Expenditure and Service Delivery Survey in Education 2002 - Zambia

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

This study is a part of a larger project on education in Zambia. The overall project included the schools survey, provincial education offices and district education offices surveys, testing of students in sampled schools, the household survey and the student survey. The project covered households in a catchment area of 36 isolated schools. Researchers also proposed to re-test students who participated in tests under 2001 National Assessment Survey (NAS). As part of the re-testing exercise, 3,200 pupils formed the initial sample for the administration of tests in English, mathematics, and vernacular. In addition, 20 randomly chosen students from Grade V and Grade VI were interviewed for the student survey.

Documented here are datasets covering schools, Provincial Education Offices (PEO) and District Education Offices (DEO).

Education expenditures in Zambia (apart from teacher salaries) are distributed through an administrative hierarchy consisting of PEOs and DEOs. The survey data contain a detailed tracking of resources allocated by the government through this hierarchy to schools. Approaches of public expenditure tracking surveys (PETS) and quantitative service delivery surveys (QSDS) are integrated in this study.

The study was carried out by the Government of Zambia and the World Bank. It covered 184 primary (grades 1-7) and basic (grades 1-9) schools, 33 DEOs and four PEOs in four provinces: Lusaka, Copperbelt, Northern, and Eastern.

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

Lusaka, Copperbelt, Northern, and Eastern provinces.

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

- District Education Offices;
- Provincial Education Offices;
- Schools;
- Teachers;
- Head-Teachers;
- Students.

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

Sample survey data [ssd]

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

A random sample stratified on the basis of urban and rural locations included 184 schools in 33 districts. The schools surveyed were chosen from a frame of primary (grades 1-7) and basic (grades 1-9) schools in four Zambian provinces: Lusaka, Copperbelt, Northern, and Eastern. The choice of these four provinces was dictated primarily by the variation in educational attainments, regional incomes, and administrative structures. Specifically, Lusaka and Copperbelt are the two richest provinces in Zambia, with high enrollment rates, and Northern and Eastern provinces are the poorest, with enrollment rates only marginally better than the worst performing Central province.

Since this study is linked to 2001 National Assessment Survey (NAS), the choice of schools in the sample was restricted by the sampling methodology of the NAS. The NAS sampling was based on a probability-proportional to size methodology. For this study, researchers surveyed about half the schools covered in NAS 2001. For details on sampling methodology, please refer to "Zambia ESDS 2002 Sampling Note " in external resources.

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

Face-to-face [f2f]

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

The following survey instruments are available:

- District Education Office Questionnaire (Part I)
- District Education Office Questionnaire (Part II)
- Provincial Education Office Questionnaire (Part I)
- Provincial Education Office Questionnaire (Part II)
- General School Questionnaire
- Teacher Roster Card
- Head-Teacher Questionnaire
- Teacher Questionnaire, Grades I-VI
- Head-Teacher Matching Roster
- Pupil Matching Roster.

The Teacher Questionnaire was designed to examine two sorts of inputs that may impact on the performance of a teacher: first are the teacher-inputs such as demographic and educational characteristics and second are institutional inputs (primarily teacher salaries). The teacher questionnaire thus focuses on obtaining a basic demographic and educational profile of the teacher, and then moves on to asking about salary and allowances, as well as delays in the receipt of payments.

The Head-Teacher Questionnaire starts with the same sections as the teacher questionnaire. Several additional sections then probe the characteristics of teachers, pupils, parents (through the PTA section) and the administrative structure (through the Relationship with DEO/PEO sections). Finally the head-teacher questionnaire also contains a section on decisions and shortages, where we try to understand the nature of the financial constraints that schools are operating under.

The General School Questionnaire has a three-fold purpose. First, researchers believed that school infrastructure and location themselves may be important for learning achievement; the first few sections of this questionnaire thus systematically ask about the availability and condition of infrastructure in the school. Second, the questionnaire examines the characteristics of the student population in the school such as the overall profile of attendance and grade-repetition. Finally, the questionnaire links to the DEO/PEO questionnaire to complete the Public Expenditure Tracking exercise. To enable researchers to track the flow of resources, the questionnaire then asks about the receipt of resources from other levels of the administration such as the DEO and the PEO's offices.

The District Education Office and Provincial Education Office Questionnaires like the General School and the Head-Teacher Questionnaires, are designed to address two different components of the survey. Part II is concerned with the tracking of public expenditure-how much do the DEO and PEO offices receive? What are the primary expenses in these offices? Part I is similar in form to the head-teacher questionnaire, and asks about the demographic and educational characteristics of the DEO/PEO, before moving on to examine the views of the DEO on their relationship with schools (through visits and inspections) and the overall educational administration.

