National Survey on Household Budget, Consumption and Standard of Living, EBCNV 2010 - Tunisia

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

<p style="border:solid thin black;"> THE CLEANED AND HARMONIZED VERSION OF THE SURVEY DATA PRODUCED AND PUBLISHED BY THE ECONOMIC RESEARCH FORUM REPRESENTS 100% OF THE ORIGINAL SURVEY DATA COLLECTED BY THE NATIONAL INSTITUTE OF STATISTICS - TUNISIA (INS)
</p>

The National Survey on Household Budget, Consumption, and Standard of Living is a quinquennial survey. The 2010 survey is the ninth of its kind that was carried out by the National Institute of Statistics (INS) in Tunisia.
The eight previous surveys were conducted in 1968, 1975, 1980, 1985, 1990, 1995, 2000 and 2005, concurrently with the preparatory work for the Tunisian development plans.

The survey aims at providing detailed information on the procurement of goods and services for consumption. Its data was collected from direct observation of household consumption to allow for having the necessary elements to assess the situation & changes in the living standards & conditions of the households.

The National Survey on Household Budget, Consumption, and Standard of Living consists of three fundamental parts; the budget survey, the nutrition survey and the access to community services survey. Thus, it tackles three areas of study:
1- Households expenses and acquisitions during the survey period.
2 - Food consumption and nutritional status of households.
3 - Household access to health and education community services.

The main objectives of the "budget survey" are:
a- Estimate the levels of expenditure on the household level: The total expenditure of the household is not only an indicator on household income, but it is also a quantitative assessment of the standard of living index.
b- Evaluate the income distribution: Due to the absence of data on income distribution, the mass distribution of expenditure between the different categories of the population constitutes a first sketch for the income distribution in the country.
c- Assess the structure of expenditure: Detailed information collected on expenditures per product are used to establish the structures of the household expenditure, as well as the budget coefficients according to different levels of classifications of goods and services. These coefficients are particularly useful in the revision and development of the Consumer Prices Index (CPI) weights.
d- Predict the demand of households: The household behavior, assessed in terms of product demand, is synthesized by the coefficients of income elasticity, which, according to the model of consumption retained and under the assumptions of the growth of income and population, allows predicting future household demand.
e- Analyze the importance of consumer subsidies: analysis of the consumption of subsidized goods by expenditure deciles allows identifying the impact of direct consumer subsidies. It also allows evaluating the effectiveness of public policies grants.

The main objectives of "the nutrition survey" are:
a- Provide estimates of food consumption by product for different groups of households according to their demographic and socio-economic characteristics.
b- Estimate food consumption of each product by collecting data on the quantities consumed of each product by source, whether purchased or own produced.
c- Identify the nutritional status of the population according to its demographic, geographic and socio-economic level. The comparison between the standards needs of nutrients to those acquired by the household enables assessing of the nutritional status and thus deficits in different nutrients such as calories, protein, vitamins, calcium, ... can also be captured.
d- Estimate the calorie intake and energy needs of the Tunisian population: This estimate is indispensible in the calculation of the food component of the poverty line and, in consequence, the threshold of global poverty.

The main objective of "the access to community services survey" is to provide an overview on the state of morbidity of the Tunisian population, from one hand, and on the households' access to various health and education public services on other hand.

The raw survey data provided by the Statistical Agency were cleaned and harmonized by the Economic Research Forum, in the context of a major project that started in 2009. During which extensive efforts have been exerted to acquire, clean, harmonize,
preserve and disseminate micro data of existing household surveys in several Arab countries.

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

Covering a sample of all urban, small and medium towns and rural areas.

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

1- Household/family.
2- Individual/person.

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

The survey covered a national sample of households and all individuals permanently residing in surveyed households.

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

Sample survey data [ssd]

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

Sampling method
===========
The National Survey on Household Budget, Consumption and Standard of Living, 2010 has focused initially on a sample of 13,392 households drawn using a two stages stratified random sampling in each governorate.
The sampling frame follows that of the General Census of Population and Housing in 2004 which was updated during the implementation of the National Population and Employment Survey in 2009.

Stratification criteria:
The sampling frame is stratified by two geographical criteria: namely the governorate and the living area. The latter is stratified as follows: large cities, medium and small cities, and non-communal areas.

These stratification criteria (governorate, living area and size of city) represent variables that differentiate between surveyed households' lifestyles. Thus, the 3 strata types used are as follows:

Stratum of large cities (stratum 1): This stratum is formed of large urban centers corresponding to municipalities with more than 100.000 inhabitants and neighboring municipalities.

Stratum of medium and small cities (stratum 2): This stratum includes all medium and small sized cities other than those classified in the stratum of large cities.

