Time Use Survey 2007 - Pakistan

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

A primary objective of the national Time Use Survey in Pakistan is to account for the 24 hours time in term of the full spectrum of activities carried out during the duration. The objectives of the survey are specified as under:-
- To profile the quantum and distribution of paid/unpaid work as a means to infer policy/programme implications from the perspective of gender equity.
- To collect and analyze the time use pattern of the individuals in order to help draw inferences for employment and welfare programmes.
- To collect and analyze the comprehensive information about the time spent by people on marketed and non-marketed economic activities covered under the 1993-SNA, non-marketed non-SNA activities within the General Production Boundary and personal care and related activities that cannot be delegated to others.
- To use the data in generating more reliable estimates on work force.

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

The survey covers all urban and rural areas of the four provinces of Pakistan defined as such by 1998 Population Census excluding Federally Administered Tribal Areas (FATA) and certain administrative areas of NWFP. The population of geographic areas excluded from the survey constitutes about 2 percent of the total population as enumerated in 1998 Population Census. The population excluded is located in difficult terrain and its enumeration through personal interview is not possible within the given constraints of time, access and cost.

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

Households
Individuals

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

The universe consists of all urban and rural areas of the four provinces of Pakistan, defined as such by Population Census 1998, excluding FATA & Military Restricted Areas. The population of excluded area constitutes about 3% of the total population and is located in different terrain.

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

Sampling Frame
Federal Bureau of Statistics has developed its own sampling frame for all urban areas of the country. Each city/town has been divided into a number of enumeration blocks. Each enumeration block consists of 200-250 households on the average with well-defined boundaries and maps. The sampling frame i.e. lists of enumeration blocks as up-dated through Economic Census 2003-04 and the lists of villages/mouzas/dehs published by Population Census Organization as a result of 1998 Population Census have been taken as sampling frame. Enumeration blocks and villages are considered as primary sampling unites (PSUs) for urban and rural domain respectively.

Stratification
a) Urban Domain
i) Large Sized Cities
Karachi, Lahore, Gujranwala, Faisalabad, Rawalpindi, Multan, Sialkot, Sargodha, Bahawapur, Hyderabad, Sukkur, Peshawar, Quetta and Islamabad are considered as large sized cities. Each of these cities constitutes a separate stratum which is further sub-stratified according to low, middle, high income groups based on the information collected in respect of each enumeration block at the time of demarcation/up-dating of urban area sampling frame.
ii) Remaining urban areas
After excluding the population of large sized cities from the population of respective administrative division, the remaining urban population of administrative division of four provinces is grouped together to form a stratum called other urban. Thus ex-division in remaining urban areas in the four provinces constitutes a stratum.
b) Rural Domain
In rural domain, each administrative district in the Punjab, Sindh and NWF Provinces is considered as independent and explicit stratum whereas, in Balochistan, each administrative division constitutes a stratum.

Sample size and its Allocation
Keeping in view the resources available, a sample size of 19600 sample households has been considered appropriate to provide estimates of key characteristics at the desired level. The entire sample of households (SSUs) has been drawn from 1388 Primary Sampling Units (PSUs) out of which 652 are urban and 736 are rural. In order to control seasonal variation etc. sample has been distributed evenly over four quarters. This has facilitated to capture the variation due to any seasonal activity as urban population is more heterogeneous therefore, a higher proportion of sample size has been allocated to urban domain. Similarly NWFP and Balochistan being the smaller province, have been assigned higher proportion of sample in order to get reliable estimates. After fixing the sample size at provincial level, further distribution of sample PSUs to different strata in rural and urban domains in each province has been made proportionately.

Sample Design
A three-stage stratified sample design has been adopted for the survey.
Sample Selection Procedure
a) Selection of Primary Sampling Unites (PSUs)
Enumeration blocks in urban domain and mouzas/dehs/villages in rural domain are taken as primary sampling unites (PSUs). In the urban domain, sample PSUs from each ultimate stratum/sub-stratum is selected with probability proportional to size (PPS) method of sampling scheme. In urban domain, the number of households in enumeration block as up-dated through Economic Census 2003-04 and population of 1998 Census for each village/mouza/deh are considered as measure of size.
b) Section of Secondary Sampling Units (SSUs)
Households within sample PSUs are taken as secondary sampling unites (SSUs). A specified number of households i.e. 12 from each urban sample PSU and 16 from each rural sample PSU are selected with equal probability using systematic sampling technique with a random start. Different households are selected in each quarter.
c) Selection of Third Stage Sampling Units i.e. Individuals/Persons (TSUs)
From the sample households, individuals/persons aged 10+ years within each sample households (SSUs) have been taken as third stage sampling units (TSUs). Two individuals aged 10 years and above among the eligible individuals/persons from each sample household have been interviewed using a selection grid.The grid and selection steps are detailed on p13 of the survey report available under external resources.