Head-Teacher Matching Roster and Pupil Matching Roster were designed to match students, who were tested in math and English and other subjects in 2001 and 2002, with teachers, and to carefully identify the changes that could have potentially affected students during the last year. The datasets documented here do not include test scores.

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
---------------------------
本项研究为赞比亚教育领域一项宏大项目的组成部分。该整体项目涵盖学校调查、省级教育办公室及区级教育办公室调查、对抽样学校学生进行的测试、家庭调查以及学生调查。项目涉及36所孤立学校的服务区域内的家庭。研究人员还提议对在2001年国家评估调查（NAS）中参加测试的学生进行再测试。作为再测试活动的一部分，3,200名小学生构成了英语、数学和方言测试的初始样本。此外，还随机选择了五年级和六年级的20名学生进行学生调查。

本处记录的数据集包括学校、省级教育办公室（PEO）和区级教育办公室（DEO）的数据。
赞比亚的教育支出（除教师工资外）通过由PEO和DEO组成的行政层级进行分配。本项研究整合了公共支出追踪调查（PETS）和定量服务交付调查（QSDS）的方法。

该研究由赞比亚政府和世界银行共同实施，覆盖了四个省份（卢萨卡、铜带、北部和东部）的184所小学（1-7年级）和基础教育学校（1-9年级），33个DEO和4个PEO。

地理覆盖范围
---------------------------
卢萨卡、铜带、北部和东部省份。
分析单元
---------------------------
- 区级教育办公室;
- 省级教育办公室;
- 学校;
- 教师;
- 校长;
- 学生。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
基于城市和农村位置的随机分层抽样包括33个区中的184所学校。所调查的学校来自四个赞比亚省份（卢萨卡、铜带、北部和东部）的初级（1-7年级）和基础教育学校（1-9年级）的框架。选择这四个省份主要是由教育成就、区域收入和管理结构的差异所决定的。具体而言，卢萨卡和铜带是赞比亚最富有的两个省份，入学率较高，而北部和东部省份是最贫穷的，入学率仅略高于表现最差的中央省份。

由于本项研究与2001年国家评估调查（NAS）相关联，因此样本中学校的选取受到NAS抽样方法的影响。NAS抽样基于概率比例大小方法。对于本项研究，研究人员调查了NAS 2001所覆盖学校的一半左右。有关抽样方法的详细信息，请参阅外部资源中的“赞比亚ESDS 2002抽样说明”。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
以下调查工具可供使用：
- 区级教育办公室问卷（第一部分）
- 区级教育办公室问卷（第二部分）
- 省级教育办公室问卷（第一部分）
- 省级教育办公室问卷（第二部分）
- 学校通用问卷
- 教师名册卡
- 校长问卷
- 教师问卷，一至六年级
- 校长匹配名册
- 学生匹配名册。
教师问卷旨在考察可能影响教师表现的两种输入：首先是教师输入，如人口统计和教育特征，其次是制度性输入（主要是教师工资）。因此，教师问卷侧重于获取教师的基本人口统计和教育背景，然后转向询问工资和津贴，以及支付延迟。
校长问卷以与教师问卷相同的部分开始。随后，几个附加部分深入探究教师的特征、学生、家长（通过PTA部分）和管理结构（通过与DEO/PEO的关系部分）。最后，校长问卷还包含一个关于决策和短缺的部分，旨在了解学校在运营中所面临的财务约束的性质。
学校通用问卷具有三项目的。首先，研究人员认为学校基础设施和位置本身可能对学习成绩很重要；因此，问卷的前几部分系统地询问学校基础设施的可用性和状况。其次，问卷检查学校学生群体的特征，如出勤和留级的整体概况。最后，问卷与DEO/PEO问卷相连接，以完成公共支出追踪活动。为了使研究人员能够追踪资源的流动，问卷还询问从其他管理层级（如DEO和PEO办公室）接收资源的情况。
与学校通用问卷和校长问卷类似，区级教育办公室和省级教育办公室问卷旨在解决调查的两个不同组成部分。第二部分涉及公共支出的追踪——DEO和PEO办公室收到的资金有多少？这些办公室的主要支出是什么？第一部分的形式与校长问卷相似，询问DEO/PEO的人口统计和教育特征，然后转向检查DEO对学校关系的看法（通过访问和检查）以及整体教育管理。
校长匹配名册和学生匹配名册旨在将2001年和2002年在数学、英语和其他科目中进行测试的学生与教师进行匹配，并仔细识别可能影响学生在过去一年中发生的变化。此处记录的数据集不包括测试成绩。
数据清理操作
---------------------------
关于数据编辑程序的详细信息可在外部资源中的“PETS/QSDS调查数据清理指南”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。