Labour Force Survey 2010 - Sri Lanka

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

The employment/unemployment data are required at very short intervals to monitor the program made in the employment generating policies of the government. To satisfy this need, Sri Lanka Labour Force Survey was designed as a quarterly basis survey to measure the levels and trends of employment, unemployment and labour force in Sri Lanka. Thus the survey is repeated four times each year since the first quarter of 1990.

Key objectives of the survey:
- To study the economically active / inactive population;
- To analyze employment by major industry group and employment status;
- To determine unemployment rates by level of education and by age group;
- To study the informal sector employment;
- To determine the underemployment rates by sector and by major industries

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

National coverage. [Excluding the Northern and the Eastern Provinces]

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

Individuals from the population aged 10 years or more

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

This survey covered persons living in housing units only and, it excludes the institutional population. Working age population (10 years and above) living in the non-institutional households in Sri Lanka.

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

Sample survey data [ssd]

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

Sampling plan and the sampling frame
A two stage stratified sampling procedure was adopted to select a sample of 22,500 housing units to be enumerated in the survey. The master sampling frame prepared for 2001 Census of Population and Housing was used as the sampling frame for the selection of sample for the labour force survey 2010.This main frame was updated after the Tsunami incident in 2007 using quick count method.

Sample size
The quarterly survey was conducted since 1990 using a sample of 2,000 housing units per quarter in areas other than North and East, and the sample size was increased to 4,000 housing units per quarter in 1996 and continued thereafter. However, in 1992 and 1997, an annual sample of 20,000 housing units was selected to give reliable estimates of district level. In 2004 again 20,000 housing units were selected for the survey. However, in order to provide district level estimates precisely, it was decided to use 20,000 - 25,000 housing units as the annual sample from 2006. As such, in 2010, the annual sample of 22,500 housing units were selected.

Sample Allocation
In 2010, 2,250 Primary Sampling Units (PSU's) are allocated to each district and each sector by using the Neymann Allocation method which considers the variance of unemployment rate as usually. The allocated sample for each district then equally distributed among the 12 months. Table 2.1, Chapter 2 of the Labour Force Survey Annual Report 2010 gives the sample distributions by district for year 2010.

Selection of Primary Sampling Units (PSU)
Primary sampling units are the census blocks prepared for the Census of Population and Housing - 2001. The sample frame, which was updated after Tsunami, has been used for the selection of primary sampling units.

Selection of Secondary Sampling Units (SSU)
Secondary Sampling Units are the housing units in the selected 2,250 primary sampling units (census blocks). From each primary sampling unit 10 housing units (SSU) are selected for the survey using random sampling method.

Sampling deviation
---------------------------

The field work of the survey for the year 2010 was done throughout the twelve months of the year. However, in 2010, even though it was planned to conduct the labour force survey in Northern Province, it was a failure due to the prevailing conditions. So the national level estimates based on the data collected in year 2010 do not cover the Northern Province.

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

Face-to-face [f2f]

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

Current survey concepts and methods are very similar to those introduced at the beginning. However, some changes have been made over the years to improve the accuracy and usefulness of the data.

In 2006, some significant improvements were made to LFS schedule to fulfill the requirements of data users and also to provide additional information for planning purposes. This was done after carefully reviewing the LFS schedule with technical assistance provided by the Economic and Social Commission for Asia and the Pacific (ESCAP) Adviser. That was revised focusing on literacy, household economic activities, informal sector employment and underemployment etc. This revised version of the questionnaire has been using since first quarter, 2006.

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

Standard errors and coefficients of variation were calculated for selected labour force variables. These variables include general labour force characteristics, employment by major industry groups, and unemployment by level of education.

Details on these sampling error statistics can be found in the Explanatory Notes under the Special Statistical Appendix of the Labour Force Survey Annual Report 2010.

摘要
---------------------------
为监测政府就业创造政策所制定的项目，需要以极短的时间间隔获取就业/失业数据。为此，斯里兰卡劳动力调查被设计为基于季度的调查，以衡量斯里兰卡就业、失业和劳动力的水平和趋势。因此，自1990年第一季度以来，该调查每年重复四次。

调查的主要目标：
- 研究经济活跃/不活跃的人口；
- 分析按主要行业组和就业状态划分的就业情况；
- 确定按教育水平和年龄组划分的失业率；
- 研究非正规部门就业；
- 确定按部门和主要行业划分的过度就业率。

地理覆盖范围
---------------------------
全国覆盖。 [不包括北部和东部省份]

分析单元
---------------------------
10岁及以上的人口个体

总体
---------------------------
本调查仅涵盖居住在住宅单位内的人员，并排除机构人口。居住在斯里兰卡非机构家庭中的10岁及以上的工作年龄人口。

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

抽样程序
---------------------------
抽样计划和抽样框架
采用两阶段分层抽样程序，选择22,500个住宅单位进行调查。用于劳动力调查2010年样本选择的抽样框架是2001年人口和住房普查的主要抽样框架。该主要框架在2007年海啸事件后使用快速计数法进行了更新。

样本量
自1990年以来，季度调查使用2,000个住宅单位的样本进行，范围除北和东地区外，并在1996年将样本量增加到每季度4,000个住宅单位，并继续沿用。然而，在1992年和1997年，选择了20,000个住宅单位的年度样本，以提供可靠的地区级估计。在2004年再次选择了20,000个住宅单位进行调查。然而，为了精确地提供地区级估计，决定从2006年开始使用20,000至25,000个住宅单位的年度样本。因此，在2010年，选择了22,500个住宅单位的年度样本。

样本分配
在2010年，使用Neymann分配方法，根据失业率的方差分配了每个地区和每个部门的2,250个基本抽样单位（PSU）。然后，分配给每个地区的样本在12个月内均匀分配。表2.1，第2章的劳动力调查年度报告2010年提供了2010年按地区分配的样本分布。

基本抽样单位的选定
基本抽样单位是为2001年人口和住房普查准备的普查区。更新后的样本框架已用于选择基本抽样单位。

次级抽样单位的选定
次级抽样单位是在选定的2,250个基本抽样单位（普查区）中的住宅单位。从每个基本抽样单位中，使用随机抽样方法选择10个住宅单位（SSU）进行调查。

抽样偏差
---------------------------
2010年调查的现场工作在全年十二个月进行。然而，尽管计划在北部省份进行劳动力调查，但由于当时的情况，调查未能进行。因此，基于2010年收集的数据的全国级估计不包括北部省份。

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

研究工具
---------------------------
当前调查的概念和方法与最初引入的非常相似。然而，多年来已经进行了一些修改，以提高数据的准确性和实用性。

在2006年，对LFS调查表进行了重大改进，以满足数据用户的需求，并为规划目的提供额外的信息。这是在仔细审查LFS调查表并获得亚太经济社会委员会（ESCAP）顾问的技术援助后进行的。这次修订重点放在识字、家庭经济活动、非正规部门就业和过度就业等方面。自2006年第一季度起，已经开始使用修订后的问卷。

抽样误差估计
---------------------------
计算了选定劳动力变量的标准误差和变异系数。这些变量包括一般劳动力特征、按主要行业组划分的就业以及按教育水平划分的失业。

有关这些抽样误差统计数据的详细信息，可在2010年劳动力调查年度报告的特别统计附录下的解释说明中找到。