Public Expenditure Tracking Survey in Education 2006, First Round - Madagascar

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

Madagascar had low school enrollment rates: only 60% of the urban children and 12% of the rural children completed primary school (World Bank, 2002). To improve the enrollment and completion rates as well as the quality of education, Madagascar government had substantially increased investments in the education sector. It committed itself to the Education For All (EFA) initiative and started to fully subsidize the tuition fees through the so-called "caisse ecole," and to provide school kits for all students in public primary schools. The Government also raised the districts' budgets for school material and started distributing free textbooks to schools.

This study investigated the different resource flows in the financing of the public primary education sector in Madagascar.

The survey was conducted in two rounds. The first round was carried out in October-November 2006 and the second round in April-May 2007. The study was implemented using stratified random sampling. Data from more than 200 schools in 28 districts was analyzed.

Public Expenditure Tracking Survey among Madagascar health care facilities and workers was conducted at the same time with PETS in Education.

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

Provinces: Antananarivo, Fianarantsoa, Toamasina, Mahajanga, Toliara and Antsiranana.

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

- Cisco (Circonscription Scolaire/District Education Facility);
- Schools;
- School Principals;
- Teachers.

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

Sample survey data [ssd]

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

The study was conducted using stratified random sampling.

The stratified sample was set up in such a way to be representative at the national level. Madagascar has 22 regions and 111 districts, and at least one district was visited in each region. Two districts were selected in the six largest regions. Hence, 28 districts were visited in total. The selected districts were obtained through random selection, giving greater (less) weight to districts with more (less) public primary schools within the district. In each district, three communes were randomly selected, giving greater weight to the communes with more schools. Within each commune, three public primary schools were randomly selected. By ranking schools from large to small and ensuring that a school was picked out of each tercile, a representative sample of school sizes was chosen.

In the First Round, 252 schools were visited. Six percent of the visited schools were closed at the time of the survey and researchers ended up with reliable data on 238 schools.

In the province of Antananarivo 54 schools were visited, 63 schools were visited in Fianarantsoa, 36 - in Toamasina, 45 - in Mahajanga, 36 - in Toliara and 18 - in Antsiranana.

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

Face-to-face [f2f]

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

The following survey instruments are available:

- Enquête Au Niveau Des Établissements Scolaires, Enquete Cisco;
- Enquête Au Niveau Des Établissements Scolaires, Enquete Directeur Ecole, Visite 1er Jour;
- Enquête Au Niveau Des Établissements Scolaires, Enquete Directeur Ecole, Visite 2ème Jour;
- Enquête Au Niveau Des Établissements Scolaires, Enquete Enseignant.

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 can also be found in external resources.

摘要
---------------------------
马达加斯加的学龄儿童入学率较低：仅有60%的城市儿童和12%的农村儿童完成了小学教育（世界银行，2002年）。为了提高入学率和完成率，以及教育质量，马达加斯加政府大幅增加了对教育领域的投资。政府致力于实现全民教育（EFA）倡议，并开始通过所谓的“学校基金”全额补贴学费，并为公立小学的所有学生提供学习用品。政府还提高了各区的学校物资预算，并开始向学校免费分发教科书。

本研究调查了马达加斯加公立小学教育领域的不同资源配置方式。

调查分为两轮进行。第一轮于2006年10月至11月进行，第二轮于2007年4月至5月进行。研究采用分层随机抽样方法实施。分析了28个区超过200所学校的数据。

在实施PETS教育调查的同时，对马达加斯加医疗设施和工作人员进行的公共支出跟踪调查。

地理覆盖范围
---------------------------
省份：塔那那利佛、菲亚纳兰楚阿、塔马塔夫、马哈贾加、托利阿拉和安齐兰纳。

分析单位
---------------------------
- Cisco（学区/教育设施区）
- 学校
- 校长
- 教师。

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

抽样程序
---------------------------
本研究采用分层随机抽样方法。

分层样本的设置旨在在全国范围内具有代表性。马达加斯加有22个地区和111个区，每个地区至少访问一个区。在六个最大的地区中，各选择两个区。因此，总共访问了28个区。选定的区通过随机选择获得，对区内公立小学数量较多（较少）的区给予更大的（较小的）权重。在每个区，随机选择三个公社，对学校数量较多的公社给予更大的权重。在每个公社内，随机选择三所公立小学。通过将学校按规模从大到小排序，并确保从每个三分之一中选出一所学校，选择了一个具有代表性的学校规模样本。

在第一轮中，访问了252所学校。调查时，6%的访问学校已关闭，研究人员最终获得了关于238所学校的可靠数据。

在塔那那利佛省，访问了54所学校，在菲亚纳兰楚阿访问了63所学校，在塔马塔夫访问了36所学校，在马哈贾加访问了45所学校，在托利阿拉访问了36所学校，在安齐兰纳访问了18所学校。

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

研究工具
---------------------------
以下调查工具可用：
- 学校水平调查，Cisco调查
- 学校水平调查，学校校长调查，第一天访问
- 学校水平调查，学校校长调查，第二天访问
- 学校水平调查，教师调查。

数据清理操作
---------------------------
有关数据编辑程序详细信息的“PETS/QSDS调查数据清理指南”可在外部资源中找到。
STATA清理do文件和数据质量报告也可在外部资源中找到。