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

Face-to-face [f2f]

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

The questionnaire has been framed in the light of contemporary precedents and practices in vogue in the developing countries. The recommendations of Gender Responsive Budgeting Initiatives (GRBI) expert who visited Pakistan in June 2006 have been taken into account. Further, the advice of local experts hailing both from data producing and using agencies has also been considered. Survey Questionnaire and Manual of Instructions, for the Supervisors & Enumerators, was finalized jointly by Federal Bureau of Statistics and GRBI Project staff. The questionnaire was also pre-tested and reviewed accordingly. The questionnaire adopted for the survey is given at Annexure-A.
All the households selected in the sample stand interviewed. Diary part of the questionnaire is filled-in from two respondents selected from each of the enumerated households. The questionnaire consists of the following six parts.
Section-1: Identification of the area, respondents, detail of field visits and staff entrusted with supervision, editing and coding.
Section-2: Detailed information about the socio-economic and demographic particulars of the selected households and individuals. Some of the important household characteristics i.e. ownership status and type of the household, earthquake damage, household items, sources of energy, drinking water, transport, health & education facilities, sources of income, monthly income, age and sex composition of the population.
Section-3: Demographic detail such as age, sex, marital status, educational level, having children, employment status, source of income etc. of the selected respondent of that household
Section-4: Comprised of diary to record the activities performed by the first selected respondent through the 24 hours period between 4.00 a.m. of the day preceding the day of interview and 3.00 a.m. on the day of the interview. Section-5 and 6 pertain to the second selected respondent of the selected household.
The diary which is the core instrument of the time use study is divided into forty eight half-hour slots. An open ended question about the activities performed during the thirty minutes was asked from the respondent. Provision for minimum of recording three activities through half hour slot was made. In case of reporting more than one activity, the respondent was probed whether these activities were carried out simultaneously or one after the other. Similarly, the two locations of performing the activities were also investigated in the diary part of the questionnaire. The activities recorded in the diary are then coded by the field enumerator according to the activity classification given at Annex-B.

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

Soon after data collection, the field supervisors manually clean, edit and check the filled in questionnaire and refer back to field where necessary. This does not take much time since most of the manual editing is done in the field.
Further editing is done by the subject matter section at the Headquarter. Also during data entry, further editing of error identified by applying computer edit checks is done. In edit checks, data ranges in numerical values are used to eliminate erroneous data as a result of mistakes made during coding. Thus, the survey records are edited and corrected through a series of computer processing stages.