Stratum of non-communal areas (stratum 3): It includes agglomerations in rural areas that are classified as major agglomerations in the General Census of Population and Housing 2004 and the National Population and Employment Survey in 2009. In addition to other areas that are located outside the territory of main municipalities and cities.
Households in these areas reside in scattered dwellings or are grouped in small agglomerations.

Survey type
========
The sampling frame is divided on the level of each governorate according to strata previously defined. On the stratum level, a two-stage random sampling is planned for the selection of the survey sample of households.
This process allows to breakdown the sample into clusters of 12 households relatively little distant from each other, thereby facilitating the conduct of the survey at the time of the information collection in the field.

In the first stage, a sample of 1,116 primary units is drawn in proportion to the number of households identified in the 2009National Population and Employment Survey. Taking into consideration that the primary units correspond to the districts that have been defined in the General Census of Population and Housing in 2004, which are geographic areas comprising on average 70 households.

In the second stage, from each primary unit (or cluster), twelve households are drawn through a simple random sampling technique.
A substitutive sample of 12 additional households is further drawn from each primary unit. Those additional households constituting a substitutive list are used to cover for unidentified households at the time of the survey, given the mobility of households and the period between the date on which the sample is drawn and the date on which the survey is conducted.

Sample size
========
The size of the sample drawn in the first stage is 1,116 primary sampling units (PSU) corresponding to 13,392 households. The samples in the second stage are 12 households per primary unit.
To optimize the use of logistic and material resources available, a sample of at least 36 PSU was selected from the less populated governorates, 3 PSU per month (the survey is conducted over a 12 months period).
This represents the monthly work of the survey team (3 interviews and 1 supervisor to whom a car is assigned).
Moreover, as the number of households varies from one governorate to another, it was agreed to adopt different rate of sampling from one governorate to another.

The following table shows the regional distribution of the sample and the corresponding sampling rates.

Regional Distribution of the Survey Sample
<table class='xsl-table'>
<tr>
<td>Region </td><td COLSPAN="2">Total</td> <td COLSPAN="2"> Sample size </td> <td> Second stage sampling rate </td></tr>
<tr><td></td> <td> District</td><td> Households </td><td>District</td><td> Households</td> <td>Household sample (%)</td></tr>
<tr><td>Grand Tunis </td> <td> 7863 </td> <td> 268113 </td><td> 240 </td> <td> 2880 </td> <td> 0.45 </td></tr>
<tr><td>North East </td> <td> 4446 </td> <td> 370812 </td><td> 156 </td> <td> 1872 </td> <td> 0,50 </td></tr>
<tr><td>North West </td> <td> 3821 </td> <td> 269466 </td><td> 144 </td> <td> 1728 </td> <td> 0,58 </td></tr>
<tr><td>Centre East </td> <td> 7379 </td> <td> 606287 </td><td> 216 </td> <td> 1728 </td> <td> 0,29 </td></tr>
<tr><td>Centre West </td> <td> 3871 </td> <td> 300223 </td><td> 144 </td> <td> 2592 </td> <td> 0,86 </td></tr>
<tr><td>South East </td> <td> 2711 </td> <td> 213471 </td><td> 108 </td> <td> 1296 </td> <td> 0,61 </td></tr>
<tr><td>South West </td> <td> 1644 </td> <td> 130371 </td><td> 108 </td> <td> 1296 </td> <td> 0,99 </td></tr>
<tr><td>Total </td> <td> 31735 </td> <td> 2553157 </td><td> 1116 </td> <td> 13392 </td> <td> 0,52 </td></tr>
</table>

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

Face-to-face [f2f]

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

The National Survey on Household Budget, Consumption, and Standard of Living questionnaire consists of three parts:
(1) The budget survey.
(2) The nutrition survey.
(3) The access to community services survey.

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

Raw Data
=======
The sampling units were uniformly distributed over the 12 months of the survey year. Thus, a constant volume of work is conducted at each stage of the survey.
The data entry of the survey records collected was conducted in the INS regional offices using the CsPro software.
It should be also noted that the data coding and entry work was done in parallel with the fieldwork.
The data processing, calculations and outputs are performed using STATA.
The analysis of survey data can then allow for further elaboration of final results.

Harmonized Data
============
- The Statistical Package for Social Science (SPSS) is used to clean and harmonize the datasets.
- The harmonization process starts with cleaning all raw data files received from the Statistical Agency.
- Cleaned data files are then all merged to produce one data file on the individual level containing all variables subject to harmonization.
- A country-specific program is generated for each dataset to generate/ compute/ recode/ rename/ format/ label harmonized variables.
- A post-harmonization cleaning process is then conducted on the data.
- Harmonized data is saved on the household as well as the individual level, in SPSS and converted to STATA format.

