Population and Housing Census 2000 - Ghana

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

Population censuses have been conducted in Ghana at approximately ten-year intervals since 1891 except in 1941, when the series was interrupted as a result of World War II but was resumed in 1948. The first post-independence census was conducted in 1960 and the next in 1970, with the expectation that a decennial census programme would be maintained. Due to circumstances beyond the control of the statistical organization, however, the third post-independence census could not be conducted until 1984. Similarly, the next census which was expected to have been conducted in 1994 was delayed. Only in 1995 was it possible to have the needed commitment to ensure the conduct of the fourth post-independence census which was scheduled for the year 2000.

The 2000 Population and Housing Census was undertaken to update current information on the size, sex, age, composition and other characteristics of Ghana's population and to ascertain the specific changes in these characteristics which had taken place since the last census was conducted in 1984. The Census was expected to ensure the continuation of a time series of demographic and socio-economic benchmark data at the national and sub-national levels and enhance the capability-building programme of the Statistical Service.

The main objective of the 2000 Population and Housing Census was to update the statistical information on the characteristics of the population of Ghana. The 2000 Population and Housing Census was the first time a full-scale housing census was conducted with a population census in one single operation.

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

National

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

- Households
- Individuals
- Dwellings

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

Sample survey data [ssd]

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

Face-to-face [f2f]

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

Consultation with Users
Work on the census questionnaire started in 1998 bearing in mind the data needs of the country. A simple questionnaire was sent to the ministries, relevant government departments, research institutions, relevant departments in the universities, private business associations and other users seeking information on the following:
· whether the organization had used any previous census data
· the specific census data used
· what use the census data were put
· any data that were needed but had not been provided in previous censuses
· general comments on population censuses.
Response to the questionnaire was encouraging; some respondents sent in the completed forms while others came over to discuss their data needs.

Selection of Topics
Selecting topics for inclusion in the questionnaire involved the review and consideration of the following:
· topics covered in the 1984 population census,
· recommended topics from the United Nations Principles and Recommendations for the 2000 round of Population and Housing Censuses,
· data requests and suggestions from users based on the answers to the questionnaire sent to them,
· list of users' requests compiled by the Statistical Service over a period of time.

A number of meetings were held at both the Census Secretariat and the Technical Advisory Committee levels to discuss the topics and requests. Decisions on topics for inclusion were based on the relevance of topics and the data needs of the country as well as practical considerations of application of concepts.

The final questionnaire consisted of 15 questions on housing characteristics and 20 questions on population covering the following areas:
· household characteristics
· geographical location and internal migration
· demographic and social characteristics
· economic characteristics
· literacy and education
· fertility and mortality.

All the population topics investigated in 1970 and 1984 censuses were maintained, because they were considered as still relevant to the country's data needs, especially in terms of maintaining a time series of socio-economic data.

The questionaires were published in English.

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

The Census data editing was implemented at three levels:

1. Field editing by interviewers and supervisors
2. Office editing and coding
3. Data cleaning and imputation

Data editing was partly manual and partly automatic.

Editing of the census data involved correcting errors from the field and those introduced during the capturing process. Both Structural Edits and Within Record Edits were used to clean the census data.

a) Structural Edits

- Structure edits check coverage and relationships between different units: persons, households, housing units, enumeration areas, etc. Specifically, they checked that:
· all households and collective quarters records within an enumeration area were present and were in the proper order;
· all occupied housing units have person records, but vacant units have no person records;
· households have neither duplicate person records, nor missing person records;
· enumeration areas have neither duplicate nor missing housing records.

- Each EA have the right geographic codes (region, district, locality, EA number, etc.)
- Every housing unit in an EA is entered and every record has a valid EA code

The Structural edit looked at the following situations:

· Geography edits
· Hierarchy of records
· Correspondence between housing and population records
· Editing relationships in a household
· Family nuclei

b) Within Record Edits: This consisted of validity checks and consistency edits.

· Validity checks: were performed to see if the values of individual variables are plausible or lie with a reasonable range.

· Consistency edits were performed to ensure that there is coherence between two or more variables.

The Top-down editing approach, which starts by editing top priority variables, (such as age, sex, etc.) and moves sequentially through all variables in decreasing priority was used to edit the census data.

The Hot Deck or Dynamic Imputation was also used for both missing data and inconsistent/invalid items.

The Census Secretariat carefully developed Editing and Imputation rules with written sets of consistency rules and corrections. These rules were translated into three CONCOR editing applications (Pop-Edit.exe, Hse-Edit.exe and Fertility.exe), which were used to 'clean' the data. This was done at the Regional level.

Data appraisal
---------------------------

A post Enumeration Survey (PES) was conducted to assess the extent of coverage and content error.