摘要
---------------------------
巴基斯坦全国时间利用调查的主要目标是在24小时内对所进行活动的全谱进行分析。调查的目标如下所述：
- 通过描绘有偿和无偿工作的量及其分布，从性别平等的角度推断政策/项目的影响。
- 收集和分析个人时间利用模式，以帮助为就业和福利项目得出结论。
- 收集和分析关于人们在市场上和非市场上的经济活动所花费时间的信息，这些活动包括1993年SNA涵盖的市场和非市场活动，以及位于一般生产边界内的非SNA非市场活动，以及无法委托给他人的个人护理和相关活动。
- 利用数据生成关于劳动力更可靠的估计。
地理覆盖范围
---------------------------
调查涵盖了巴基斯坦四个省份的所有城市和农村地区，这些地区由1998年人口普查定义为如此，不包括联邦管理部落地区（FATA）和北西部的某些行政区域。被调查范围外的人口占1998年人口普查总数的约2%。被排除的人口位于地形险峻的地区，在给定的时间、可访问性和成本限制内无法通过个人访谈进行统计。
分析单位
---------------------------
家庭
个人
总体
---------------------------
总体包括巴基斯坦四个省份的所有城市和农村地区，这些地区由1998年人口普查定义，不包括FATA和军事禁区。被排除地区的人口占总体人口的约3%，位于不同的地形上。
抽样程序
---------------------------
抽样框架
联邦统计局为全国所有城市地区开发了其自己的抽样框架。每个城市/城镇都被划分为若干个普查区。每个普查区平均有200-250户家庭，边界和地图清晰界定。抽样框架，即通过2003-04年经济普查更新的普查区清单和人口普查组织根据1998年人口普查发布的村庄/ mouzas/dehs 列表，已被用作抽样框架。普查区和村庄被视为城市和农村领域的各自初级抽样单位（PSU）。
分层
a) 城市领域
i) 大型城市
卡拉奇、拉合尔、古吉拉纳瓦拉、费萨拉巴德、拉瓦尔品第、木尔坦、西亚尔科特、萨戈达、巴豪帕尔、海得拉巴、苏库尔、佩什瓦、奎达和伊斯兰堡被视为大型城市。这些城市中的每一个都是一个独立的层，并根据每个普查区在划分/更新城市地区抽样框架时收集的信息进一步细分为低、中、高收入群体。
ii) 其他城市地区
在从各自行政区域的居民中排除大型城市居民后，四个省份的剩余城市人口被合并在一起，形成一个称为其他城市的层。因此，四个省份中剩余城市地区的行政划分构成了一个层。
b) 农村领域
在农村领域，旁遮普、信德和西北部省的每个行政区被视为独立的和明确的层，而在俾路支斯坦，每个行政区构成一个层。
样本大小及其分配
考虑到可用的资源，19600个样本家庭被认为是一个合适的样本大小，可以提供所需水平的关键特征估计。整个家庭样本（SSU）是从1388个初级抽样单位（PSU）中抽取的，其中652个是城市，736个是农村。为了控制季节性变化等，样本在四个季度中均匀分配。这有助于捕捉任何季节性活动的变化，因为城市人口更加异质，因此，分配给城市领域的样本大小比例更高。同样，由于西北部和俾路支斯坦是较小的省份，因此分配了更高的样本比例，以获得可靠的估计。在省一级确定样本大小后，将样本PSU按比例分配到每个省份农村和城市领域的不同层中。
样本设计
调查采用了三阶段的分层样本设计。
样本选择程序
a) 初级抽样单位的选取
在城市领域，将普查区作为初级抽样单位（PSU）。在城市领域，使用抽样方案中的大小成比例（PPS）抽样方法从每个最终层/亚层选择样本PSU。在城市领域，将经济普查2003-04年更新的普查区家庭数量以及每个村庄/ mouza/deh 的1998年普查人口视为规模的衡量标准。
b) 二级抽样单位的选取
样本PSU内的家庭被视为二级抽样单位（SSU）。从每个城市样本PSU中选择了指定数量的家庭，即12户，从每个农村样本PSU中选择了16户，使用系统抽样技术以随机开始的方式以等概率选择。在每季度选择不同的家庭。
c) 第三阶段抽样单位即个人/人员的选取
从样本家庭中，选取了每个样本家庭（SSU）中10岁及以上年龄的个人/人员作为第三阶段抽样单位（TSU）。从每个样本家庭中符合条件的个人/人员中选取了两个10岁及以上的个人进行访谈，使用选择网格。网格和选择步骤在调查报告的第13页上有详细说明。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
问卷是根据发展中国家当代先例和实践制定的。已考虑了2006年6月访问巴基斯坦的性别敏感预算倡议（GRBI）专家的建议。此外，还考虑了来自数据产生和使用机构的地方专家的建议。联邦统计局和GRBI项目工作人员共同确定了调查问卷和指示手册。问卷还进行了预测试并根据需要进行审查。用于调查的问卷见附件-A。
样本中选择的所有家庭都进行了访谈。问卷的日记部分由从每个被普查家庭中选出的两名受访者填写。问卷由以下六个部分组成。
部分-1：识别区域、受访者、现场访问的详细情况和负责监督、编辑和编码的员工。
部分-2：关于选定家庭和个人社会经济和人口特征的详细信息。一些重要的家庭特征，即家庭所有权状况和类型、地震破坏、家庭物品、能源来源、饮用水、交通、卫生和教育设施、收入来源、月收入、人口年龄和性别构成。
部分-3：选定家庭的选定受访者的年龄、性别、婚姻状况、教育水平、是否生育、就业状况、收入来源等人口统计详细信息。
部分-4：包括记录在访谈前一天上午4:00至访谈当天上午3:00之间24小时期间第一选定受访者所执行活动的日记。部分-5和6涉及选定家庭的第二选定受访者。
时间利用研究的核心工具——日记，分为四十八个半小时时段。从受访者那里询问了关于30分钟内执行的活动的问题。通过半小时时段记录至少三个活动的规定。在报告多个活动的情况下，受访者被调查这些活动是否同时进行或依次进行。同样，在问卷的日记部分还调查了执行活动的两个地点。然后，现场调查员根据附件-B中给出的活动分类对日记中记录的活动进行编码。
数据清理操作
---------------------------
在数据收集后不久，现场监督员手动清理、编辑和检查填写的问卷，并在必要时回访现场。这不会花费太多时间，因为大部分手动编辑都是在现场完成的。
总部的主管部门还进行了进一步的编辑。此外，在数据输入过程中，通过应用计算机编辑检查识别出的错误进行了进一步的编辑。在编辑检查中，使用数据范围在数值中的数据来消除由于编码过程中出现的错误而导致的错误数据。因此，调查记录通过一系列计算机处理阶段进行编辑和纠正。