Household Income and Expenditure Survey 2019 - Kiribati

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

The purpose of the Household Income and Expenditure Survey (HIES) survey is to obtain information on the income, consumption pattern, incidence of poverty, and saving propensities for different groups of people in Kiribati. This information will be used to guide policy makers in framing socio-economic developmental policies and in initiating financial measures for improving economic conditions of the people. Some more specific outputs from the survey are listed below:

a) To obtain expenditure weights and other useful data for the revision of the consumer price index;
b) To supplement the data available for use in compiling official estimates of household accounts in the systems of national accounts;
c) To supply basic data needed for policy making in connection with social and economic planning;
d) To provide data for assessing the impact on household living conditions of existing or proposed economic and social measures, particularly changes in the structure of household expenditures and in household consumption;
e) To gather information on poverty lines and incidence of poverty throughout Kiribati.

In addition, newly developed modules were incorporated in the 2019 HIES including:
-Person Details;
-Anaemia & Diabetic Test;
-Food Recall;
-Food Away From Home;
-Partaker;
-Non-Food Recall;
-Household Details;
-Dietary Recall;
-Disability, Healthy Living & Time-Use;
-Deprivation And Financial Inclusion;
-Migrant Worker;
-Geographic Information + Photo;
-Market Survey;
-Village Resource Survey (Vrs).

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

National coverage

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

Households

Universe
---------------------------

The survey covered all persons resident in private households.

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

Sample survey data [ssd]

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

SAMPLE SIZE:

In determining an appropriate sample size for a survey of this nature, numerous factors come into the equation. These include:

a) The degree of accuracy required for key estimates;
b) The population size of the country;
c) The manner in which the sample is selected;
d) Cost or staffing constraints which may exist;
e) Whether or not estimates are required for sub-populations;
f) The level of variability in the data being collected.

Each of these factors have different magnitudes of importance, but the major priority should always be on selecting a sample big enough to produce results of suitable accuracy. Many of these issues are generally known as well - for instance:

· A user group may pre-specify what level of accuracy they may wish to achieve for the survey
· The population of a country can normally be estimated to a reasonable level of accuracy
· The sample selection technique adopted is known
· Cost and staff constraints are generally known, and
· A user group can once again provide information on whether estimates for sub-populations are required.

The Kiribati 2019 Household Income and Expenditure Survey (HIES) aims to release outputs at the island division level and National level. The targeted sample size has been determined around 2,000 households based on the results of the previous 2006 HIES that provided the following Relative Sampling Error (RSE) at the strata level:

-Sth. Tarawa: 230 with an average cluster size of 10,5 and a Relative Sampling Error (RSE) of: 6,3%;
-Northern: 245 with an average cluster size of 11,1 and a Relative Sampling Error (RSE) of: 5,7%;
-Central: 217 with an average cluster size of 12,1 and a Relative Sampling Error (RSE) of: 6,7%;
-Southern: 244 with an average cluster size of 11,1 and a Relative Sampling Error (RSE) of: 19,0%;
-Line Is. & Phoenix: 225 with an average cluster size of 13,2 and a Relative Sampling Error (RSE) of: 14,7%;
-TOTAL: 1,161 with an average cluster size of 11,6 and a Relative Sampling Error (RSE) of: 5,4%.

The 2006 Kiribati HIES was based on stratified cluster sampling strategy. The selection of households was based on a three stages selection: island, EA and households. This is the reason why the RSE are above 5% in all domains. In order to improve the quality of the 2019 HIES results, decisions were made to increase the total sample size (making sur we will not over pass the allocated budget) and to use a stratified cluster sampling strategy based on a 2 stage selection (selection of EA and households). The optimal allocation of 1,800 households was used as a first step to the sample allocation, and through several adjustments, the total sample size is 2,180 households with a cluster size of 12 households.

