Living Standards Survey 2007 - Tajikistan

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

National

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

- Households
- Individuals
- Communites

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

Sample survey data [ssd]

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

A detailed description of the sampling methodology is available in appendix to the document "Basic Information Document".

The TLSS sample was designed to allow reliable estimation of poverty and most variables for a variety of other living standard indicators at the various domains of interest based on a representative probability sample on the level of:
• Tajikistan as a whole
• Total urban and total rural areas
• The five main administrative regions (oblasts) of the country: Dushanbe, Rayons of Republican Subordination (RRS), Sogd, Khatlon, and Gorno-Badakhshan Autonomous Oblast (GBAO)

The last census was conducted in 2000 and covered all five main administrative regions (oblasts) of the country (Dushanbe, RRS, Sogd, Khatlon, and GBAO). Each oblast was further subdivided into smaller areas called census section, instructor's sector and enumeration sector (ES). Each ES is either totally urban or rural. The list of ESs has census information on the population of each ES, and the ES lists were grouped by oblast.

In 2005, UNICEF implemented a Multiple Indicator Cluster Survey (MICS05) in Tajikistan during which an electronic database of the ES information was created. Information in this database included: oblast, rayon, jamoat, settlement type, city/village, ES code, and population. Information from this database was used in the sample design of the TLSS07.

The total number of clusters for the TLSS07 was established as 270 and total number of households per cluster was established as 18, resulting in a sample size of 4,860. The sample size was determined by taking into account:
• The reliability of the survey estimates on both regional and national level
• Quality of the data collected for the survey
• Cost in time for the data collection
• An oversample in 7 rayons in Khatlon

The final cluster allocation is as follows:

Region: Urban / Rural / Total
Dushanbe 50 / 0 / 50
RRP 9 / 45 / 54
Sogd 18 / 38 / 56
Khatlon 12 / 59 / 71
GBAO 6 / 33 / 39
Total 95 / 175 / 270

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

Face-to-face [f2f]

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

Three questionnaires were used to collect information for the TLSS07: a household questionnaire, a female questionnaire for recording information about women of child bearing age, and a community questionnaire. These questionnaires were based on the TLSS questionnaires used in 2003, but had some changes. Questions were added to existing modules and new modules were added to collect information to be used for MICS analyses. These included HIV/AIDS awareness, and Immunizations and Anthropometric Measurements for children 0 to 5 years old. Other new modules on Migration, Financial Services, Subjective Poverty and Food Security, and Subjective Beliefs were also added. The Labor Market Module was changed substantially from 2003 to better look at the informal labor market. The food expenditures module included additional food products. The HIV/AIDS questions were removed from the female questionnaire and were applied to all household members 12 to 49 years old.

The Second Round Household Questionnaire was shorter and was used primarily to collect additional information that was not possible to collect in the First Round. Because the First Round questionnaire was very long, it was decided to collect some information in a second round of visits to the households. The Household Questionnaire was the main instrument used during the Second Round. The female questionnaire was only used if females were added to the household after the First Round and the community questionnaire was not repeated. In the Second Round Household Questionnaire, the time reference period for the Food
Security module was reduced from 4 weeks to 2 weeks. This was done because in the households visited at the beginning of the Second Round, a 4 week period would have included the last portion of the Ramadan period.

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

Data Entry and Cleaning

The data entry program was designed using CSPro, a data entry package developed by the US Census Bureau. This software allows programs to be developed to perform three types of data checks: (a) range checks; (b) intra-record checks to verify inconsistencies pertinent to the particular module of the questionnaire; and (c) inter-record checks to determine inconsistencies between the different modules of the questionnaire.

The data from the First Round were key entered at the Goskomstat headquarters in Dushanbe starting 4 October 2007 through 25 November 2007. The Second Round and Sughd data were key entered from 26 November 2007 through 12 December 2007. All of the data were double entered with both the First Round, Second Round and Sughd re-collection double entry being completed by 22 January 2008.

