Labour Force Survey 2009-2010 - Timor-Leste

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

This report presents the results of the first national Labour Force Survey (LFS) to be conducted in Timor-Leste since the country gained its full independence in 2002.

The survey provides data on a variety of key employment issues:
- the labour force, in terms of age, sex, and education;
- the employed population, in terms of occupation, economic sector and multiple job-holding;
- employment conditions, in terms of job permanency, public/private sector, hours worked, underemployment, and net monthly earnings;- informal sector and informal employment, in terms of contractual conditions, size of establishment, benefits of employment, etc.;
- the unemployed, including duration of unemployment, and methods of seeking work; and
- persons not in the labour force, their reasons for not being available to work, and their previous work experience.

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

National coverage

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

Households
Individuals

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

Household members ages 10 and over.
Excludes institutional population: persons living in military installations, correctional and penal institutions, dormitories of schools and universities, religious institutions, hospitals, and so forth. In the case of the armed forces, this means that they were included if they lived as members of a private household, but they were excluded if they lived in dormitories, barracks or similar accommodation.

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

Sample survey data [ssd]

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

DNE made use of the sample of 300 census enumeration areas (EAs) that had been selected for the Timor-Leste Survey of Living Standards 2007 (SLS 2007). Those EAs had been selected with probability proportional to size (PPS), where the measure of size was the 2004 EA census count. The actual number of EAs covered in SLS 2007 was in fact 269 instead of 300, because some large EAs were selected twice (or occasionally three or more times) and therefore received the corresponding number of workloads.

For SLS 2007, an up-to-date listing of households in the selected EAs was prepared in the field, and 24 households were selected in each EA, using a random start. These 24 households were then divided up into three 'tasks' (A, B and C), one for each of the three interviewers in the team. Task A received the first eight named households, Task B the next eight, and Task C the last eight. Each interviewer was required to interview five households, but they had a reserve list of three households in case they could not contact or interview any of the first five households on their list.

For LFS 2010, attempts were made to cover exactly the same EAs as was done in SLS 2007. Occasionally an EA was missed, and in a few cases the number of workloads covered in an EA was greater than the number covered in SLS 2007. Table 1.1 shows the number of urban and rural EA workloads covered in each district, and the expected number of households. Exactly the same listing sheets were used as had been used in SLS 2007, with the same names of household heads from SLS 2007 shown on the lists.

Based on the information provided on the cover sheet of each questionnaire, the sample of 4665 households contained 12,088 males and 12,000 females, giving an average household size of 5.2 (unweighted). Among these households there were 8,610 males aged 10 and over and 8,538 females aged 10 and over; these were the people to whom most of the survey questions were directed, after the basic household listing information had been collected.

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

Face-to-face [f2f]

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

The questionnaires were carefully designed by the ILO team that visited Dili in March 2009. The questions were worded in the correct fashion, allowing the calculation of many statistical indicators that are fully in line with current international standards in labour statistics.
Two questionnaires were used - a household questionnaire and an individual questionnaire.

DNE had organized the translation of the questionnaires into Tetum and the final questionnaires were in booklet form.

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

It is difficult to measure the true response rate on this survey because of the method used for selecting the households to be interviewed. Three interviewers were working in each EA. Each interviewer had five specific households to visit, and kept three households in reserve. If any of those five households could not be located, or had moved, or was out at the time of the interviewer’s visits (even after repeated visits), or refused, or was otherwise not available, the interviewer was allowed to take the first replacement household. There was officially zero nonresponse, with all quotas successfully filled and all household questionnaires being marked with code 1 (‘completed - fully responding household’).

Sampling error estimates
---------------------------

Because the LFS is a sample survey, all estimates are subject to sampling error. Sampling errors have not been included in this report, but all reported figures have been rounded to the nearest thousand in order to make some allowance for the effects of sampling error.

摘要
---------------------------
本报告展示了自2002年东帝汶国家完全独立以来，首次在该国进行的全国劳动力调查（LFS）的结果。
调查提供了关于各种关键就业问题的数据：
- 劳动力，按年龄、性别和教育程度划分；
- 就业人口，按职业、经济部门和多重就业情况划分；
- 就业条件，包括工作稳定性、公共/私营部门、工作时间、非充分就业和净月收入；
- 非正规部门和就业，包括合同条件、机构规模、就业福利等；
- 失业人员，包括失业持续时间和工作寻找方法；
- 以及不在劳动力范围内的人员，他们无法工作的原因以及他们的先前工作经历。
地理覆盖范围
---------------------------
全国范围
分析单元
---------------------------
家庭
个人
总体
---------------------------
10岁及以上的家庭成员。
不包括机构人口：居住在军事设施、矫正和惩罚机构、学校和大专院校宿舍、宗教机构、医院等人员。在武装部队的情况下，这意味着如果他们作为私人家庭的一员居住，则包括在内，但如果他们居住在宿舍、营房或类似住所，则不包括在内。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
DNE利用了300个普查区（EA）的样本，这些样本已被选用于2007年东帝汶生活水平调查（SLS 2007）。这些EA是按规模比例抽样（PPS）选择的，其中规模的衡量标准是2004年的EA人口普查计数。SLS 2007中实际涵盖的EA数量实际上是269个，而不是300个，因为一些大型EA被选两次（或偶尔三次或更多次），因此接收了相应数量的工作量。
对于SLS 2007，现场准备了选定EA中家庭的最新清单，并在每个EA中随机选择24户家庭。这些24户家庭然后被分为三个‘任务’（A、B和C），每个团队中的三个访问者各一个。任务A接收前八个命名的家庭，任务B接收接下来的八个，任务C接收最后的八个。每个访问者都需要访问五个家庭，但如果他们无法联系或访问名单上的前五个家庭，则有一个备用名单上的三个家庭。
对于LFS 2010，试图涵盖与SLS 2007中完全相同的EA。偶尔会错过一个EA，在少数情况下，一个EA中的工作量覆盖数量大于SLS 2007中的覆盖数量。表1.1显示了每个区涵盖的城市和农村EA工作量以及预期的家庭数量。与SLS 2007中使用的完全相同的清单表被使用，SLS 2007中清单表上的户主姓名也显示在清单上。
基于每个问卷封面上的信息，包含4665户家庭的样本中，有12088名男性和12000名女性，平均家庭规模为5.2（未加权）。在这些家庭中，有8610名10岁及以上的男性和8538名10岁及以上的女性；在收集了基本家庭清单信息之后，大多数调查问题都是针对这些人提出的。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
问卷由访问2009年3月蒂利的ILO团队精心设计。问题措辞得当，允许计算许多与国际劳工统计标准完全一致的统计指标。
使用了两种问卷——家庭问卷和个人问卷。
DNE组织了问卷翻译成达语，最终问卷以小册子形式出现。
应答率
---------------------------
由于选择要访谈的家庭所使用的方法，很难测量此次调查的真正应答率。三名访谈者在每个EA中工作。每个访谈者有五个特定的家庭要访问，并保留三个家庭作为备用。如果这五个家庭中的任何一个无法找到，或已搬迁，或访谈者访问时不在家（即使经过重复访问），或拒绝，或以其他方式不可用，访谈者被允许选择第一个替代家庭。官方应答率为零，所有配额均成功填补，所有家庭问卷均标记为代码1（‘完成 - 完全应答的家庭’）。
抽样误差估计
---------------------------
由于LFS是样本调查，所有估计值都受抽样误差的影响。抽样误差未包含在本报告中，但所有报告的数字均已四舍五入到最接近的千位，以对抽样误差的影响做出一定的补偿。