-Sth. Tarawa: 600 with an average cluster size of 12, a number of EAS of 50 and RSE is: 3,8%;
-Northern: 400 with an average cluster size of 12, a number of EAS of 33 and RSE is: 4,3%;
-Central: 300 with an average cluster size of 12, a number of EAS of 25 and RSE is: 5,2%;
-Southern: 480 with an average cluster size of 12, a number of EAS of 40 and RSE is: 13,1%;
-Line Is. & Phoenix: 400 with an average cluster size of 12, a number of EAS of 33 and RSE is: 9,3%;
-Total: 2,180 with a number of EAS of 182 and RSE is: 2,9%.

SAMPLE SELECTION:
The random selection of PSU (EAs) was based on a probability proportional to size selection within each domain.
Within each selected EA, a total of 18 households are selected in order to have a replacement list of 6 households (list B) and 12 to contact in priority (list A).

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

Computer Assisted Personal Interview [capi]

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

The questionnaires were published in English. There are 16 sections in the Household Income and Expenditure Survey questionnaire which relate to the following:

-1. Household ID
-2. Household member roster
-3. Person details
-4. Anaemia & diabetic test
-5. Food recall
-6. Food away from home
-7. Partaker
-8. Non-food recall
-9. Household details
-10. Dietary recall
-11. Disability, healthy living & time-use
-12. Deprivation and financial inclusion
-13. Migrant worker
-14. Geographic information + photo
-15. Market survey
-16. Village resource survey (vrs).

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

Data editing was done using the software Stata Version 15. The completed questionnaires were entered into the Survey Solutions CAPI data entry system.

Response rate
---------------------------

Below is a table showing the response rates based on list A (households selected from the sample):
-South Tarawa: 85.8%;
-Northern: 91.2%;
-Central: 80%;
-Is. & Phoenix: 87.5%;
-TOTAL: 81.1%.

Below is the table showing the completion rates based on valid households from lists A and B (households from the sample + replacements):
-South Tarawa: 99.7%;
-Northern: 100%;
-Central: 100%;
-Is. & Phoenix: 100%;
-TOTAL: 100%.

