National Labor Force Survey 2003 - Indonesia

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

National Labor Force Survey (SAKERNAS) is a survey that is designed to observe the general situation of workforce and also to understand whether there is a change of workforce structure between the enumeration period. Since the survey was initiated in 1976, it has undergone a series of changes affecting its coverage, the frequency of enumeration, the number of households sampled and the type of information collected. It is the largest and most representative source of employment data in Indonesia. For each selected household, the general information about the circumstances of each household member that includes the name, relationship to head of household, sex, and age were collected. Household members aged 10 years and over will be prompted to give the information about their marital status, education and employment.

SAKERNAS is aimed to gather informations that meet three objectives: 1.Employment by education, working hours, industrial classification and employment status, 2.Unemployment and underemployment by different characteristics and efforts on looking for work, 3.Working age population not in the labor force (e.g. attending schools, doing housekeeping and others).

The data for annual SAKERNAS was gathered in August 2003 covered all provinces in Indonesia with 67.072 households, scattered both in rural and urban areas and representative until provincial level. The main household data is taken from core questionnaire of SAK2003-AK.

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

National coverage*, including urban and rural area, representative until provincial level.

*) Although covering all of Indonesia, there are some circumstances when not all provincial were covered. For example, in 2000, the Province of Maluku excluded in SAKERNAS because horizontal conflicts occurred there. Also, the separation of East Timor from Indonesia in 1999 also changed the scope of SAKERNAS for the years to come. After that, due to the expansion of regional autonomy as a consequence, the proportion of samples per Province is also changed, as in 2006 when the number of provinces are already 33. However, the difference is only on the number of influential scope/level but not to the pattern. On the other hand, changes in the methodology (including sample size) over time is likely to affect the outcome, for example in years 2000 and 2001, when sample size is only 32.384 and 34.176 households, the level of data presentation is only representative to island level, (insufficient sample size even to make it representative to provincial level).

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

Individual

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

The survey covered all de jure household members (usual residents), aged 10 years and over that resident in the household. However, Diplomatic Corps households, households that are in the specific enumeration area and specific households in the regular enumeration area are not chosen as a sample.

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

Sample survey data

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

Annual SAKERNAS 2003 was implemented in the whole territory of the Republic of Indonesia with a total sample of about 67.072 households, both in rural and urban areas and representative until provincial level. Diplomatic Corps households, households that are in the specific enumeration area and specific households in the regular enumeration area are not chosen as a sample.

The sampling method* for annual SAKERNAS 2003 is two-stages cluster sampling design with census block as the primary sampling unit (PSU) and households as the ultimate sampling unit. PSUs were selected with probability proportional to size. A number of households were taken randomly from selected PSUs. However, there is documentation explained about how the sample size was determined at the domain level, or stratification measures that were implemented and also, the sample size allocation across strata, and also detail information about sample frame**.

The sampling for the urban areas and rural areas is done separately, and by following this procedure:
1. In the first stage, from the sample frame of census block, selected some census block number with probability proportional to size (pps) to the number of household size.
2. At the second stage, from each selected census blocks selected some households in linear systematic household sampling.The first stage sample selection is done by the BPS, while the second level is done by the supervisor/examiner of SAKERNAS.

*) Sampling method used is varied in different years. For example, in SAKERNAS period of 1986-1989 sampling method used is the method of rotation, where most of the households selected at one period was re-elected in the following period. This often happens on quarterly SAKERNAS on that period. At other periods often use multi-stages sampling method (two or three stages depend on whether sub block census included or not), or a combination of multi stages sampling also with rotation method (e.g. SAKERNAS 2006).

**) Commonly, annual SAKERNAS sample frame comes from the last population census result undertaken before SAKERNAS. For example, for annual SAKERNAS 2003 used sample frame derived from "listing process" of household results of Population Census 2000. Also can refer to sampling frame of some periodic household based census like Economic Census, e.g. block census sample frame of SAKERNAS 2007 formed using Economic Census 2006 result. In the other hand sample frame used for quarterly SAKERNAS is from the list of households obtained from National Socio-Economic Survey (SUSENAS) Core activities held before Sakernas. For example, for quarterly SAKERNAS 2002/2003 activities, which used sample frame derived from the household of the selected districts of SUSENAS 2002.

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

Face-to-face

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

In SAKERNAS, the questionnaire has been designed in a simple and concise way. It is expected that respondents will understand the aim of question of survey and avoid the memory lapse and uninterested respondents during data collection. Furthermore, the design of SAKERNAS's questionnaire remains stable in order to maintain data comparison.
A household questionnaire was administered in each selected household, which collected general information of household members that includes name, relationship with head of the household, sex and age. Household members aged 10 years and over were then asked about their marital status, education and occupation.

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

Stages of data processing in Sakernas are through process of:
- Batching
- Editing
- Coding
- Data Entry
- Validation
- Tabulation

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

Sampling error results are presented at the end of the publication of The State of Labor Force in Indonesia and in publication of The State of Workers in Indonesia.

