. From each stratum, 4 independent sub-samples of villages were drawn, with the sample being allocated over the strata proportionately to the total number of households in the strata.

At the school level, questionnaires were administered to each school director. One teacher per grade was randomly selected from each school to interview. Three School Support Committee (SSC) members from each school and 6 parents per school were also randomly chosen to participate in the study.

Mode of data collection
---------------------------

Face-to-face [f2f]

Research instrument
---------------------------

The following survey instruments are available:

- Student Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- School Director Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Parent Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Teacher Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- School Support Committee (SSC) Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- District Education Office Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Provincial Education Office Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Provincial Treasury Questionnaire, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Literacy Test, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Numeracy Test, Grade 4, Cambodia Public Expenditure Tracking Survey in Primary Education 2004;
- Numeracy Test, Grade 6, Cambodia Public Expenditure Tracking Survey in Primary Education 2004.

The core of the questionnaires for provincial treasuries, provincial education offices, district education offices and school directors focuses on resource flows (i.e., resources received from an upper tier of government and resources transferred to a lower tier of government), which allowed for some triangulation or cross checking of the data. The questionnaire for school directors also contained a variety of information on schools, teachers and students.

The teacher, School Support Committee (SSC) and parental questionnaires were designed to collect information on the degree of knowledge of and involvement in school matters (particularly PAP flows and use).

In addition, 20 Grade 4 students and 20 Grade 6 students were tested in numeracy and literacy skills; background socio-economic information was also collected from them.

Cleaning operations
---------------------------

Detailed information about data editing procedures is available in "Data Cleaning Guide for PETS/QSDS Surveys" in external resources.

STATA cleaning do-files and data quality reports on all ten datasets can also be found in external resources.

摘要
---------------------------
本研究旨在分析柬埔寨基础教育的优先行动项目（PAP 2.1）。柬埔寨公共支出管理系统中的缺陷，包括资金难以分配至服务提供者，导致了改善社会福利成果方面的巨大低效。为应对这些问题，柬埔寨王国政府于2000年启动了优先行动项目（PAP）。

PAP教育项目于2000年在10个省份启动，并于2001年扩展至所有省份。基础教育的优先行动项目（PAP 2.1）为学校提供运营预算，占整个教育PAP预算的四分之一以上。PAP 2.1的资金旨在支持学年初的运营成本以及日常学校运营。PAP 2.1的资金用于覆盖基本教学材料、体育、学校环境改善和其他小学通常发生的少量支出。固定拨款部分预计将在任何给定预算年度的第一时间发放，以便学校能够在每个学年初（10月）购买教学材料和举行开学典礼。

本研究的主要目标是分析PAP 2.1在为柬埔寨学校提供及时且可预测的资源方面的成效，确保这些资源得到按预期使用。除了效率问题（资源流动和使用）之外，柬埔寨基础教育公共支出跟踪调查还考察了PAP 2.1资源的分配公平性及其对学生成果的影响。

此外，本研究分析了常规预算系统中第11章资金的流动和使用情况，并将其与PAP 2.1资金进行了比较。第11章资金旨在用于一般运营和维护支出。

柬埔寨基础教育公共支出跟踪调查是一项复杂的调查，涉及省级财政部门（PT）、省级教育局（PEO）、区教育局（DEO）、学校校长、家长和学生。调查覆盖了21个区和7个省份的200所小学的随机样本。同时，随机选取了20名四年级学生和20名六年级学生在算术和读写能力方面进行了测试。

地理覆盖范围
---------------------------
7个省份（班迭棉吉、磅湛、磅湛省、磅同、克雷蒂、波东奔和西哈努克省）。

分析单元
---------------------------
- 学生
- 教师
- 家长
- 学校校长
- 小学
- 省级财政部门
- 省级教育局
- 区教育局。

数据类型
---------------------------
样本调查数据 [ssd]

抽样程序
---------------------------
从12个省份和34个区中随机抽取了220所小学作为初始样本。由于一些学校无法访问且无法找到替代学校，最终样本包括7个省份的200所学校。

基于2002-2003学年人口普查的分层随机抽样被用于最终样本。每个省份根据学校数量进行加权。从24个省份的人口中随机选择了5个早期（自2000年起获得资金）和7个后期（自2001年起获得资金）的省份。然后，根据每个省份在学校总数中的比例，为每个选定的12个省份分配了一定数量的学校。为了使调查工作可行，决定从每个省份中选择2个或3个区（具体数量取决于该省份的总区数）。与省份的选择类似，每个区根据学校数量进行加权。此后，从每个省份内的区人口中随机选择了2个或3个区，最终形成了34个区纳入最终样本。

抽样的第二阶段基于柬埔寨2003-2004年家庭调查（HSES）。HSES抽样框架包括900个村庄和15,000户家庭。抽样设计涉及将国家划分为五个地理区域（金边、平原、洞里萨湖、沿海和高原山区），并将每个区域划分为独立的城镇和农村层。从每个层中抽取了4个独立的村庄子样本，样本在层之间的分配比例与层中的总家庭数成比例。

在学校层面，对每位学校校长进行了问卷调查。从每所学校随机选择一名教师进行访谈。从每所学校随机选择了3名学校支持委员会（SSC）成员和6名家长参与研究。

数据收集方式
---------------------------
面对面 [f2f]

研究工具
---------------------------
以下调查工具可用：
- 学生问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 学校校长问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 家长问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 教师问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 学校支持委员会（SSC）问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 区教育局问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 省级教育局问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 省级财政部门问卷，柬埔寨基础教育公共支出跟踪调查2004年；
- 读写能力测试，柬埔寨基础教育公共支出跟踪调查2004年；
- 四年级算术测试，柬埔寨基础教育公共支出跟踪调查2004年；
- 六年级算术测试，柬埔寨基础教育公共支出跟踪调查2004年。

省级财政部门、省级教育局、区教育局和学校校长的问卷核心关注资源流动（即从上级政府接收资源和向下级政府转移资源），这允许对数据进行一些三角测量或交叉检查。学校校长的问卷还包含有关学校、教师和学生的各种信息。

教师、学校支持委员会（SSC）和家长问卷旨在收集有关对学校事务（尤其是PAP流动和使用）的了解程度和参与度的信息。

此外，对20名四年级学生和20名六年级学生进行了算术和读写能力测试；还从他们那里收集了背景社会经济信息。

数据清理操作
---------------------------
有关数据编辑程序详细信息的详细信息可在“PETS/QSDS调查数据清理指南”的外部资源中找到。

STATA清理do文件和所有十个数据集的数据质量报告也可在外部资源中找到。