摘要
---------------------------
家庭收入与支出调查（HIES）的目的在于收集基里巴斯不同人群的收入、消费模式、贫困发生率和储蓄倾向等信息。这些信息将被用于指导政策制定者制定社会经济发展战略，并采取金融措施以改善民众的经济状况。以下是调查的一些具体成果：
a) 获取消费权重和其他有用数据，以修订消费者价格指数；
b) 补充可用于编制国家账户体系中家庭账户官方估计的数据；
c) 提供政策制定所需的基本数据，涉及社会和经济规划；
d) 提供数据以评估现有或拟议的经济和社会措施对家庭生活条件的影响，尤其是家庭支出结构和家庭消费的变化；
e) 收集基里巴斯全境贫困线和贫困发生率的信息。
此外，2019年HIES中纳入了新开发的模块，包括：
- 个人详细信息；
- 贫血与糖尿病检测；
- 食物回忆；
- 家庭外食物；
- 参与者；
- 非食物回忆；
- 家庭详细信息；
- 饮食回忆；
- 残疾、健康生活与时间利用；
- 贫困与金融包容；
- 流动工人；
- 地理信息+照片；
- 市场调查；
- 村庄资源调查（Vrs）。
地理覆盖范围
---------------------------
全国覆盖
分析单元
---------------------------
家庭
总体
---------------------------
调查涵盖了所有居住在私人家庭中的个人。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
样本量：
在确定此类调查的适当样本量时，需要考虑众多因素。这些因素包括：
a) 对关键估计所需准确度的程度；
b) 国家的总人口规模；
c) 样本选择的模式；
d) 可能存在的成本或人员限制；
e) 是否需要为子群体提供估计；
f) 收集数据中的变异程度。
这些因素的重要性各不相同，但始终应将选择足够大的样本以产生合适准确度的结果作为主要优先事项。其中许多问题通常是众所周知的，例如：
· 用户组可能预先指定他们希望调查达到的准确度水平
· 一个国家的总人口通常可以估计到合理的准确度
· 采样的选择技术是已知的
· 成本和人员限制通常是已知的
· 用户组可以再次提供有关是否需要为子群体提供估计的信息。
基里巴斯2019年家庭收入与支出调查（HIES）旨在在岛屿分部和全国层面发布成果。根据2006年HIES的结果，目标样本量已确定为约2000户家庭，该调查在分层层面上提供了以下相对抽样误差（RSE）：
- 斯塔·塔拉瓦：230户，平均集群规模为10.5，相对抽样误差（RSE）为：6.3%；
- 北部：245户，平均集群规模为11.1，相对抽样误差（RSE）为：5.7%；
- 中心：217户，平均集群规模为12.1，相对抽样误差（RSE）为：6.7%；
- 南部：244户，平均集群规模为11.1，相对抽样误差（RSE）为：19.0%；
- 线岛与凤凰岛：225户，平均集群规模为13.2，相对抽样误差（RSE）为：14.7%；
- 总计：1,161户，平均集群规模为11.6，相对抽样误差（RSE）为：5.4%。
2006年基里巴斯HIES基于分层集群抽样策略。家庭的选择基于三个阶段的选择：岛屿、EA和家庭。这就是为什么所有领域的RSE都高于5%。为了提高2019年HIES结果的质量，决定增加总样本量（确保不会超过分配的预算），并采用基于两阶段选择的分层集群抽样策略（选择EA和家庭）。第一步将1,800户家庭的最佳分配用于样本分配，经过多次调整，总样本量为2,180户，集群规模为12户。
- 斯塔·塔拉瓦：600户，平均集群规模为12，EA数量为50，相对抽样误差（RSE）为：3.8%；
- 北部：400户，平均集群规模为12，EA数量为33，相对抽样误差（RSE）为：4.3%；
- 中心：300户，平均集群规模为12，EA数量为25，相对抽样误差（RSE）为：5.2%；
- 南部：480户，平均集群规模为12，EA数量为40，相对抽样误差（RSE）为：13.1%；
- 线岛与凤凰岛：400户，平均集群规模为12，EA数量为33，相对抽样误差（RSE）为：9.3%；
- 总计：2,180户，EA数量为182，相对抽样误差（RSE）为：2.9%。
样本选择：
PSU（EA）的随机选择基于每个领域内按规模成比例的概率选择。
在每个选定的EA中，总共选择18户家庭，以便拥有6户家庭的替换名单（名单B）和12户家庭的优先联系名单（名单A）。
数据收集方式
---------------------------
计算机辅助个人访谈 [capi]
研究工具
---------------------------
问卷以英文发布。家庭收入与支出调查问卷包含16个部分，涉及以下内容：
-1. 家庭ID
-2. 家庭成员名单
-3. 个人详细信息
-4. 贫血与糖尿病检测
-5. 食物回忆
-6. 家庭外食物
-7. 参与者
-8. 非食物回忆
-9. 家庭详细信息
-10. 饮食回忆
-11. 残疾、健康生活与时间利用
-12. 贫困与金融包容
-13. 流动工人
-14. 地理信息+照片
-15. 市场调查
-16. 村庄资源调查（vrs）。
数据清理操作
---------------------------
使用Stata Version 15软件进行数据编辑。完成的问卷被输入到Survey Solutions CAPI数据录入系统中。
响应率
---------------------------
以下表格显示了基于名单A（从样本中选出的家庭）的响应率：
- 南塔拉瓦：85.8%；
- 北部：91.2%；
- 中心：80%；
- 线岛与凤凰岛：87.5%；
- 总计：81.1%。
以下表格显示了基于名单A和B（样本中的家庭+替换家庭）的有效家庭的完成率：
- 南塔拉瓦：99.7%；
- 北部：100%；
- 中心：100%；
- 线岛与凤凰岛：100%；
- 总计：100%。