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

During the National Survey on Household Budget, Consumption and Standard of Living, 2010, 11,281 out of 13,992 households responded to the survey questionnaire. This represents a response rate of 84.2%.

The percentage of non-response is either due to the exceptional circumstances that occured in Tunisia during the revolution period (January 14, 2011), or due to the refusal of some households to be surveyed.
It should be noted that the nutrition survey was only collected from half of the sample only to avoid annoying the surveyed households.

摘要
---------------------------

<p style="border:solid thin black;">由突尼斯国家统计局（INS）收集并发布的关于家庭预算、消费和 生活水平的调查数据经过清洗与协调，构成了100%的原始调查数据。</p>
</p>

国家家庭预算、消费和生活水平调查是一项五年期调查。2010年的调查是突尼斯国家统计局（INS）进行的第九次此类调查。
前八次调查分别在1968年、1975年、1980年、1985年、1990年、1995年、2000年和2005年进行，与突尼斯发展计划的准备工作同时进行。

该调查旨在提供关于消费商品和服务的详细采购信息。其数据通过直接观察家庭消费收集，以便获得评估家庭生活水平状况及变化的必要要素。

国家家庭预算、消费和生活水平调查包括三个基本部分：预算调查、营养调查和社区服务可及性调查。因此，它涵盖了三个研究领域：
1- 调查期间家庭的支出和采购。
2 - 家庭的食品消费和营养状况。
3 - 家庭对健康和教育社区服务的可及性。

“预算调查”的主要目标是：
a- 估算家庭层面的支出水平：家庭的总支出不仅是家庭收入的指标，也是生活水平指数的定量评估。
b- 评估收入分配：由于缺乏收入分配数据，不同人口类别之间的支出大规模分布构成了国家收入分配的第一份草图。
c- 评估支出结构：通过收集按产品分类的详细支出信息，建立家庭支出结构，以及根据不同商品和服务分类的预算系数。这些系数在修订和发展消费者价格指数（CPI）权重方面特别有用。
d- 预测家庭需求：从产品需求的角度评估的家庭行为，通过收入弹性系数综合，根据保留的消费模型和收入及人口增长的假设，允许预测未来的家庭需求。
e- 分析消费者补贴的重要性：通过分析按支出十等分消费的补贴商品，可以确定直接消费者补贴的影响。它还允许评估公共政策的有效性。

“营养调查”的主要目标是：
a- 提供不同家庭群体根据其人口和社会经济特征的食品消费估计。
b- 通过收集每个产品按来源的消耗量数据（无论是购买的还是自产的），估算每个产品的食品消费。
c- 根据人口、地理和社会经济水平识别人口的营养状况。通过与家庭获得的营养标准需求的比较，可以评估营养状况以及不同营养素（如卡路里、蛋白质、维生素、钙等）的不足。
d- 估算突尼斯人口的卡路里摄入量和能量需求：这一估算对于计算贫困线的食品成分以及全球贫困的阈值至关重要。

“社区服务可及性调查”的主要目标是概述突尼斯人口的发病状况，一方面，以及家庭对各种健康和教育公共服务的可及性，另一方面。

国家统计局提供的原始调查数据经过经济研究论坛在2009年开始的一个大型项目的背景下进行清洗和协调。

地理覆盖范围
---------------------------
涵盖所有城市、中小城镇和农村地区的样本。

分析单位
---------------------------
1- 家庭/家庭。
2- 个人/人。

总体
---------------------------
调查覆盖了全国家庭的样本以及永久居住在调查家庭中的所有个人。

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

抽样程序
---------------------------
抽样方法
===========
2010年国家家庭预算、消费和标准水平调查最初关注使用两阶段分层随机抽样在各个省份抽取的13,392个家庭的样本。
抽样框架遵循2004年人口和住房普查的框架，该框架在2009年国家人口和就业调查的实施期间进行了更新。

分层标准：
抽样框架根据两个地理标准进行分层：即省份和居住区。后者按以下方式分层：大城市、中小城市和非社区地区。

这些分层标准（省份、居住区和城市规模）代表区分调查家庭生活方式的变量。因此，使用的三个层类型如下：

大城市层（层1）：该层由大型城市中心组成，对应于拥有超过100,000居民的市镇及其邻近市镇。

中小城市层（层2）：包括除大城市层中分类的城市以外的所有中小型城市。

非社区地区层（层3）：包括2004年人口和住房普查以及2009年国家人口和就业调查中归类为主要聚集区农村地区的聚集。还包括位于主要市镇和城市领土之外的其它地区。这些地区的家庭居住在分散的住宅中或成小聚集。