The data cleaning process began in February 2008 and was completed at the end of May 2008.

地理覆盖范围
---------------------------
国家层面

分析单元
---------------------------
- 家庭
- 个人
- 社区

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

抽样程序
---------------------------
详细抽样方法描述见文档《基本信息文档》的附录。
TLSS样本设计旨在通过在以下层面上的代表性概率样本，实现对贫困以及多种生活标准指标变量的可靠估计：
• 塔吉克斯坦整体
• 总体城市和总体农村地区
• 国家的五个主要行政区域（州）：杜尚别、共和国直属地区（RRS）、索格德、哈特隆和戈尔诺-巴达赫尚自治州（GBAO）。
2000年进行的最后一次人口普查涵盖了国家的五个主要行政区域（州）（杜尚别、RRS、索格德、哈特隆和GBAO）。每个州进一步细分为更小的区域，称为普查区、指导员区及普查小区（ES）。每个ES要么完全为城市，要么完全为农村。ES名单包含了每个ES的人口普查信息，并按州分组。
2005年，联合国儿童基金会（UNICEF）在塔吉克斯坦实施了多项指标集群调查（MICS05），在此过程中创建了ES信息的电子数据库。该数据库中的信息包括：州、地区、社区、定居类型、城市/村庄、ES代码和人口。该数据库中的信息被用于TLSS07的样本设计。
TLSS07的总集群数为270，每个集群的家庭总数为18，样本量达到4,860。样本量是根据以下因素确定的：
• 对区域和国家层面的调查估计的可靠性
• 收集数据的质
• 数据收集的时间成本
• 在哈特隆的7个地区进行超样本。
最终的集群分配如下：
地区：城市 / 农村 / 总计
杜尚别 50 / 0 / 50
RRP 9 / 45 / 54
索格德 18 / 38 / 56
哈特隆 12 / 59 / 71
GBAO 6 / 33 / 39
总计 95 / 175 / 270

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

研究工具
---------------------------
用于收集TLSS07信息的工具包括三个问卷：家庭问卷、记录育龄妇女信息的女性问卷以及社区问卷。这些问卷基于2003年使用的TLSS问卷，但有所变化。在现有模块中添加了问题，并新增了模块以收集用于MICS分析的信息。这些包括对HIV/AIDS的认识、0至5岁儿童的免疫接种和人体测量学以及关于移民、金融服务、主观贫困和食物安全、主观信念等的新模块。劳动市场模块从2003年大幅改变，以更好地观察非正式劳动力市场。食物支出模块包括了额外的食品产品。HIV/AIDS问题从女性问卷中移除，并应用于12至49岁的所有家庭成员。
第二次家庭问卷较短，主要用于收集在第一轮中无法收集的附加信息。由于第一轮问卷非常长，决定在家庭第二次访问时收集一些信息。家庭问卷是第二轮的主要工具。只有当第一轮后家庭中增加了女性时才使用女性问卷，社区问卷不再重复。在第二轮家庭问卷中，食物安全模块的时间参考期从4周缩短到2周。这是因为，在第二轮开始时访问的家庭中，4周的时间将包括斋月最后一部分。

数据清理操作
---------------------------
数据录入与清理
数据录入程序使用CSPro设计，这是美国人口普查局开发的数据录入软件包。该软件允许开发程序执行三种类型的数据检查：（a）范围检查；（b）记录内检查，以验证与问卷特定模块相关的矛盾；（c）记录间检查，以确定问卷不同模块之间的矛盾。
第一轮数据于2007年10月4日开始，至11月25日在杜尚别国家统计局总部录入。第二轮和苏赫德的数据于2007年11月26日开始，至12月12日录入。所有数据都进行了双录入，第一轮、第二轮和苏赫德重收集的双录入于2008年1月22日完成。
数据清理过程于2008年2月开始，并于5月底完成。