摘要
---------------------------
自1891年以来，加纳每十年进行一次人口普查，除1941年因第二次世界大战中断外，普查系列在1948年得以恢复。独立后的第一次普查于1960年进行，紧接着的是1970年的普查，预计将维持每十年一次的普查计划。然而，由于统计机构无法控制的特殊情况，第三次独立后的普查直至1984年才得以实施。同样，原计划于1994年进行的下一次普查也被推迟。直到1995年，才有可能获得必要的承诺，以确保2000年进行的第四次独立后普查按计划进行。
2000年人口与住房普查旨在更新加纳人口规模、性别、年龄、构成及其他特征的最新信息，并确定自上次1984年普查以来这些特征的具体变化。普查预期将确保国家和次国家层面人口与社会经济基准数据的时序连续性，并增强统计服务能力建设计划。
2000年人口与住房普查的主要目标是更新加纳人口特征的统计数据。2000年人口与住房普查是首次在一次单一操作中同时进行人口普查和住房普查。
地理覆盖范围
---------------------------
全国
分析单元
---------------------------
- 家庭
- 个人
- 居住单元
数据类型
---------------------------
样本调查数据 [ssd]
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
与用户协商
普查问卷的工作始于1998年，考虑到国家的数据需求。向各部委、相关政府部门、研究机构、大学的相关部门、私营商业协会和其他用户发送了一份简单的问卷，以了解以下信息：
· 组织是否使用了任何以前的普查数据
· 具体使用了哪些普查数据
· 普查数据被用于什么目的
· 需要但以前普查中未提供的数据
· 对人口普查的一般评论。
问卷的回复令人鼓舞；一些受访者提交了完成的表格，而其他人则过来讨论他们的数据需求。
选择主题
问卷中包含的主题选择涉及以下内容的审查和考虑：
· 1984年人口普查涵盖的主题
· 联合国关于2000年人口和住房普查原则和建议中推荐的主题
· 基于发送给他们的问卷的答案，用户的数据请求和建议
· 统计服务在一定时期内编制的用户请求清单。
在普查秘书处和技术顾问委员会两个层面召开了多次会议，讨论主题和请求。关于包含主题的决定基于主题的相关性和国家的数据需求，以及实际应用概念的可操作性。
最终的问卷包括15个关于住房特征的提问和20个关于人口的提问，涵盖以下领域：
· 家庭特征
· 地理位置和内部迁移
· 人口和社会特征
· 经济特征
· 文盲和教育
· 妇女和儿童。
1970年和1984年普查中调查的所有人口主题均得到保留，因为它们被认为仍然与国家的数据需求相关，特别是在维持社会经济数据时序方面。
问卷以英文出版。
数据清理操作
---------------------------
普查数据的编辑在三个层面上实施：
1. 访谈员和主管的现场编辑
2. 办公室编辑和编码
3. 数据清理和插补
数据编辑部分是手工的，部分是自动的。
普查数据的编辑涉及纠正来自现场以及在捕获过程中引入的错误。使用了结构编辑和记录内编辑来清理普查数据。
a) 结构编辑
- 结构编辑检查不同单元（如人员、家庭、住房单元、普查区等）的覆盖范围和关系：具体而言，它们检查以下内容：
· 普查区内所有家庭和集体宿舍记录都存在，且顺序正确；
· 所有被占用的住房单元都有人员记录，但空置单元没有人员记录；
· 家庭没有重复的人员记录，也没有缺失的人员记录；
· 普查区没有重复或缺失的住房记录。
- 每个普查区都有正确的地理代码（地区、地区、地方、普查区编号等）。
- 每个普查区的每个住房单元都被录入，每条记录都有一个有效的普查区代码。
结构编辑审查以下情况：
· 地理编辑
· 记录层次
· 住房和人口记录之间的对应关系
· 家庭中的编辑关系
· 家庭核心。
b) 记录内编辑：这包括有效性检查和一致性编辑。
· 有效性检查：用于检查个体变量的值是否合理或在一个合理的范围内。
· 一致性编辑：用于确保两个或多个变量之间的一致性。
采用自上而下的编辑方法，首先编辑优先级最高的变量（如年龄、性别等），然后按优先级递减的顺序依次编辑所有变量。
还使用了热盘或动态插补来处理缺失数据和不一致/无效项。
普查秘书处精心制定了编辑和插补规则，形成了成套的一致性规则和纠正措施。这些规则被翻译成三个CONCOR编辑应用程序（Pop-Edit.exe、Hse-Edit.exe和Fertility.exe），用于'清理'数据。这在区域层面完成。
数据评估
---------------------------
进行了一次人口普查后的调查（PES），以评估覆盖范围和内容错误的程度。