调查类型
========
抽样框架根据每个省份的层水平进行划分。在层水平上，计划进行两阶段随机抽样，以选择家庭调查样本。
此过程允许将样本分解为彼此相对较近的12户家庭的集群，从而在信息收集阶段便于进行调查。

在第一阶段，根据2009年国家人口和就业调查中识别的家庭数量，抽取了1,116个一级单元的样本。考虑到一级单元对应于2004年人口和住房普查中定义的区，这些区是平均包含70户家庭的地理区域。

在第二阶段，从每个一级单元（或集群）中通过简单随机抽样技术抽取12户家庭。
从每个一级单元进一步抽取12户额外的家庭作为替代样本。这些额外的家庭构成替代名单，用于在调查时覆盖未识别的家庭，考虑到家庭的流动性和抽样日期与调查日期之间的时间间隔。

样本量
========
第一阶段抽取的样本量为1,116个一级抽样单位（PSU），对应于13,392户家庭。第二阶段的样本量是每个一级单元12户家庭。
为了优化使用可用的物流和物质资源，从人口较少的省份中选择了至少36个PSU的样本，每月3个PSU（调查在12个月内进行）。
这代表了调查团队每月的工作量（3次访谈和1名配备车辆的监督员）。

此外，由于各省份的家庭数量不同，同意采用不同省份的抽样率。

以下表格显示了样本的区域分布和相应的抽样率。

调查样本的区域分布
<table class='xsl-table'>
<tr>
<td>地区 </td><td COLSPAN="2">总计</td><td COLSPAN="2">样本量</td><td>第二阶段抽样率</td></tr>
<tr><td></td><td>区</td><td>家庭</td><td>区</td><td>家庭</td><td>家庭样本 (%)</td></tr>
<tr><td>大突尼斯 </td><td> 7863 </td><td> 268113 </td><td> 240 </td><td> 2880 </td><td> 0.45 </td></tr>
<tr><td>东北部 </td><td> 4446 </td><td> 370812 </td><td> 156 </td><td> 1872 </td><td> 0,50 </td></tr>
<tr><td>西北部 </td><td> 3821 </td><td> 269466 </td><td> 144 </td><td> 1728 </td><td> 0,58 </td></tr>
<tr><td>中东部 </td><td> 7379 </td><td> 606287 </td><td> 216 </td><td> 1728 </td><td> 0,29 </td></tr>
<tr><td>中西部 </td><td> 3871 </td><td> 300223 </td><td> 144 </td><td> 2592 </td><td> 0,86 </td></tr>
<tr><td>东南部 </td><td> 2711 </td><td> 213471 </td><td> 108 </td><td> 1296 </td><td> 0,61 </td></tr>
<tr><td>西南部 </td><td> 1644 </td><td> 130371 </td><td> 108 </td><td> 1296 </td><td> 0,99 </td></tr>
<tr><td>总计 </td><td> 31735 </td><td> 2553157 </td><td> 1116 </td><td> 13392 </td><td> 0,52 </td></tr>
</table>

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

研究工具
---------------------------
国家家庭预算、消费和生活水平调查问卷包括三个部分：
(1) 预算调查。
(2) 营养调查。
(3) 社区服务可及性调查。

数据清洗操作
---------------------------
原始数据
=======
调查单元在调查年度的12个月中均匀分布。因此，在每个调查阶段的每个阶段都进行着恒定的工作量。
调查记录的录入在国家统计局的地区办公室使用CsPro软件进行。
还应注意的是，数据编码和录入工作与实地工作并行进行。
数据处理、计算和输出使用STATA进行。
然后，对调查数据进行分析，以便进一步阐述最终结果。

协调数据
============- 使用统计软件包（SPSS）对数据集进行清洗和协调。
- 协调过程从清洗从统计机构收到的所有原始数据文件开始。
- 清洗后的数据文件随后全部合并，生成一个包含所有待协调变量的个体层面数据文件。
- 为每个数据集生成一个特定于国家的程序，以生成/计算/重新编码/重命名/格式化/标记协调变量。
- 然后对数据进行后协调清洗。
- 协调数据以家庭和个体层面保存于SPSS中，并转换为STATA格式。

响应率
---------------------------
在2010年国家家庭预算、消费和生活水平调查中，13,992户家庭中的11,281户响应了调查问卷。这代表着84.2%的响应率。

非响应率的百分比要么是由于突尼斯在革命时期（2011年1月14日）发生的特殊情况，要么是由于一些家庭拒绝接受调查。
值得注意的是，营养调查只从样本的一半收集，以避免打扰被调查的家庭。