摘要
---------------------------
国家劳动力调查（SAKERNAS）是一项旨在观察劳动力总体状况并了解在统计周期内劳动力结构是否发生变化的调查。自1976年启动以来，该调查经历了多次变化，影响了其覆盖范围、统计频率、样本家庭数量和信息收集类型。它是印度尼西亚最大的、最具代表性的就业数据来源。对于每个选定的家庭，收集了关于每个家庭成员情况的详细信息，包括姓名、与户主的亲属关系、性别和年龄。10岁及以上的家庭成员将被要求提供其婚姻状况、教育和就业信息。
SAKERNAS旨在收集满足以下三个目标的信息：1.按教育、工作时间、工业分类和就业状况划分的就业情况；2.按不同特征和寻找工作的努力程度划分的失业和就业不足情况；3.不在劳动力中的适龄人口（例如，在校学习、家务劳动等）。
2003年8月收集的年度SAKERNAS数据覆盖了印度尼西亚所有省份，共67,072个家庭，分布于农村和城市地区，并在省级层面具有代表性。主要家庭数据来自SAK2003-AK的核心问卷。
地理覆盖范围
---------------------------
全国范围，包括城市和农村地区，并在省级层面具有代表性。

* 虽然覆盖了整个印度尼西亚，但在某些情况下并非所有省份都包括在内。例如，在2000年，由于发生水平冲突，马鲁古省被排除在SAKERNAS之外。此外，1999年东帝汶从印度尼西亚分离也改变了未来SAKERNAS的范围。此后，由于地方自治权的扩大，各省份的样本比例也发生了变化，例如在2006年，当省份数量已达到33个时。然而，这种变化仅在影响力范围/层面上有所不同，而不是在模式上。另一方面，随着时间的推移，方法（包括样本量）的变化可能会影响结果，例如在2000年和2001年，样本量仅为32,384和34,176个家庭时，数据的呈现水平仅代表岛屿级别（样本量不足，甚至无法代表省级水平）。
分析单位
---------------------------
个人

总体
---------------------------
该调查涵盖了所有法定家庭成员（常驻居民），年龄在10岁及以上，居住在家庭中。然而，外交使团家庭、特定统计区域的家庭以及常规统计区域中的特定家庭不被选为样本。
数据类型
---------------------------
样本调查数据

抽样程序
---------------------------
2003年的年度SAKERNAS在整个印度尼西亚领土上实施，总样本量约为67,072个家庭，分布于农村和城市地区，并在省级层面具有代表性。外交使团家庭、特定统计区域的家庭以及常规统计区域中的特定家庭不被选为样本。
2003年年度SAKERNAS的抽样方法为两阶段聚类抽样设计，以人口普查街区为一级抽样单位（PSU），以家庭为最终抽样单位。PSU以规模成比例的概率进行选择。从选定的PSU中随机抽取一定数量的家庭。然而，有文件说明了如何确定领域层面的样本量，或实施的分层措施，以及各层之间的样本量分配，以及样本框架的详细信息。
城市地区和农村地区的抽样是分开进行的，并遵循以下程序：1.在第一阶段，从人口普查街区的样本框架中，以规模成比例的概率（pps）选择一些人口普查街区编号，以家庭规模的数量成比例。2.在第二阶段，从每个选定的街区中，通过线性系统抽样选择一些家庭。第一阶段样本选择由BPS进行，而第二阶段由SAKERNAS的监督员/检查员进行。

* 在不同的年份，使用的抽样方法各不相同。例如，在1986-1989年的SAKERNAS期间，使用的抽样方法是轮换法，其中大部分在一个时期内被选中的家庭在下个时期将被重新选举。这种情况在那一时期的季度SAKERNAS中经常发生。在其他时期通常使用多阶段抽样方法（二或三个阶段取决于是否包括子街区人口普查），或结合多阶段抽样方法以及轮换法（例如，SAKERNAS 2006）。

**) 通常，年度SAKERNAS的样本框架来自SAKERNAS之前的最后一次人口普查结果。例如，对于2003年的年度SAKERNAS，使用的样本框架来自2000年人口普查的家庭“列举过程”。也可以参考一些周期性家庭基于普查的样本框架，例如经济普查，例如SAKERNAS 2007的街区普查样本框架是使用2006年的经济普查结果形成的。另一方面，用于季度SAKERNAS的样本框架来自国家社会经济调查（SUSENAS）核心活动之前获得的家庭清单。例如，对于2002/2003年的季度SAKERNAS活动，使用的样本框架来自SUSENAS 2002所选地区的家庭。
数据收集方式
---------------------------
面对面

研究工具
---------------------------
在SAKERNAS中，问卷被设计得简单而简洁。预计受访者将理解调查问题的目的，避免在数据收集过程中出现记忆缺失和缺乏兴趣的受访者。此外，SAKERNAS问卷的设计保持稳定，以便维持数据比较。
在每个选定的家庭中进行了家庭问卷，收集了家庭成员的一般信息，包括姓名、与户主的亲属关系、性别和年龄。然后询问10岁及以上的家庭成员关于其婚姻状况、教育和职业。
数据清理操作
---------------------------
Sakernas数据处理阶段通过以下过程进行：
- 批处理
- 编辑
- 编码
- 数据录入
- 验证
- 制表
抽样误差估计
---------------------------
抽样误差结果在《印度尼西亚劳动力状况》和《印度尼西亚工人状况》的出版物结束时呈现。